Search
Find anything on CostMon
Every page, guide, glossary term, and FAQ answer on CostMon lives here. Search below, or jump to a kind in the index.
689 entries across 20 kinds.
Pages 44
- CostMon — cloud and AI cost monitoring with spike alertsHome/
- Flat pricing — never a cut of your savingsPricing/pricing
- See it, get alerted, cut itProduct/features
- Three steps to your first alertProduct/how-it-works
- Anatomy of a runaway billCase studies/runaway-bills
- Catch it before the invoiceProduct/cost-alerts
- Should you build your own cloud & AI cost monitoring — or buy it?Build vs. buy/build-vs-buy
- LLM API pricing: Claude vs. GPT vs. GeminiReference/llm-pricing
- AWS Bill Decoder: what every line item actually meansReference/aws-bill-decoder
- The Cloud & AI Cost Optimization PlaybookThe Playbook/playbook
- The Cloud & AI Cost Optimization ChecklistChecklist/checklist
- Cloud & AI Cost GlossaryGlossary/glossary
- Everything CostMon has written about not overpaying for your stackResources/resources
- Every free CostMon tool, in one placeTools/tools
- Every price change that moved a cloud or AI billReference/price-changes
- What CostMon's own price change catalogue actually showsOriginal research/price-change-index
- FOCUS: the open standard for normalizing your billReference/focus-finops-standard
- The spend guardrails you can turn on todayGuardrails/cost-guardrails
- Built to be trusted with your billsSecurity/security
- How CostMon checks accessibility, and where it falls shortAccessibility/accessibility
- What costmon.com does with visitor dataPrivacy/privacy
- Every connector CostMon monitorsSolutions/solutions
- How CostMon comparesComparisons/compare
- See CostMon in actionLive demo/demo
- Find anything on CostMonSearch/search
- Field notes on not overpaying for your stackBlog/blog
- See what your stack really costs, todayGet started/start
- Get in touchContact/contact
- The exact access every connector asks forConnector docs/docs/connect
- Exactly what CostMon asks for to read AWSConnector docs: AWS/docs/connect/aws
- Exactly what CostMon asks for to read GCPConnector docs: GCP/docs/connect/gcp
- Exactly what CostMon asks for to read AzureConnector docs: Azure/docs/connect/azure
- Exactly what CostMon asks for to read AnthropicConnector docs: Anthropic/docs/connect/anthropic
- Exactly what CostMon asks for to read OpenAIConnector docs: OpenAI/docs/connect/openai
- Exactly what CostMon asks for to read Amazon BedrockConnector docs: Amazon Bedrock/docs/connect/bedrock
- Exactly what CostMon asks for to read Vertex AI / GeminiConnector docs: Vertex AI / Gemini/docs/connect/vertex
- Exactly what CostMon asks for to read HeliconeConnector docs: Helicone/docs/connect/helicone
- Exactly what CostMon asks for to read SnowflakeConnector docs: Snowflake/docs/connect/snowflake
- Exactly what CostMon asks for to read DatabricksConnector docs: Databricks/docs/connect/databricks
- Exactly what CostMon asks for to read DatadogConnector docs: Datadog/docs/connect/datadog
- Exactly what CostMon asks for to read GitHubConnector docs: GitHub/docs/connect/github
- Exactly what CostMon asks for to read VercelConnector docs: Vercel/docs/connect/vercel
- Exactly what CostMon asks for to read FOCUS export (CSV)Connector docs: FOCUS export (CSV)/docs/connect/focus
- Push your own data (POST /v1/costs)Connector docs/docs/connect/api
Guides 8
- The AI agent cost guide: why agentic AI blows up your billGuide/ai-agent-costs
- How to choose a cloud & AI cost monitoring toolGuide/how-to-choose-a-cost-monitoring-tool
- How to cut your LLM bill without downgrading your productGuide/reduce-ai-costs
- Context trimmingEngineering time to find and cut redundant context: repeated history, unused tool schemas, dead system-prompt sections. Done right, it has no latency or qualit…/reduce-ai-costs
- Output disciplineA max_tokens ceiling or a stricter output format can truncate a long answer that would otherwise have been correct. Output is priced several times higher than…/reduce-ai-costs
- Prompt cachingZero latency or quality cost on a hit, but a real write premium (1.25x–2x base input) on every miss. Below the break-even hit rate above, this lever is activel…/reduce-ai-costs
- Model routingRouting to a cheaper model only pays off if that model finishes the task. A retry, an escalation to the frontier model, or a wrong answer reaching a customer a…/reduce-ai-costs
- Batch APILittle to no engineering cost beyond an asynchronous code path, but a response can take up to 24 hours to come back. That makes this lever unusable for anythin…/reduce-ai-costs
Calculators 19
- What does your AI usage actually cost?Calculator/llm-cost-calculator
- What is your stack actually wasting?Calculator/savings-calculator
- What does it actually cost to serve one customer?Calculator/cost-per-customer
- Where is your cloud & AI spend actually heading?Calculator/cost-forecasting
- When do your cloud credits run out?Calculator/cloud-credits
- A cost tool that takes a cut of your bill still costs youCalculator/flat-vs-percentage-pricing
- What is your AWS bill actually telling you?Calculator/aws-bill-analyzer
- What is your Claude or OpenAI bill actually telling you?Calculator/ai-spend-analyzer
- What is AWS data transfer actually costing you?Calculator/aws-data-transfer-costs
- What is your AWS storage actually costing you?Calculator/aws-storage-costs
- What is an idle GPU node costing you right now?Calculator/gpu-costs
- What is your observability stack actually costing you?Calculator/observability-costs
- What is your data warehouse actually costing you?Calculator/data-warehouse-costs
- Savings Plans, RIs, or Spot — what should you actually commit to?Calculator/commitment-discounts
- Who is actually paying for your idle cluster?Calculator/kubernetes-costs
- What is your Kubernetes cluster actually wasting?Calculator/kubernetes-waste-analyzer
- Who actually spent this?Calculator/cost-allocation-tagging
- Which of your dollars has no owner?Calculator/tag-coverage-analyzer
- What is your FOCUS export actually telling you?Calculator/focus-export-analyzer
Blog 10
- Catching Runaway Spend Before the InvoiceWhy invoices arrive late, which cost signals move first, what to check every day, and why one common schema makes alerting possible./blog/catching-runaway-spend-before-the-invoice/
- Cut AWS Costs Without Slowing Down EngineeringIdle resources, rightsizing, commitments, storage tiering, and egress: the practical AWS cost levers that keep engineering fast./blog/cut-aws-costs-without-slowing-down/
- FinOps for Teams That Aren't HugeWhat FinOps means, the crawl/walk/run maturity model, why attribution matters, and how a small team gets 80% of the value without a dedicated headcount./blog/finops-for-teams-that-arent-huge/
- One Claude Model, Three BillsOne Claude model, sold by Anthropic, Amazon Bedrock and Google Vertex AI. The token rate is close. The invoice, surcharges and reports differ./blog/one-claude-model-three-bills/
- Read an AWS Bill in 20 MinutesA guided walkthrough of the AWS Bill Analyzer's sample bill, section by section: the total, the spike, and what the numbers mean./blog/read-an-aws-bill-in-twenty-minutes/
- The Monitoring Bill Nobody Budgets ForDatadog and CloudWatch bills grow apart from the infrastructure they watch. Host high-water marks, metric cardinality, and log ingest costs, explained./blog/the-monitoring-bill-nobody-budgets-for/
- The Thirty-Minute Cost ReviewThe agenda, the attendees, the four numbers to open with, and the two decisions that make a monthly cost review worth running./blog/the-thirty-minute-cost-review/
- Thinking Tokens: The Hidden Cost in AI BillsAnthropic, OpenAI and Gemini bill hidden reasoning as output tokens. Learn where that cost hides, which usage field shows it, and how to cap it./blog/thinking-tokens-the-ai-cost-you-cannot-see/
- AI Coding Tools: What One Developer Really CostsA seat price is the small number. See the three ways AI coding tools bill, a worked monthly figure per developer, and where the usage line shows up./blog/what-ai-coding-tools-cost-per-developer/
- What Does Your AI Actually Cost?Token pricing, input vs. output, caching, model choice, and batch processing. See why AI spend hides in most FinOps tools and how to bring it into view./blog/what-does-your-ai-actually-cost/
Solutions 13
- Catch AWS overspend before the invoice doesCostMon connects to AWS Cost Explorer and turns your raw billing data into a daily, normalized view that finance and engineering both trust. A runaway service…/solutions/aws-cost-monitoring/
- See what your AI actually costs before the invoice doesCostMon connects to Anthropic's Admin cost report. It normalizes your Claude API spend into the same daily view as the rest of your stack. A runaway agent loop…/solutions/ai-cost-monitoring/
- Catch GCP overspend across every projectCostMon reads your GCP billing export straight from BigQuery. It turns spend from many projects into one daily view. A runaway BigQuery job or an egress spike…/solutions/gcp-cost-monitoring/
- Catch Azure overspend before it reaches the invoiceCostMon connects to Azure Cost Management's Query API. It turns the cost of each subscription into one daily view. A jump in one service is visible within days…/solutions/azure-cost-monitoring/
- Catch OpenAI overspend before the invoice doesCostMon connects to the OpenAI Admin API's usage and costs reports and merges them per model and per day. A runaway agent loop or an o-series reasoning-token s…/solutions/openai-cost-monitoring/
- See which model is driving your Bedrock billCostMon joins AWS Cost Explorer's Bedrock spend with CloudWatch's per-model token metrics. A model swap that changes your per-token rate shows up in days. So d…/solutions/bedrock-cost-monitoring/
- See what your Gemini usage costs, per modelCostMon reads Vertex AI spend from your GCP billing export. It joins that spend with Cloud Monitoring's per-model token counts. A model that costs more each da…/solutions/vertex-ai-cost-monitoring/
- See which customer, app, or prompt is driving your AI spendCostMon reads cost, tokens and model from Helicone's gateway request log. The gateway knows which user, app or prompt a call belongs to. CostMon adds that deta…/solutions/helicone-cost-monitoring/
- Catch Snowflake credit burn before the invoice doesCostMon reads Snowflake's organization usage views. It turns daily credit use into one view by service type, usage type and account. A warehouse-compute line t…/solutions/snowflake-cost-monitoring/
- Catch a job running on the wrong Databricks clusterCostMon reads Databricks' system.billing.usage table daily. It normalizes DBU use by SKU and workspace. A scheduled job on a pricier all-purpose cluster is vis…/solutions/databricks-cost-monitoring/
- Catch a Datadog cardinality spike before the invoice doesCostMon reads Datadog's Usage Metering API. This is the data behind your Datadog invoice. CostMon splits it by product and by committed or on-demand spend. It…/solutions/datadog-cost-monitoring/
- Catch a macOS-heavy CI matrix before it drains your minutesCostMon reads GitHub's billing usage API each day. It splits usage by product and runner SKU. A CI matrix that burns minutes on pricier macOS and Windows runne…/solutions/github-cost-monitoring/
- Catch a Vercel traffic spike before it hits three meters at onceCostMon reads Vercel's FOCUS-formatted billing charges API each day, one line item at a time. A viral traffic spike can hit Fast Data Transfer, Image Optimizat…/solutions/vercel-cost-monitoring/
Comparisons 7
- CostMon vs. VantageVantage is a broad, developer-friendly multi-cloud cost visibility platform. Below is how it compares to CostMon, and when each one is the better fit./compare/costmon-vs-vantage/
- CostMon vs. CloudZeroCloudZero is built around engineering-led unit economics and cost-per-customer allocation. This page shows how it compares to CostMon, and when each one is the…/compare/costmon-vs-cloudzero/
- CostMon vs. KubecostKubecost is a Kubernetes cost monitoring tool built on the open-source OpenCost project. The comparison below shows how it compares to CostMon, and when each o…/compare/costmon-vs-kubecost/
- CostMon vs. AWS Cost ExplorerAWS Cost Explorer is the native billing console built into every AWS account. Below, we compare it to CostMon and cover when each one is the better fit./compare/costmon-vs-aws-cost-explorer/
- CostMon vs. Datadog Cloud Cost ManagementDatadog Cloud Cost Management is the cost-observability module built into the Datadog platform that many teams already run for monitoring. Here is how it compa…/compare/costmon-vs-datadog-cloud-cost-management/
- CostMon vs. FinoutFinout is a FinOps platform built to unify cloud, Kubernetes, SaaS, and AI spend into one cost model. It is the closest thing to CostMon's own pitch among the…/compare/costmon-vs-finout/
- CostMon vs. InfracostInfracost estimates infrastructure cost before it is ever deployed, straight from your Terraform plan. CostMon tracks what you actually spent after the fact. T…/compare/costmon-vs-infracost/
Glossary 50
- Committed use discountA discount a provider gives you for committing to a minimum spend or usage level for a fixed term./glossary/committed-use-discount
- Reserved instanceA commitment to one instance type in one region or zone, priced below on-demand./glossary/reserved-instance
- Savings planA commitment to a dollar-per-hour spend level. It applies automatically across compute, whatever the instance family./glossary/savings-plan
- On-demand vs. spot instancesOn-demand pricing charges the full published rate with no commitment. Spot pricing sells unused capacity at a steep discount, but the provider can reclaim it o…/glossary/on-demand-vs-spot
- Data egressThe fee a cloud provider charges for data that leaves its network. On most providers, this includes transfer between availability zones./glossary/data-egress
- Amortized, blended, and unblended costThree bases for reporting the same spend. Unblended is the raw rate per resource. Blended averages a rate across accounts. Amortized spreads upfront commitment…/glossary/cost-allocation-methods
- Billed vs. effective vs. list vs. contracted costThe four cost columns FOCUS defines for one charge: cash-basis, amortized, pre-discount, and negotiated-rate. This is why two dashboards can quote two numbers…/glossary/billed-effective-list-contracted-cost
- RightsizingMatching a resource's provisioned size to the usage you observe, not to the peak it was first sized for./glossary/rightsizing
- Idle and orphaned resourcesResources that keep running and billing while they do no useful work. Idle resources are underused. Orphaned resources have lost their owner and purpose./glossary/idle-orphaned-resources
- Storage classA pricing and durability tier for stored data. The same bytes cost a different amount, and take a different time to read back, depending on the class./glossary/storage-class
- Lifecycle policyA rule that moves or deletes stored data on its own after a set condition, most often an age threshold./glossary/lifecycle-policy
- Incremental snapshotA backup that stores only the blocks changed since the previous snapshot, not a full copy. Deleting one frees far less space than its listed size suggests./glossary/incremental-snapshot
- Allocatable capacityWhat is left for pods to request on a Kubernetes node after the kubelet, container runtime, OS, and DaemonSets reserve their share./glossary/allocatable-capacity
- Requests vs. limitsA request is what the scheduler reserves for a container. It decides whether the container fits on a node. A limit is a ceiling that the kubelet enforces at ru…/glossary/requests-vs-limits
- Cluster idleNode capacity that is schedulable but that no pod requested. It is a bin-packing and autoscaling problem, not a request-sizing one./glossary/cluster-idle
- Workload idleCapacity that a pod requested and had reserved but does not use. It is a request-sizing problem, fixed by sizing to observed usage./glossary/workload-idle
- Bin packingHow tightly a scheduler packs pods onto the fewest nodes, given their requests. Loose packing strands capacity as cluster idle across many half-empty nodes./glossary/bin-packing
- Cost allocation tagsKey-value labels on cloud resources. They let you group spend by team, project, environment, or customer, not see one undifferentiated total./glossary/cost-allocation-tags
- Tag coverageThe share of spend, measured in dollars and not in resource count, that carries a required set of tags./glossary/tag-coverage
- Unallocated spendThe share of a bill that you cannot attribute to a real owner. The cause is a missing tag, a wrong tag, or spend that cannot carry a tag at all./glossary/unallocated-spend
- Showback vs. chargebackShowback reports each team's cloud spend for visibility only. Chargeback bills that spend back to the team's budget./glossary/showback-vs-chargeback
- Unit economicsInfrastructure cost expressed per customer, per request, or per transaction, not as one company-wide total./glossary/unit-economics
- RI/SP coverage and utilizationCoverage is the share of eligible usage that a commitment discounts. Utilization is the share of the commitment you bought that you actually use./glossary/ri-sp-coverage-utilization
- Observability taxThe cost of the monitoring stack itself, such as hosts, custom metrics, traces, and logs. It grows on its own, apart from the infrastructure it watches./glossary/observability-tax
- Metric cardinalityThe number of unique combinations of metric name and tag value that a monitoring platform stores and bills for. It is not the number of metric names your code…/glossary/metric-cardinality
- High-water-mark billingA billing method that charges close to a resource's peak usage in a period, not its average. A spike that lasts long enough can set the charge for the whole mo…/glossary/high-water-mark-billing
- Log ingest vs. indexTwo separate charges that most observability platforms bill independently. Ingest is charged when a log event arrives. Index, or retention, is a recurring char…/glossary/log-ingest-vs-index
- Snowflake creditSnowflake's unit of compute billing. A virtual warehouse uses credits every second it runs, at a rate that doubles with each warehouse size./glossary/snowflake-credit
- Auto-suspend / resume minimumSnowflake's default of suspending an idle warehouse to stop billing, then resuming it on the next query. Each resume bills at least 60 seconds, however short t…/glossary/auto-suspend-resume-minimum
- BigQuery slotA unit of query-processing capacity in BigQuery. On-demand queries use a shared pool billed by bytes scanned. An Editions reservation buys dedicated slots bill…/glossary/bigquery-slot
- Bytes scanned & partition pruningOn-demand BigQuery bills by how much data a query reads, not by how long it runs. Partition pruning and column selection are the two levers that shrink that nu…/glossary/bytes-scanned-partition-pruning
- Time Travel & Fail-safeTwo Snowflake features that keep changed or deleted data recoverable, and keep billing storage for it. Time Travel covers a configurable window. Fail-safe cove…/glossary/time-travel-failsafe
- TokenThe basic unit of text that an LLM API meters and bills by. It is roughly three to four characters of English text, not a whole word./glossary/token
- Input vs. output tokensMost LLM APIs price input tokens, the prompt you send, and output tokens, the response generated, separately. Output usually costs several times more./glossary/input-output-tokens
- Context windowThe maximum number of tokens a model can process in one request. It counts the prompt, the conversation history, and the generated output together./glossary/context-window
- Prompt cachingA pricing and speed feature that discounts and speeds up tokens repeated across requests. You do not pay the full input rate each time./glossary/prompt-caching
- Batch APIAn asynchronous request mode. It trades a longer turnaround, from minutes to hours, for a discount against real-time API calls./glossary/batch-api
- Cost per 1M tokensThe standard unit LLM providers use to publish prices: the dollar cost of one million input or output tokens. A single token is too small a unit to price./glossary/cost-per-million-tokens
- Per-request costThe full cost of one API call, shown as one number. It includes input tokens, output tokens, and any tool calls./glossary/per-request-cost
- Fine-tuning vs. inference costFine-tuning cost is a one-time or periodic training expense to specialize a model. Inference cost is the ongoing, usage-based cost of running it./glossary/fine-tuning-vs-inference-cost
- Model tieringRouting each request to the cheapest model that can handle it well, not sending every request to the most capable and most expensive model./glossary/model-tiering
- FinOpsThe practice of managing cloud, and increasingly AI, spend as a shared job between finance and engineering, on an ongoing schedule./glossary/finops
- FOCUSThe FinOps Open Cost and Usage Specification. It is an open, vendor-neutral standard for the shape of a provider's cost and usage export. The FinOps Foundation…/glossary/focus
- Cost anomaly detectionFlagging spend that moves far from a service's or workload's own history. It happens automatically, so you do not wait for a person to notice it on the invoice./glossary/cost-anomaly-detection
- Budget and budget alertA budget is a target spend ceiling for a scope of cost. A budget alert tells an owner when actual or forecast spend nears or crosses it./glossary/budget-alert
- Cost forecastingProjecting future spend from the historical trend and known upcoming changes. You use it to catch a budget overrun before it happens, not after./glossary/cost-forecasting
- Run-rateCurrent spend annualized, or projected to month-end, as if the recent pace continued unchanged. It is a quick but naive look forward./glossary/run-rate
- Total cost of ownershipThe full cost of a system or vendor decision over its useful life. It includes costs beyond the sticker price: migration, integration, ongoing operation, and e…/glossary/total-cost-of-ownership
- Bill shockAn invoice that arrives far higher than you expected, with no earlier warning that the amount was building./glossary/bill-shock
- Percentage-of-savings pricingA cost-tool pricing model that charges a cut of the spend it manages, or of the savings it finds, not a flat fee./glossary/percentage-of-savings-pricing
AWS bill line items 50
- DataTransfer-Out-Bytes — Internet data transfer outA per-gigabyte charge for data that leaves an AWS region for the public internet./aws-bill-decoder/internet-egress
- USE1-USW2-AWS-Out-Bytes — Inter-region data transferA charge for data sent from one AWS region (here, us-east-1) to a different AWS region (here, us-west-2)./aws-bill-decoder/inter-region-transfer
- DataTransfer-Regional-Bytes — Cross-AZ (regional) data transferA charge for data that moves between Availability Zones in the same AWS region. AWS bills both the sending side and the receiving side./aws-bill-decoder/cross-az-transfer
- NatGateway-Bytes — NAT gateway data processingA per-gigabyte charge for every byte that passes through your NAT gateway. It adds to the flat hourly uptime charge. It also adds to any standard data-transfer…/aws-bill-decoder/nat-gateway
- PublicIPv4:InUseAddress — Public IPv4 address chargeAn hourly charge for every public IPv4 address you hold. The charge applies if the address serves traffic. It also applies if the address is allocated but idle./aws-bill-decoder/public-ipv4-address
- VpcEndpoint-Bytes — VPC interface endpoint (PrivateLink)An hourly fee per endpoint per Availability Zone, plus a per-gigabyte fee for data processed through an interface VPC endpoint (AWS PrivateLink)./aws-bill-decoder/vpc-endpoint
- TransitGateway-Bytes — Transit Gateway attachment and data processingAn hourly fee for each attachment (VPC, VPN, Direct Connect, peering) on your Transit Gateway, plus a per-gigabyte fee for data processed through it./aws-bill-decoder/transit-gateway
- LoadBalancerUsage — Load balancer hourly usage and LCU chargeAn hourly rate per load balancer, plus a rate per Load Balancer Capacity Unit (LCU) used. The highest dimension in that hour sets the LCUs. The dimensions are…/aws-bill-decoder/load-balancer-lcu
- US-DataTransfer-Out-Bytes — CloudFront data transfer and requestsA per-gigabyte charge for content that CloudFront serves from its edge locations to end users, plus a per-10,000-requests charge. Rates depend on the edge-loca…/aws-bill-decoder/cloudfront-data-transfer
- BoxUsage:m5.large — On-Demand instance hoursThe hourly running time of an On-Demand EC2 instance of the given type, in the given region. This is the base compute rate before other EC2 charges (storage, d…/aws-bill-decoder/ec2-on-demand-boxusage
- SpotUsage:m5.large — Spot instance hoursThe hourly running time of an EC2 Spot Instance of the given type. AWS bills it at the Spot market price, which changes, instead of the fixed On-Demand rate./aws-bill-decoder/ec2-spot-spotusage
- HeavyUsage:m5.large — Reserved Instance recurring feeThe recurring charge for a Reserved Instance you've already committed to. It's billed whether or not a matching instance is running that hour./aws-bill-decoder/ec2-ri-heavyusage
- UnusedBox:m5.large — Unused Capacity ReservationThe instance-hours of an On-Demand Capacity Reservation that were held open but never had an instance launched into them. These bill at the full On-Demand rate…/aws-bill-decoder/ec2-unused-capacity-reservation
- DedicatedUsage:m5.large — Dedicated Instance hoursThe hourly running cost of an EC2 instance launched with dedicated tenancy. This is single-tenant physical hardware. AWS bills it at a premium over the same in…/aws-bill-decoder/ec2-dedicated-instance-dedicatedusage
- Fargate-vCPU-Hours:perCPU — Fargate vCPU and memory hoursThe vCPU-hours and the companion memory GB-hours requested by your container tasks, billed per second from task start to stop. It's metered by resource request…/aws-bill-decoder/fargate-vcpu-gb-hours
- AmazonEKS-Hours:perCluster — EKS cluster control-plane feeA flat hourly fee charged per EKS cluster for the managed Kubernetes control plane, independent of how many nodes or pods run underneath it./aws-bill-decoder/eks-cluster-hourly-fee
- Lambda-GB-Second — Lambda compute duration (GB-seconds)The memory-allocated × duration charge for Lambda execution. AWS meters it in GB-seconds from invocation (including cold start) to return. The x86 line and the…/aws-bill-decoder/lambda-gb-second-x86-arm
- EBS:VolumeUsage.gp2 — EBS General Purpose SSD storage (gp2/gp3)Billing for the provisioned capacity (GB-month) of General Purpose SSD EBS volumes. The types are gp2 and its newer, usually cheaper sibling, gp3./aws-bill-decoder/ebs-volume-storage-gp2-gp3
- EBS:VolumeP-IOPS.piops — EBS Provisioned IOPS charge (io1/io2)A separate per-IOPS-month charge for the performance you provisioned on io1/io2 volumes, billed apart from the volume's GB-month storage charge./aws-bill-decoder/ebs-provisioned-iops
- EBS:SnapshotUsage — EBS snapshot storageGB-month charge for EBS snapshots in the Standard tier, billed incrementally on the changed blocks each snapshot stores./aws-bill-decoder/ebs-snapshot-storage
- TimedStorage-ByteHrs — S3 Standard storageThe core "you store data, you pay for it" charge. It covers GB-months of object data and metadata held in S3 Standard storage. AWS meters it hourly and average…/aws-bill-decoder/s3-standard-storage
- TimedStorage-SIA-ByteHrs — S3 Standard-Infrequent Access storageGB-month storage charge for objects held in the S3 Standard-IA class. It's cheaper per-GB than Standard, but with per-GB retrieval fees and a 30-day minimum st…/aws-bill-decoder/s3-standard-ia-storage
- TimedStorage-GlacierByteHrs — S3 Glacier Flexible Retrieval storageGB-month storage charge for archived objects held in the S3 Glacier Flexible Retrieval class, AWS's low-cost long-term archive tier./aws-bill-decoder/s3-glacier-storage
- Requests-Tier1 — S3 request charges (PUT/COPY/POST/LIST and GET)Per-request API charges, billed per thousand requests. Tier1 covers PUT, COPY, POST, and LIST calls. A separate Tier2 line covers GET and most other read reque…/aws-bill-decoder/s3-request-charges
- EarlyDelete-ByteHrs — S3 Glacier early-deletion chargeA prorated penalty for an object in S3 Glacier Flexible Retrieval that you delete, overwrite, or transition before its 90-day minimum storage commitment ends.…/aws-bill-decoder/s3-early-delete-charge
- InstanceUsage:db.m5.large — RDS instance hoursThe hourly on-demand compute charge for a single-AZ RDS database instance of a given class, billed from launch until the instance is stopped or deleted./aws-bill-decoder/rds-instance-usage
- Multi-AZUsage:db.m5.large — RDS Multi-AZ instance hoursThe hourly compute charge for a Multi-AZ RDS deployment. It bills the primary instance and its synchronous standby replica in another Availability Zone./aws-bill-decoder/rds-multi-az-usage
- RDS:GP3-Storage — RDS gp3 storageThe per-GB-month charge for General Purpose SSD (gp3) storage provisioned on a single-AZ RDS instance, separate from the instance-hour and IOPS/throughput char…/aws-bill-decoder/rds-gp3-storage
- RDS:ChargedBackupUsage — RDS backup storage overageThe per-GB-month charge for RDS automated backups and manual snapshots. It starts when their combined size exceeds the free allotment (up to 100% of total prov…/aws-bill-decoder/rds-charged-backup-usage
- Aurora:ServerlessV2Usage — Aurora Serverless v2 ACU usageThe per-second (billed hourly) charge for Aurora Capacity Units (ACUs) that an Aurora Serverless v2 instance uses. Each ACU is roughly 2 GiB of memory plus mat…/aws-bill-decoder/aurora-serverless-v2-usage
- Aurora:StorageIOUsage — Aurora I/O requestsThe per-operation charge for read and write I/O requests against Aurora's distributed storage layer, under the default storage configuration./aws-bill-decoder/aurora-storage-io-usage
- WriteCapacityUnit-Hrs — DynamoDB provisioned write capacityThe hourly charge for provisioned write capacity units reserved on a DynamoDB table, billed whether or not the capacity is consumed./aws-bill-decoder/dynamodb-write-capacity-unit-hrs
- NodeUsage:cache.m5.large — ElastiCache node hoursThe hourly on-demand charge per provisioned ElastiCache node of a given instance class. Billing runs from launch until you terminate the node. Partial hours ro…/aws-bill-decoder/elasticache-node-usage
- DataProcessing-Bytes — CloudWatch Logs ingestionCharges for every byte of log data your applications, agents, or functions push into CloudWatch Logs./aws-bill-decoder/cloudwatch-logs-ingestion
- VendedLog-Bytes — CloudWatch vended logsThe per-GB charge for logs that other AWS services generate and deliver into CloudWatch Logs for you. Examples are VPC flow logs, DNS query logs, CDN logs, and…/aws-bill-decoder/cloudwatch-vended-logs
- CW:MetricMonitorUsage — CloudWatch custom metricsThe per-metric-per-month charge for every unique custom metric (and dimension combination) published to CloudWatch. The source can be your own application, emb…/aws-bill-decoder/cloudwatch-custom-metrics
- CW:AlarmMonitorUsage — CloudWatch alarmsThe per-alarm hourly charge (prorated to per-month) for every standard-resolution metric alarm that exists in your account, whether or not it has ever fired./aws-bill-decoder/cloudwatch-alarms
- CW:Requests — CloudWatch API requestsThe per-1,000-request charge for CloudWatch API calls beyond the free tier (most visibly reads of metric data), billed once you exceed the free allotment./aws-bill-decoder/cloudwatch-api-requests
- ConfigurationItemRecorded — AWS Config configuration item recordingThe per-item charge AWS Config bills every time it detects and records a change to a tracked resource's configuration. It's metered by how often your infrastru…/aws-bill-decoder/aws-config-configuration-items
- PaidEventsRecorded — CloudTrail data eventsA per-100,000-events charge for CloudTrail data events. These are fine-grained, resource-level API activity, such as object reads or function invocations. The…/aws-bill-decoder/cloudtrail-data-events
- KMS-Keys — KMS customer-managed key, monthly feeA flat monthly charge (prorated hourly) for every customer-managed key you've created in KMS, whether or not it's ever used to encrypt anything./aws-bill-decoder/kms-keys
- KMS-Requests — KMS API request charge beyond the free tierBilling for KMS API calls (encrypt, decrypt, generate-data-key, and similar) once you exceed the monthly free-tier request allotment./aws-bill-decoder/kms-requests
- AWSSecretsManager-Secrets — Secrets Manager per-secret storage feeA flat monthly charge (prorated hourly) for every secret stored in Secrets Manager, including each cross-region replica, regardless of how often it's used./aws-bill-decoder/secretsmanager-secrets
- AWSSecretsManagerAPIRequest — Secrets Manager API request chargeBilling for API calls to Secrets Manager (retrieving, describing, writing, and listing secrets) at a small per-10,000-requests rate; every operation counts, no…/aws-bill-decoder/secretsmanager-apirequest
- PaidEventsAnalyzed-Bytes — GuardDuty log analysis, per GBGuardDuty's charge for the volume (in GB) of network flow and DNS query log data it analyzes each month. A tiered per-GB schedule applies after any free-trial…/aws-bill-decoder/guardduty-paideventsanalyzed-bytes
- USE1-Claude4.6Sonnet-input-tokens — Bedrock input tokens (per model)Every token sent into a Bedrock model as part of the prompt is metered. This includes system instructions, chat history, retrieved context, and few-shot exampl…/aws-bill-decoder/bedrock-input-tokens
- USE1-Claude4.6Sonnet-output-tokens — Bedrock output tokens (per model)Tokens the model generates in its response are billed as a separate line item from input tokens, usually at a materially higher per-token rate./aws-bill-decoder/bedrock-output-tokens
- USE1-Claude4.6Sonnet-cache-read-input-token-count — Bedrock prompt-cache read tokensTokens that Bedrock serves from its prompt cache appear as their own usage type. Bedrock does not reprocess them as full input. AWS bills them at a steep disco…/aws-bill-decoder/bedrock-cache-read-tokens
- USE1-Host:ml.g5.xlarge — SageMaker real-time endpoint hoursBilled for every hour a SageMaker real-time inference endpoint's instance is running, regardless of whether it's actively serving prediction traffic./aws-bill-decoder/sagemaker-hosting-instance-hours
- USE1-Notebk:ml.g4dn.8xlarge — SageMaker notebook instance hoursBilled for every hour a classic SageMaker notebook instance sits in an active state, independent of whether anyone is actively working in it./aws-bill-decoder/sagemaker-notebook-instance-hours
AWS data transfer paths 12
- Internet data transfer in — $0.00/GBIngress/aws-data-transfer-costs#internet-in
- Global data-transfer-out free tier — First 100 GB/mo freeEgress/aws-data-transfer-costs#global-free-tier
- Internet data transfer out — $0.090/GB ≤10 TB · $0.085/GB next 40 TB · $0.070/GB next 100 TB · $0.050/GB >150 TBEgress/aws-data-transfer-costs#internet-egress
- Cross-AZ (regional) transfer — $0.010/GB, each direction ($0.02/GB total moved)Both directions billed: sender's egress and receiver's ingress/aws-data-transfer-costs#cross-az
- PrivateLink / Transit Gateway / Client VPN cross-AZ transfer — $0.00/GBBoth directions/aws-data-transfer-costs#xaz-waiver
- Inter-region transfer out — $0.010–$0.020/GB depending on destination regionEgress, source-side only (the destination region's inbound leg is $0.00)/aws-data-transfer-costs#inter-region
- NAT gateway — $0.045/hour per gateway + $0.045/GB processedProcessing fee, on top of any standard data-transfer charge the same bytes already incur/aws-data-transfer-costs#nat-gateway
- Public IPv4 address — $0.005/hour per addressFlat holding fee/aws-data-transfer-costs#public-ipv4
- VPC interface endpoint (PrivateLink) — $0.01/hour per AZ + $0.01/GB processedProcessing fee/aws-data-transfer-costs#vpc-interface-endpoint
- Gateway VPC endpoint (S3, DynamoDB) — $0.00No charge/aws-data-transfer-costs#gateway-vpc-endpoint
- Transit Gateway — $0.05/hour per attachment + $0.02/GB processedProcessing fee, independent of the inter-AZ waiver above/aws-data-transfer-costs#transit-gateway
- CloudFront (US price class) — $0.085/GB ≤10 TB, tapering to $0.020/GB at high volume; AWS-origin fetch $0.00/GBEgress, at rates below direct internet egress; origin fetch from an AWS origin is free/aws-data-transfer-costs#cloudfront
AWS storage rates 24
- S3 Standard storage — $0.023/GB-mo ≤50 TB · $0.022/GB-mo next 450 TB · $0.021/GB-mo >500 TBTimedStorage-ByteHrs/aws-storage-costs#s3-standard
- S3 Standard-IA storage — $0.0125/GB-mo30-day minimum storage duration./aws-storage-costs#s3-standard-ia
- S3 Standard-IA retrieval — $0.01/GBRetrieval-SIA/aws-storage-costs#s3-standard-ia-retrieval
- S3 Standard-IA early-delete penalty — $0.0125/GB-mo, prorated for the days short of 30EarlyDelete-SIA/aws-storage-costs#s3-standard-ia-early-delete
- S3 One Zone-IA storage — $0.01/GB-mo30-day minimum storage duration; data survives in a single AZ only./aws-storage-costs#s3-one-zone-ia
- S3 Glacier Instant Retrieval storage — $0.004/GB-mo90-day minimum storage duration./aws-storage-costs#s3-glacier-instant-retrieval
- S3 Glacier Instant Retrieval retrieval fee — $0.03/GBRetrieval-GIR/aws-storage-costs#s3-glacier-instant-retrieval-fee
- S3 Glacier Flexible Retrieval storage — $0.0036/GB-mo90-day minimum storage duration./aws-storage-costs#s3-glacier-flexible
- Intelligent-Tiering: Frequent Access — $0.023/GB-mo ≤50 TB, same ladder as StandardAutomatic. It's the tier every object starts in. Objects under 128 KB stay here permanently; see the traps below./aws-storage-costs#s3-intelligent-frequent
- Intelligent-Tiering: Infrequent Access — $0.0125/GB-moAutomatic after 30 consecutive days with no access./aws-storage-costs#s3-intelligent-infrequent
- Intelligent-Tiering: Archive Instant Access — $0.004/GB-moAutomatic after 90 consecutive days with no access, with no retrieval fee and no loss of millisecond access. Nothing to activate./aws-storage-costs#s3-intelligent-archive-instant
- Intelligent-Tiering: Archive Access — $0.0036/GB-moOpt-in tier, 90+ days with no access. It trades millisecond access for a 3–5 hour restore./aws-storage-costs#s3-intelligent-archive-access
- Intelligent-Tiering: Deep Archive Access — $0.00099/GB-moOpt-in tier, 180+ days with no access. It's the only Deep Archive rate this page can verify first-party; see the module note on plain Glacier Deep Archive stor…/aws-storage-costs#s3-intelligent-deep-archive
- Intelligent-Tiering monitoring fee — $0.0025 per 1,000 objects/moNever charged on objects under 128 KB. They're excluded from monitoring and auto-tiering both./aws-storage-costs#s3-intelligent-monitoring
- Glacier Deep Archive staging storage — $0.021/GB-moA transient staging area, not standalone Deep Archive storage./aws-storage-costs#s3-glacier-deep-archive-staging
- gp3 volume storage — $0.08/GB-moIncludes 3,000 IOPS and 125 MiB/s throughput free./aws-storage-costs#ebs-gp3
- gp2 volume storage — $0.10/GB-moExactly 20% more than gp3 for the same size./aws-storage-costs#ebs-gp2
- io1/io2 volume storage — $0.125/GB-moio1 bills as .piops, not .io1. There is no .io1 usage type. Provisioned IOPS billed separately, see below./aws-storage-costs#ebs-io
- st1 (Throughput Optimized HDD) storage — $0.045/GB-moEBS:VolumeUsage.st1/aws-storage-costs#ebs-st1
- sc1 (Cold HDD) storage — $0.015/GB-moEBS:VolumeUsage.sc1/aws-storage-costs#ebs-sc1
- gp3 provisioned IOPS beyond the free 3,000 — $0.005/IOPS-moEBS:VolumeP-IOPS.gp3/aws-storage-costs#ebs-gp3-extra-iops
- gp3 provisioned throughput beyond the free 125 MiB/s — $0.04/MiBps-mo ($40.96/GiBps-mo)AWS bills this per GiBps-month. Read the unit carefully before comparing it to the free baseline, which is stated in MiB/s./aws-storage-costs#ebs-gp3-extra-throughput
- EBS snapshot storage — $0.05/GB-moThis storage is incremental. It bills on changed blocks, not each snapshot's full listed size./aws-storage-costs#ebs-snapshot-standard
- EBS snapshot archive storage — $0.0125/GB-moSeparate retrieval and early-delete fees apply; a restore takes hours, not seconds./aws-storage-costs#ebs-snapshot-archive
GPU instance rates 9
- g6.xlarge — $0.8048/hour1x NVIDIA L4, 24 GB/gpu-costs#g6-xlarge
- g5.xlarge — $1.006/hour1x NVIDIA A10G, 24 GB/gpu-costs#g5-xlarge
- g6e.xlarge — $1.861/hour1x NVIDIA L40S, 48 GB/gpu-costs#g6e-xlarge
- g5.12xlarge — $5.672/hour4x NVIDIA A10G, 96 GB/gpu-costs#g5-12xlarge
- g6e.12xlarge — $10.49264/hour4x NVIDIA L40S, 192 GB/gpu-costs#g6e-12xlarge
- p4d.24xlarge — $21.957642/hour8x NVIDIA A100, 320 GB HBM2/gpu-costs#p4d-24xlarge
- p5.48xlarge — $55.04/hour8x NVIDIA H100, 640 GB HBM3/gpu-costs#p5-48xlarge
- p5en.48xlarge — $63.296/hour8x NVIDIA H200, 141 GB HBM3e/gpu-costs#p5en-48xlarge
- p6-b200.48xlarge — $113.9328/hour8x NVIDIA B200, 1432 GB HBM3e/gpu-costs#p6-b200-48xlarge
Observability rates 22
- Infrastructure host (Pro plan) — $15/host/mo annual · $18/host/mo on-demandBilled at the high-water mark: the max of the LOWER 99% of the month's hourly host counts. A spike lasting more than roughly 1% of the month's hours (about 7 h…/observability-costs#dd-host
- Custom metrics included per host — 100/host (Pro) · 200/host (Enterprise)Covers both ingested and indexed volume. A custom metric is any unique metric-name + tag-value combination not submitted through one of Datadog's 1,000+ built-…/observability-costs#dd-custom-metrics-included
- Custom metrics ingested beyond allotment — $0.10 per 100 ingested custom metrics/moThe only overage rate Datadog publishes. Indexed-metric overage for accounts NOT using Metrics without Limits is priced per your contract, and Datadog doesn't…/observability-costs#dd-custom-metrics-overage
- Histogram/Distribution metric fan-out — 5x custom metrics per unique series by defaultCovers max, median, avg, p95, and count. Adding percentile aggregations on a Distribution stacks another 5x on top, up to 10x total./observability-costs#dd-histogram-multiplier
- APM host — $31/host/mo annual · $48/host/mo on-demandPer host, per month, bundled with Infrastructure Monitoring/observability-costs#dd-apm-host
- Indexed spans included per APM host — 1M spans/host/moAt the default 15-day retention/observability-costs#dd-apm-spans-included
- Indexed spans beyond allotment, 7-day retention — $1.27/M spans annual · $1.91/M on-demandPer million spans, per month/observability-costs#dd-apm-spans-7d
- Indexed spans beyond allotment, 15-day retention — $1.70/M spans annual · $2.55/M on-demandPer million spans, per month/observability-costs#dd-apm-spans-15d
- Indexed spans beyond allotment, 30-day retention — $2.50/M spans annual · $3.75/M on-demandPer million spans, per month/observability-costs#dd-apm-spans-30d
- Log ingestion — $0.10/GB ingested or scannedFlat: Datadog quotes the same figure whether billed annually or on-demand./observability-costs#dd-log-ingest
- Log indexing, 15-day retention — $1.70/M events annual · $2.55/M on-demandA separate meter from ingestion. Every other retention tier Datadog offers links out to "Contact Us for Pricing" rather than a published rate./observability-costs#dd-log-index-15d
- Flex Storage (rehydratable cold tier) — $0.05/M events annual · $0.075/M on-demand1–15 month retention, searchable on demand rather than always-indexed./observability-costs#dd-log-flex
- Log ingestion, Standard class — $0.50/GBDataProcessing-Bytes/observability-costs#cw-log-ingest-standard
- Log ingestion, Infrequent Access class — $0.25/GBHalf the ingestion cost, but that log group loses metric filters, anomaly detection, and sensitive-data protection. Logs Insights queries still work on it./observability-costs#cw-log-ingest-ia
- Log storage, Standard class — $0.03/GB-moTimedStorage-ByteHrs/observability-costs#cw-log-storage-standard
- Log storage, Infrequent Access class — $0.018/GB-moTimedStorage-IA-ByteHrs/observability-costs#cw-log-storage-ia
- Logs Insights query scanning — $0.005/GB scannedDataScanned-Bytes/observability-costs#cw-logs-insights-scan
- Custom metrics — $0.30/metric-mo ≤10,000 · $0.10/metric-mo next 240,000 · $0.05/metric-mo next 750,000 · $0.02/metric-mo over 1,000,000Tiers apply account-wide, not per namespace. The marginal rate on your next metric depends on how many you already publish./observability-costs#cw-custom-metrics
- GetMetricData API requests — $0.01 per 1,000 requestsEvery dashboard auto-refresh and every external tool polling CloudWatch counts here./observability-costs#cw-getmetricdata
- GetMetricWidgetImage API requests — $0.02 per 1,000 requestsCW:GMWI-Metrics/observability-costs#cw-getmetricwidgetimage
- Other API requests (PutMetricData, ListMetrics, etc.) — $0.01 per 1,000 requestsThe first 1,000,000 requests/month are free; only the overage bills./observability-costs#cw-api-requests
- OTel metrics via the CloudWatch OTLP endpoint — $0.50/GB ingestedA separate, byte-based meter from per-metric custom-metrics pricing. It's the CloudWatch parallel to Datadog's OTel-defaults-to-custom trap below./observability-costs#cw-otel-bytes
Data warehouse rates 20
- Virtual warehouse compute, X-Small to 6X-Large — 1 to 512 credits/hour, doubling at every sizeX-Small=1, Small=2, Medium=4, Large=8, X-Large=16, 2X-Large=32, 3X-Large=64, 4X-Large=128, 5X-Large=256, 6X-Large=512. Snowflake does not publish the $/credit…/data-warehouse-costs#sf-warehouse-credits
- Compute billing granularity — Per second, after the resume minimumWarehouse compute/data-warehouse-costs#sf-per-second-billing
- Resume minimum — 60 secondsEvery resume from auto-suspend bills at least 60 seconds of compute, even if the query that woke it finishes in a few seconds./data-warehouse-costs#sf-resume-minimum
- Auto-suspend / auto-resume — Both enabled by defaultSnowflake suspends a warehouse after its configured idle timeout, and resumes it automatically when a new query needs it./data-warehouse-costs#sf-auto-suspend-resume-defaults
- Cloud services credits — Billed only above 10% of daily compute creditsAccount-level daily threshold/data-warehouse-costs#sf-cloud-services-threshold
- Time Travel retention, Standard Edition — 1 day, fixedPer table/schema/database/data-warehouse-costs#sf-time-travel-standard
- Time Travel retention, Enterprise Edition — 0 to 90 days, configurablePer table/schema/database/data-warehouse-costs#sf-time-travel-enterprise
- Fail-safe — 7 days, non-configurablePermanent tables only, after Time Travel ends/data-warehouse-costs#sf-failsafe
- On-demand analysis — $6.25/TiB scannedThe first 1 TiB scanned per month is free./data-warehouse-costs#bq-ondemand-scan
- Minimum billed size per query — 10 MiBPer query, per referenced table/data-warehouse-costs#bq-min-bytes-per-query
- Editions, Standard — $0.04/slot-hour PAYG · $0.036 1-yr · $0.032 3-yrPer slot, per hour/data-warehouse-costs#bq-editions-standard
- Editions, Enterprise — $0.06/slot-hour PAYG · $0.054 1-yr · $0.048 3-yrPer slot, per hour/data-warehouse-costs#bq-editions-enterprise
- Editions, Enterprise Plus — $0.10/slot-hour PAYG · $0.09 1-yr · $0.08 3-yrPer slot, per hour/data-warehouse-costs#bq-editions-enterprise-plus
- Storage, active, logical billing — $0.023/GiB/moUncompressed bytes/data-warehouse-costs#bq-storage-active-logical
- Storage, long-term, logical billing — $0.016/GiB/moUncompressed bytes, unmodified 90+ days/data-warehouse-costs#bq-storage-longterm-logical
- Storage, active, physical billing — $0.04/GiB/moCompressed bytes/data-warehouse-costs#bq-storage-active-physical
- Storage, long-term, physical billing — $0.02/GiB/moCompressed bytes, unmodified 90+ days/data-warehouse-costs#bq-storage-longterm-physical
- Storage free tier — First 10 GiB/monthAll storage billing modes/data-warehouse-costs#bq-storage-free-tier
- Storage Read API — $1.10/TiBFirst 300 TiB/month free./data-warehouse-costs#bq-storage-read-api
- Storage Write API — $0.025/GiBFirst 2 TiB/month free./data-warehouse-costs#bq-storage-write-api
Playbook 7
- Why stack costs get out of controlCost overruns are rarely one bad decision. They're the compounding effect of dozens of normal decisions made without a daily signal./playbook#quiet-cost-creep
- One normalized view across cloud, AI, and SaaSNormalization (one schema, one currency, one cadence, one dedupe rule) is what turns scattered bills into a single number everyone trusts./playbook#one-normalized-view
- Cut AWS & cloud costs without slowing the team downThe biggest reclaimable cloud costs are almost always idle or oversized resources nobody is actively watching, not the workloads doing real work./playbook#cut-cloud-costs
- Understand and control AI / LLM spendAI spend doesn't behave like a cloud bill. It's usage-based and can multiply overnight, so it needs a daily, per-workload view, not a monthly total./playbook#control-ai-spend
- FinOps that fits a team that isn't hugeFinOps is a set of habits, not a headcount. A team of five can run the core loop with one rotating owner and thirty minutes a month./playbook#finops-for-small-teams
- Pick tooling that doesn't take a cut of your savingsA cost tool priced as a percentage of your spend, or of the savings it finds, gets more expensive as you succeed at cost control./playbook#pick-tooling-that-doesnt-take-a-cut
- Catch anomalies before the invoice: alerts & guardrailsA normalized daily view only pays off if something is watching it. Alerting turns a dashboard people could check into a signal that reaches them automatically./playbook#alerts-and-guardrails
Checklist 26
- Rightsize or stop idle computeCompare CPU/memory utilization against instance size; downsize or stop instances running well under their provisioned capacity./checklist#chk-rightsize-idle-instances
- Clear orphaned volumes, snapshots, and IPsAudit for unattached EBS volumes, snapshots past their retention need, and unassociated Elastic IPs. All are billed whether anything uses them or not./checklist#chk-clear-orphaned-resources
- Schedule down non-production environmentsStop dev, staging, and test instances outside business hours instead of leaving them running around the clock./checklist#chk-schedule-non-prod
- Revisit autoscaling policies against real trafficCheck scale-out thresholds and max instance limits against actual peak traffic, not the peak you originally provisioned for./checklist#chk-review-autoscaling-policies
- Move cold data to a cheaper storage tierIdentify objects and backups that haven't been accessed in months and move them to infrequent-access or archive tiers./checklist#chk-tier-cold-storage
- Audit cross-AZ and egress data transferCross-region, cross-AZ, and internet egress traffic bills per GB. Check whether services could be colocated to avoid it./checklist#chk-audit-data-transfer
- Set lifecycle policies instead of manual cleanupAutomate expiration and tiering rules on buckets and volumes so cleanup doesn't depend on someone remembering to do it./checklist#chk-set-lifecycle-policies
- Review backup frequency and retentionConfirm backup cadence and retention windows match your actual recovery requirements, not a default nobody has revisited./checklist#chk-review-backup-retention
- Track cost per model, not one blended totalBreak spend down by model and by workload. A blended total hides whether an increase is a deliberate upgrade or a regression./checklist#chk-track-cost-per-model
- Set an expected cost per request before shippingEstimate a rough cost per request or per task before a feature ships, so a regression is visible immediately, not a month later./checklist#chk-set-expected-cost-per-request
- Watch for agent-loop and retry fan-outA single logic bug in an agent loop or retry policy can multiply call volume as much as per-call cost. Watch call counts as well as spend./checklist#chk-watch-agent-loop-fanout
- Use prompt caching for repeated contextReuse cached context for calls that repeat the same system prompt or reference documents instead of resending it every time./checklist#chk-use-prompt-caching
- Compare models on cost-per-outcomeThe cheapest model per call isn't always the cheapest per completed task. Compare on results, not the sticker price./checklist#chk-compare-cost-per-outcome
- Revisit reserved instances & savings plans quarterlyCompare current commitment coverage against your actual footprint. Commitments sized for last year's usage rarely fit this year's./checklist#chk-revisit-committed-use
- Match commitment term to confidence in the workloadUse shorter or convertible commitments for workloads you're not confident will run steady for the full commitment term./checklist#chk-match-commitment-term
- Check for underutilized or expiring commitmentsFlag reserved capacity sitting unused, or commitments expiring soon, so you can renew, resize, or let them lapse on purpose./checklist#chk-check-underutilized-commitments
- Revisit enterprise discount agreements as spend growsProvider discount tiers and enterprise agreements are usually renegotiable once your spend has grown past the terms they were set at./checklist#chk-revisit-discount-agreements
- Normalize spend into one daily viewPull every provider's billing data into the same schema, currency, and cadence so totals are actually comparable./checklist#chk-normalize-spend-into-one-view
- Alert on deviation from a service's own baselineA single service moving sharply off its own historical trend is a stronger signal than total spend moving at all./checklist#chk-alert-on-baseline-deviation
- Route cost alerts to an owner, not a shared inboxAn alert that lands on a specific engineer's radar gets acted on; one that lands in a shared inbox decays into being ignored./checklist#chk-route-alerts-to-owners
- Turn on the hard cap and anomaly detector your provider already gives youAWS, Anthropic, and OpenAI can each stop spend outright once it crosses a number you set. Google Cloud and Azure alert but don't cap; wire your own automation…/checklist#chk-turn-on-native-guardrails
- Put a recurring spend review on the calendarA short, regular review against a normalized view catches drift while it's small, instead of at invoice time./checklist#chk-put-review-on-the-calendar
- Tag spend by service, team, or projectConsistent tagging at the point resources are created makes it possible to attribute cost instead of just seeing a total./checklist#chk-tag-spend-by-owner
- Use read-only, cost-scoped credentials for toolingAny tool that only needs to read billing data should never be granted write access to workloads or infrastructure./checklist#chk-use-read-only-credentials
- Audit unused SaaS seats and shadow toolsCross-check active users against paid seats for every SaaS subscription. An unused seat is pure waste with no offsetting benefit./checklist#chk-audit-unused-saas-seats
- Decommission cleanly when a project endsKeep a checklist for tearing down a project's resources when it ends, instead of relying on someone remembering to look later./checklist#chk-decommission-cleanly
Runaway bills 8
- A 3 MB MP3 turns into a $104,500 bandwidth billjyutping.org is a free-tier static site. It normally used well under 10 GB of bandwidth a month. A sudden traffic spike pushed daily bandwidth to a peak of 60.…/runaway-bills#netlify-mp3-ddos
- A viral week turns Cara's Vercel bill into $96,280Cara is a social app for artists. It positions itself against AI-generated art on larger platforms. It grew from 40,000 to 650,000 users in a week as creatives…/runaway-bills#cara-vercel-viral-week
- A Lambda function that triggered itself runs up ~$10,000 in a weekA Lambda function reacted to S3 API activity via CloudTrail. It wrote back to the same bucket that triggered it. This created a self-sustaining invocation loop…/runaway-bills#lambda-s3-recursive-loop
- Leaked AWS keys spin up ~600 mining instances overnightAWS access keys committed to a public GitHub repository were found by an automated bot within hours. Attackers used them to spin up roughly 600 EC2 instances a…/runaway-bills#hellyer-aws-leaked-keys
- S3 keys leak to GitHub, bots spin up 140 servers in five minutesA Rails security tool accidentally published Andrew Hoffman's Amazon S3 keys to a public GitHub repository. He pulled the keys within five minutes of noticing.…/runaway-bills#hoffman-aws-github-keys
- An enterprise skips usage limits, spends $500 million on Claude in a monthAn AI consultant told Axios that one of their enterprise clients rolled Claude out company-wide without per-user spending caps or usage limits. Engineers runni…/runaway-bills#unnamed-enterprise-claude-500m
- 100 coding agents run up $1.3 million in OpenAI tokens in 30 daysPeter Steinberger, creator of the open-source OpenClaw project, posted a screenshot showing $1,305,088.81 in OpenAI API charges over 30 days. That was 603 bill…/runaway-bills#openclaw-openai-1-3m
- AWS's own billing console briefly shows customers trillion-dollar estimatesA unit-pricing bug in AWS's estimated-billing computation subsystem briefly showed some customers cost forecasts. They ran from a few million dollars up to rou…/runaway-bills#aws-billing-console-bug
Connectors 14
- AWSCost Explorer/solutions/aws-cost-monitoring
- GCPBigQuery billing export/solutions/gcp-cost-monitoring
- AzureCost Management API/solutions/azure-cost-monitoring
- AnthropicAdmin API cost report/solutions/ai-cost-monitoring
- OpenAIOrg usage + costs API/solutions/openai-cost-monitoring
- Amazon BedrockCost Explorer + token usage/solutions/bedrock-cost-monitoring
- Vertex AI / GeminiBilling export + token usage/solutions/vertex-ai-cost-monitoring
- HeliconeLLM gateway usage/solutions/helicone-cost-monitoring
- SnowflakeOrganization usage views/solutions/snowflake-cost-monitoring
- DatabricksSQL warehouse usage/solutions/databricks-cost-monitoring
- DatadogUsage Metering API/solutions/datadog-cost-monitoring
- GitHubActions & Copilot usage/solutions/github-cost-monitoring
- VercelBilling charges API/solutions/vercel-cost-monitoring
- CSV importany tool without a native connector/solutions#manual
Price changes 32
- AWS started charging for every public IPv4 address, attached or not.Every public IPv4 address across AWS started accruing an hourly charge. This holds if the address is attached to a running instance. It also holds if the addre…/price-changes#aws-public-ipv4-charge
- AWS paired the new IPv4 charge with 750 free IPv4-hours a month, for new accounts only.New AWS accounts get 750 hours a month of public IPv4 usage at no charge when they launch an EC2 instance with a public address. That's enough to cover one alw…/price-changes#aws-ipv4-free-tier-750-hours
- AWS cut GuardDuty's S3 malware-scanning price by 85%.The per-GB price for the data-scanned dimension of Malware Protection for S3 dropped from $0.60 to $0.09 in US East (N. Virginia), applied automatically to exi…/price-changes#aws-guardduty-malware-protection-s3-cut
- AWS cut On-Demand GPU instance prices by up to 45%.On-Demand pricing dropped across the P5, P5en, P4d, and P4de instance families. P5 dropped by up to 45%. P5en dropped by up to 26%. P4d/P4de dropped by up to 3…/price-changes#aws-ec2-gpu-instances-price-cut-2025
- VPC Encryption Controls stopped being a free preview and started billing per VPC-hour.After a free preview that ran from November 2025, AWS began charging a fixed hourly rate. The rate applies to every non-empty VPC that runs Encryption Controls…/price-changes#aws-vpc-encryption-controls-new-charge
- Google Cloud raised the hourly charge on in-use external IPv4 addresses.The External IP Charge applies to every in-use external IPv4 address. For a standard VM it rose from $0.004 to $0.005 an hour. For a Spot VM it rose from $0.00…/price-changes#gce-external-ipv4-price-increase-2024
- BigQuery raised on-demand query pricing 25% the same day it retired flat-rate slots.Google retired flat-rate annual, flat-rate monthly, and flex-slot commitments. Three new editions, billed per-second, replace them. On the same date, Google ra…/price-changes#bigquery-editions-on-demand-increase-2023
- Cloud SQL began billing instances left running on end-of-life MySQL or PostgreSQL versions.Instances that still run a MySQL or PostgreSQL major version past its community end-of-life get auto-enrolled in a paid "extended support" add-on. Google bills…/price-changes#cloud-sql-extended-support-new-charge-2025
- Cloud Build repriced its e2 machines by region, raising build minutes in some and trimming others.Per-build-minute pricing for e2 machine types moved from one flat rate to a region-by-region table. In us-east1 (South Carolina), the price went up across e2-m…/price-changes#cloud-build-e2-regional-price-update-2025
- GKE folded paid multi-cluster management into the Standard tier at no extra charge.As part of moving to a single paid GKE tier, Fleets, Teams, Config Management, and Policy Controller are now included with GKE Standard. All four were previous…/price-changes#gke-single-tier-multicluster-free-2025
- Azure raised local-currency prices in the UK, EU, and Nordics to track the USD exchange rate.Microsoft moved to a twice-yearly currency-alignment cadence and raised local-currency Azure pricing to track USD: GBP up 9%, DKK/EUR/NOK up 11%, SEK up 15%./price-changes#azure-currency-price-alignment-2023
- Azure zeroed out the bandwidth charge for traffic between Availability Zones in the same region.Microsoft eliminated the per-GB charge for data moving across Availability Zones within a region, for both private and public IP traffic./price-changes#azure-inter-az-bandwidth-free-2024
- Azure cut Ultra Disk IOPS and throughput pricing in UK South by up to 80%.In UK South, provisioned IOPS on Ultra Disk dropped from $0.06205 to $0.02482 per month. Provisioned throughput dropped from $0.40588 to $0.08103 per MBPS per…/price-changes#azure-ultra-disk-price-cut-uk-south-2025
- GPT-4o launched at half the price of GPT-4 Turbo, with 5x the rate limit.OpenAI's new flagship model replaced GPT-4 Turbo as the default, matching or beating its quality at half the per-token price./price-changes#gpt-4o-half-price-vs-gpt-4-turbo
- GPT-5 replaced the GPT-4o/o-series lineup with one model family sold at three price points.GPT-5 launched at $1.25 per million input tokens and $10 per million output tokens. Mini and nano tiers sit underneath it. It is one family instead of separate…/price-changes#gpt-5-launch-pricing
- GPT-5.5 launched priced higher than GPT-5.4, betting token efficiency would offset it.GPT-5.5 launched at roughly double GPT-5.4's per-token rate. OpenAI's own announcement frames this as a deliberate increase, offset by claimed gains in token e…/price-changes#gpt-5-5-price-increase
- OpenAI cut GPT-5.6 Luna's price 80% and Terra's 20%, leaving the flagship tier untouched.Citing inference-stack efficiency gains, OpenAI dropped Terra to $2/$12 per million tokens and Luna to $0.20/$1.20, while the top-tier Sol model's price stayed…/price-changes#gpt-5-6-luna-terra-price-cut
- Anthropic launched a batch endpoint priced at half the standard API rate.The Message Batches API processes up to 10,000 queries asynchronously within 24 hours, at half the price of a standard synchronous call./price-changes#anthropic-message-batches-api-50-off
- Anthropic cut tool-use output tokens on Claude 3.7 Sonnet by up to 70%.A new token-efficient tool-use mode cut output tokens consumed on tool-calling workloads, and prompt-cache reads stopped counting against input-token-per-minut…/price-changes#anthropic-token-saving-updates-2025
- Claude Haiku 4.5 launched at a third of Sonnet 4's price for comparable coding performance.The new small model launched at a flat $1 input / $5 output per million tokens. Anthropic positions it as Sonnet-4-level coding at a third of the cost./price-changes#claude-haiku-4-5-pricing
- Claude Opus 4.5 cut Opus-tier pricing to $5/$25 per million tokens.Opus 4.5 launched at $5 input / $25 output per million tokens, a cut Anthropic frames as putting frontier-tier capability within reach of more teams./price-changes#claude-opus-4-5-pricing-cut
- Gemini 2.5 Flash's stable release raised input price but cut output price and merged two rate tiers into one.Going stable, Google removed 2.5 Flash's separate thinking vs. non-thinking price tiers, raising input from $0.15 to $0.30 per million tokens while cutting out…/price-changes#gemini-2-5-flash-stable-repricing-2025
- Gemini 3 Pro launched as Google's new flagship, at a higher output rate than 2.5 Pro.Google's new flagship reasoning and agentic model launched at $2 per million input tokens. Output costs $12 per million tokens. Both rates apply to prompts up…/price-changes#gemini-3-pro-launch-pricing-2025
- Gemini 3 Flash turned on context caching by default, cutting repeated-token cost up to 90%.Gemini 3 Flash launched at $0.50 per million input tokens and $3 per million output tokens. Context caching is standard rather than opt-in. Asynchronous Batch…/price-changes#gemini-3-flash-caching-batch-pricing-2025
- Claude Opus 5.5 launched 20% cheaper than Opus 5, with cache reads down 60%.Opus 5.5 replaced Opus 5 at $4 input and $20 output per million tokens. Both prices are 20% lower than Opus 5. Cache reads fell from $0.50 to $0.20 per million…/price-changes#anthropic-opus-5-5-price-cut
- GPT-6 Sol and Luna launched at half their GPT-5.6 promotional price.OpenAI priced GPT-6 Sol at $2 input and $10 output per million tokens. GPT-6 Luna launched at $0.10 input and $0.50 output per million tokens. Both prices are…/price-changes#openai-gpt-6-sol-luna-price-cut
- GitHub made Copilot free in VS Code, capped at 2,000 completions a month.Any GitHub account gets capped code completions and chat in VS Code. The cap is 2,000 completions and 50 chat messages a month. No trial or subscription is req…/price-changes#github-copilot-free-tier-2024
- MongoDB merged its Shared and Serverless tiers into one Flex tier with a capped bill.Atlas Flex bills an $8 base fee plus usage up to 500 ops/sec, capped at $30 a month. It replaces both the old Shared clusters and standalone Serverless instanc…/price-changes#mongodb-atlas-flex-tier-2025
- Vercel switched Functions billing to active CPU time instead of full wall-clock duration.A Standard-size function running at 100% active CPU now costs about $0.149 an hour instead of $0.318. Idle time waiting on a database call or an LLM response n…/price-changes#vercel-active-cpu-pricing-2025
- Snowflake moved Snowpipe to one flat per-GB rate, cutting ingestion cost for some teams over 50%.Snowpipe pricing, previously based on compute consumed and file count, moved to a single 0.0037-credits-per-GB rate across file ingestion and streaming./price-changes#snowflake-snowpipe-unified-pricing-2025
- GitHub replaced Copilot's flat premium-request billing with usage-based AI credits.Every Copilot plan now gets a monthly allotment of GitHub AI Credits. Actual token usage uses up the credits. Before, a flat per-request charge billed a quick…/price-changes#github-copilot-ai-credits-2026
- Vercel cut Turbo build machine pricing 16% and unified every tier to one per-CPU rate.All build machine tiers now bill at a flat $0.0035 per CPU per minute, dropping the 30-CPU Turbo machine from $0.126 to $0.105 a minute./price-changes#vercel-turbo-build-machines-price-cut-2026
FOCUS cost columns 4
- BilledCostThe charge that forms the basis for invoicing./focus-finops-standard#billed-cost
- EffectiveCostThe amortized cost after every discount is applied./focus-finops-standard#effective-cost
- ListCostList unit price × pricing quantity, no discounts applied./focus-finops-standard#list-cost
- ContractedCostContracted unit price × pricing quantity./focus-finops-standard#contracted-cost
FAQ 310
- How does CostMon connect to AWS?Through a read-only AWS Cost Explorer connector, scoped to billing data only. CostMon never requests access to your workloads, source code, or production infra…/solutions/aws-cost-monitoring/#faq-q-0
- Does CostMon replace AWS Cost Explorer?No. CostMon reads from Cost Explorer. Your AWS bill then sits beside Anthropic, OpenAI, Amazon Bedrock and any other source you add. One dashboard shows every…/solutions/aws-cost-monitoring/#faq-q-1
- How often does AWS cost data sync?CostMon's background scheduler syncs AWS Cost Explorer data on a regular cadence, so the dashboard reflects a daily view instead of a monthly snapshot./solutions/aws-cost-monitoring/#faq-q-2
- Can I see AWS costs broken down by service or account?Yes. CostMon records the service and linked account for every AWS row. Trends split by provider or by service. For team totals, write an allocation rule that m…/solutions/aws-cost-monitoring/#faq-q-3
- What does the AWS connector cost?A flat monthly rate per plan. It never charges a percentage of your AWS bill or the savings CostMon helps you find. Free includes 2 connectors; Professional an…/solutions/aws-cost-monitoring/#faq-q-4
- Which AI providers does CostMon support today?Five AI connectors are live today. Anthropic, OpenAI and Amazon Bedrock are three of them. Vertex AI / Gemini and Helicone are the others. Manual entries let y…/solutions/ai-cost-monitoring/#faq-q-0
- Can CostMon break down AI spend by model or agent?Yes. Claude API cost data syncs in with its underlying detail intact. You can see trends by model, by workload, or grouped into cost buckets by team or project./solutions/ai-cost-monitoring/#faq-q-1
- How does CostMon get Anthropic cost data?Through the Anthropic Admin API's cost report endpoint, read-only and scoped to billing data. CostMon never requests access to prompts, completions, or your An…/solutions/ai-cost-monitoring/#faq-q-2
- How is this different from the Anthropic Console?The Anthropic Console shows Claude usage on its own. CostMon normalizes that same cost data alongside AWS and every other provider you connect. AI spend sits i…/solutions/ai-cost-monitoring/#faq-q-3
- What does the Anthropic connector cost?A flat monthly rate per plan. CostMon never charges a percentage of your AI spend or the savings it helps you find. Free includes 2 connectors; Professional an…/solutions/ai-cost-monitoring/#faq-q-4
- How does CostMon connect to GCP?Through a read-only service account key scoped to bigquery.dataViewer and bigquery.jobUser on the dataset holding your FOCUS-formatted billing export. CostMon…/solutions/gcp-cost-monitoring/#faq-q-0
- Does CostMon replace the GCP Billing console?No. CostMon reads the same FOCUS billing export. GCP then shows beside AWS, Azure and your AI providers on a single daily chart./solutions/gcp-cost-monitoring/#faq-q-1
- Can CostMon see spend across multiple projects?Yes. The billing export carries project detail, and CostMon keeps it. You can break spend down by project, service, SKU or region./solutions/gcp-cost-monitoring/#faq-q-2
- Does CostMon track committed-use discounts?Not by end date. CostMon does not read commitment IDs or expiry dates from the export. It does show total GCP spend each day, so a lapsed discount shows as a r…/solutions/gcp-cost-monitoring/#faq-q-3
- What does the GCP connector cost?A flat monthly rate per plan. It's never a percentage of your GCP bill or the savings CostMon helps you find. Free includes 2 connectors; Professional and Ente…/solutions/gcp-cost-monitoring/#faq-q-4
- How does CostMon connect to Azure?Through an Azure AD app registration (service principal) granted Cost Management Reader on the subscription, read-only and scoped to billing data. CostMon neve…/solutions/azure-cost-monitoring/#faq-q-0
- Does CostMon replace Azure Cost Management?No. CostMon reads from the Cost Management Query API. Azure spend then appears on the same chart as AWS, GCP and the rest of your stack./solutions/azure-cost-monitoring/#faq-q-1
- How often does Azure cost data sync?CostMon's background scheduler syncs Azure Cost Management data on a regular cadence, so the dashboard reflects a daily view rather than a monthly statement./solutions/azure-cost-monitoring/#faq-q-2
- Can I see Azure costs broken down by service or region?Yes. Every Azure row carries a service name and a region. To roll spend up by team, match an allocation rule on the Azure source, a service name or the connect…/solutions/azure-cost-monitoring/#faq-q-3
- What does the Azure connector cost?A flat monthly rate per plan. It's never a cut of your Azure bill or the savings CostMon helps you find. Free includes 2 connectors; Professional and Enterpris…/solutions/azure-cost-monitoring/#faq-q-4
- How does CostMon connect to OpenAI?Through an organization Admin API key (sk-admin-...), scoped to the usage and costs report endpoints only. Regular project API keys can't grant this access, an…/solutions/openai-cost-monitoring/#faq-q-0
- Is OpenAI a live connector or still coming soon?Live today. CostMon merges the Admin API's usage and costs reports per day and per model into the same normalized view as every other provider you connect./solutions/openai-cost-monitoring/#faq-q-1
- Can CostMon break down OpenAI spend by model or project?Yes. Usage and cost data merge on day and model. You can optionally scope the connector to a single project. You see trends by model, by project, or grouped in…/solutions/openai-cost-monitoring/#faq-q-2
- How is this different from the OpenAI usage dashboard?The OpenAI dashboard shows usage on its own. CostMon normalizes that same cost data alongside AWS, Anthropic, and every other provider you connect. OpenAI spen…/solutions/openai-cost-monitoring/#faq-q-3
- What does the OpenAI connector cost?A flat monthly rate per plan. It never charges a percentage of your OpenAI spend or the savings CostMon helps you find. Free includes 2 connectors; Professiona…/solutions/openai-cost-monitoring/#faq-q-4
- How does CostMon connect to Amazon Bedrock?Through the same read-only AWS role as the AWS Cost Explorer connector, scoped to ce:GetCostAndUsage plus CloudWatch's ListMetrics and GetMetricData. CostMon n…/solutions/bedrock-cost-monitoring/#faq-q-0
- How does CostMon break Bedrock spend down by model?Bedrock usage types from Cost Explorer are parsed back into model IDs. They are merged with CloudWatch's per-model InputTokenCount and OutputTokenCount metrics…/solutions/bedrock-cost-monitoring/#faq-q-1
- What if CloudWatch token metrics aren't available for a model?Bedrock spend still syncs from Cost Explorer either way. Token-level detail is best-effort enrichment, so a gap in CloudWatch metrics means less granularity. T…/solutions/bedrock-cost-monitoring/#faq-q-2
- Does CostMon track Provisioned Throughput separately from on-demand usage?Provisioned Throughput commitments bill hourly regardless of traffic and show up as cost with little or no corresponding token usage. That's the pattern that f…/solutions/bedrock-cost-monitoring/#faq-q-3
- What does the Bedrock connector cost?A flat monthly rate per plan. CostMon never takes a percentage of your Bedrock spend or the savings it helps you find. Free includes 2 connectors; Professional…/solutions/bedrock-cost-monitoring/#faq-q-4
- How does CostMon connect to Vertex AI?It uses the same read-only service account as the GCP connector. That account has bigquery.dataViewer and jobUser on your billing export dataset. It also has m…/solutions/vertex-ai-cost-monitoring/#faq-q-0
- Is Vertex AI spend separate from the rest of my GCP bill?No. It comes from the same GCP billing export as everything else. CostMon filters it to Vertex AI and generative AI services. It joins the rows with per-model…/solutions/vertex-ai-cost-monitoring/#faq-q-1
- Can CostMon tell me if I should be using the Batch API?Not directly. CostMon shows daily spend and token counts for each model. A high-volume model stands out there. The token metric covers online serving only, so…/solutions/vertex-ai-cost-monitoring/#faq-q-2
- Does CostMon track context caching?No. CostMon reads two token directions from Cloud Monitoring: input and output. It has no count of cached tokens. It does show the spend for each model, so a c…/solutions/vertex-ai-cost-monitoring/#faq-q-3
- What does the Vertex AI connector cost?A flat monthly rate per plan. It never charges a percentage of your Vertex AI spend or the savings CostMon helps you find. Free includes 2 connectors; Professi…/solutions/vertex-ai-cost-monitoring/#faq-q-4
- What does the Helicone connector pull?It pulls Helicone's gateway request log: cost, tokens and model for each request. CostMon groups these by day, model, upstream provider and workload. Any other…/solutions/helicone-cost-monitoring/#faq-q-0
- What does my app need to do to get per-customer attribution?Two things. First, tag requests with Helicone custom properties (the Helicone-Property-* headers). Use one for whatever you want to attribute by, such as a cus…/solutions/helicone-cost-monitoring/#faq-q-1
- How is this different from Helicone's own dashboard?Helicone's dashboard shows gateway usage on its own. CostMon reads that same data. It puts it next to AWS, Anthropic, OpenAI and your other providers. AI spend…/solutions/helicone-cost-monitoring/#faq-q-2
- Does CostMon see my prompts or completions through Helicone?No. The connector reads cost and usage data only, scoped to a read-only Helicone API key. It never requests access to prompt or completion content./solutions/helicone-cost-monitoring/#faq-q-3
- What does the Helicone connector cost?A flat monthly rate per plan, never a cut of your AI spend or the savings CostMon helps you find. Free includes 2 connectors; Professional and Enterprise inclu…/solutions/helicone-cost-monitoring/#faq-q-4
- How does CostMon connect to Snowflake?Through a Programmatic Access Token scoped to Snowflake's built-in ORGANIZATION_USAGE_VIEWER role, read-only and scoped to usage and billing views. CostMon nev…/solutions/snowflake-cost-monitoring/#faq-q-0
- How often does Snowflake cost data sync?On CostMon's regular background schedule, reading a trailing window to account for Snowflake's own reporting lag on organization usage views. Recent days get c…/solutions/snowflake-cost-monitoring/#faq-q-1
- Can I see Snowflake spend broken down by individual warehouse?Not by warehouse name. CostMon reports at the level Snowflake's organization usage views give: service type, usage type, account and region, for each day. That…/solutions/snowflake-cost-monitoring/#faq-q-2
- Does CostMon replace Snowflake's own usage views?No. CostMon reads the same organization usage views. Snowflake credits then sit beside your cloud and AI bills, so you need no separate login. Snowflake's ware…/solutions/snowflake-cost-monitoring/#faq-q-3
- What does the Snowflake connector cost?A flat monthly rate per plan. It never charges a percentage of your Snowflake spend or the savings CostMon helps you find. Free includes 2 connectors; Professi…/solutions/snowflake-cost-monitoring/#faq-q-4
- How does CostMon connect to Databricks?Through a read-only personal access token or OAuth service-principal token that runs a single SELECT-style statement against system.billing.usage via the State…/solutions/databricks-cost-monitoring/#faq-q-0
- Are the dollar amounts CostMon shows for Databricks exact?DBU consumption is exact, sourced directly from system.billing.usage. Dollar amounts are estimated against Databricks' list price rather than your negotiated o…/solutions/databricks-cost-monitoring/#faq-q-1
- Can CostMon tell me if a job is running on the wrong cluster type?system.billing.usage records the SKU for every row, which distinguishes jobs, all-purpose, and SQL warehouse compute. CostMon shows that breakdown, so an all-p…/solutions/databricks-cost-monitoring/#faq-q-2
- Does CostMon also track the cloud VM cost underneath Databricks?On classic (non-serverless) AWS and GCP compute, the underlying VM, storage, and network cost is billed separately by the cloud provider. Connect that provider…/solutions/databricks-cost-monitoring/#faq-q-3
- What does the Databricks connector cost?A flat monthly rate per plan. It's never a percentage of your Databricks spend or the savings CostMon helps you find. Free includes 2 connectors; Professional…/solutions/databricks-cost-monitoring/#faq-q-4
- How does CostMon connect to Datadog?Through a Datadog API key and application key scoped to the Usage Metering API's read-only cost endpoints. CostMon never requests access to your monitors, dash…/solutions/datadog-cost-monitoring/#faq-q-0
- Does CostMon replace Datadog's own usage dashboard?No. CostMon reads the same Usage Metering API. Your Datadog month-to-date figure then shows next to your AWS, GCP and model spend./solutions/datadog-cost-monitoring/#faq-q-1
- Can CostMon break Datadog spend down by product?Yes. Each cost row keeps its product and its committed or on-demand charge type. You can see trends for APM, logs, infrastructure or custom metrics. An allocat…/solutions/datadog-cost-monitoring/#faq-q-2
- How often does Datadog cost data sync?The background scheduler syncs every hour by default. It uses Datadog's estimated-cost endpoint for the current and previous month. It uses historical cost for…/solutions/datadog-cost-monitoring/#faq-q-3
- What does the Datadog connector cost?A flat monthly rate per plan. It never charges a cut of your Datadog spend or the savings CostMon helps you find. Free includes 2 connectors; Professional and…/solutions/datadog-cost-monitoring/#faq-q-4
- How does CostMon connect to GitHub?Through a fine-grained personal access token scoped to Administration: read for an organization, or a classic token with admin:enterprise for enterprise-wide b…/solutions/github-cost-monitoring/#faq-q-0
- Does CostMon break Actions spend down by runner type?Yes. GitHub's billing API reports usage for each date, product and SKU. A SKU names the runner OS and size. You see Linux, Windows and macOS minutes apart inst…/solutions/github-cost-monitoring/#faq-q-1
- Does CostMon track GitHub Copilot spend too?Yes, as far as GitHub reports it. Copilot is one of the products in GitHub's billing usage API. CostMon syncs whatever that API returns for Copilot. It sits ne…/solutions/github-cost-monitoring/#faq-q-2
- Does CostMon show per-repository cost?GitHub's billing API returns usage at the organization or enterprise level. CostMon stores it at that date, product, and SKU grain, rather than per repository…/solutions/github-cost-monitoring/#faq-q-3
- What does the GitHub connector cost?A flat monthly rate per plan. CostMon never takes a percentage of your GitHub spend or the savings it helps you find. Free includes 2 connectors; Professional…/solutions/github-cost-monitoring/#faq-q-4
- How does CostMon connect to Vercel?Through a Vercel API token that only calls the read-only billing charges endpoint. CostMon never requests access to your deployments, source code, or project c…/solutions/vercel-cost-monitoring/#faq-q-0
- Does CostMon break Vercel spend down by meter?Yes. Vercel's FOCUS-formatted billing export lists line items for each service and SKU. Bandwidth, functions, Image Optimization, ISR and build minutes each ge…/solutions/vercel-cost-monitoring/#faq-q-1
- Can CostMon catch a traffic-spike bill before it lands?CostMon syncs Vercel's billing charges daily. A spike on several meters shows as a rising daily line within days. You do not wait for the next invoice./solutions/vercel-cost-monitoring/#faq-q-2
- Does CostMon replace Vercel's own usage dashboard?No. CostMon reads the same billing charges data. Vercel hosting cost then appears on one daily chart with your cloud providers./solutions/vercel-cost-monitoring/#faq-q-3
- What does the Vercel connector cost?A flat monthly rate per plan. It never charges a percentage of your Vercel spend or the savings CostMon helps you find. Free includes 2 connectors; Professiona…/solutions/vercel-cost-monitoring/#faq-q-4
- Does Vantage charge based on how much cloud spend I track?Vantage's published pricing scales with the amount of cloud spend tracked through the platform. CostMon charges a flat rate per plan, regardless of how much sp…/compare/costmon-vs-vantage/#faq-q-0
- Can Vantage track AI or LLM spend like Anthropic or OpenAI?Vantage's core focus is cloud and infrastructure cost visibility. CostMon normalizes AI/LLM provider spend (Anthropic, OpenAI, and three more AI providers) in…/compare/costmon-vs-vantage/#faq-q-1
- Is CostMon a replacement for Vantage?Not necessarily. The two solve overlapping but different problems. Vantage is a broad, deeply integrated cost visibility platform. CostMon is a smaller, flat-p…/compare/costmon-vs-vantage/#faq-q-2
- Which one is easier to set up?Both connect with read-only, cost-scoped credentials. CostMon has a smaller catalog, which tends to make first setup faster. Vantage has a broader integration…/compare/costmon-vs-vantage/#faq-q-3
- Can we run CostMon and Vantage at the same time?Yes, and some teams do. Vantage gives deep, engineering-facing cost reports across a long integration tail. CostMon gives a flat-priced, finance-facing daily n…/compare/costmon-vs-vantage/#faq-q-4
- Does Vantage do cost anomaly detection?Yes. Per Vantage's own site, cost anomaly alerts and budget notifications are an established part of the platform. CostMon also ships anomaly rules and budgets…/compare/costmon-vs-vantage/#faq-q-5
- Is CloudZero self-serve like CostMon?No. CloudZero is positioned and sold through a sales-led, quote-based process. CostMon is self-serve. You sign up, connect a provider, and see normalized spend…/compare/costmon-vs-cloudzero/#faq-q-0
- Does CostMon do cost-per-customer allocation like CloudZero?In part. CostMon lets you define a unit, such as active customers, orders, or seats. It divides spend by that unit and charts the result over time. Allocation…/compare/costmon-vs-cloudzero/#faq-q-1
- Which is cheaper, CostMon or CloudZero?CostMon publishes flat per-plan pricing on its pricing page. CloudZero's pricing is custom and quote-based, so a direct comparison needs a quote. Check both pr…/compare/costmon-vs-cloudzero/#faq-q-2
- Can I track AI/LLM spend in CloudZero?CloudZero's focus is cloud infrastructure cost allocation. CostMon normalizes AI/LLM provider spend (Anthropic, OpenAI, and three more AI providers) in the sam…/compare/costmon-vs-cloudzero/#faq-q-3
- Can we run CostMon and CloudZero at the same time?Yes. Some teams use CloudZero for deep cost-per-customer and margin reporting. They add a simpler, flat-priced view like CostMon for the daily cloud-plus-AI nu…/compare/costmon-vs-cloudzero/#faq-q-4
- How long does CloudZero take to implement?CloudZero's own materials describe an onboarding process built around mapping your cost drivers to a unit-economics model. That takes longer than a plug-and-pl…/compare/costmon-vs-cloudzero/#faq-q-5
- Does CostMon replace Kubecost?Not for Kubernetes-specific cost allocation. Kubecost goes far deeper than CostMon on pod-, namespace-, and label-level cost inside a cluster. Some teams run b…/compare/costmon-vs-kubecost/#faq-q-0
- Can Kubecost track AI/LLM API spend like Anthropic?No. Kubecost's scope is Kubernetes infrastructure cost, not external AI provider billing. CostMon normalizes AI/LLM spend (Anthropic, OpenAI, and three more AI…/compare/costmon-vs-kubecost/#faq-q-1
- Is Kubecost free?Kubecost publishes free and paid tiers on its own site, scoped to Kubernetes cost monitoring. Check its current pricing page for details, because tiers and lim…/compare/costmon-vs-kubecost/#faq-q-2
- Do I need Kubernetes to use CostMon?No. CostMon monitors cloud provider billing (AWS, GCP, and Azure today), AI/LLM provider billing, and more. It does not matter whether your workloads run on Ku…/compare/costmon-vs-kubecost/#faq-q-3
- Can we run CostMon and Kubecost together?Yes, and it is a common pairing. Kubecost handles pod-, namespace-, and label-level chargeback inside your clusters. CostMon gives the whole-stack daily total…/compare/costmon-vs-kubecost/#faq-q-4
- Does Kubecost require installing something inside my clusters?Yes. Per Kubecost's own documentation, it runs as an in-cluster agent that reads Kubernetes metrics and cost allocation data. CostMon has no cluster-side compo…/compare/costmon-vs-kubecost/#faq-q-5
- Does CostMon replace AWS Cost Explorer?Not exactly. CostMon reads from AWS Cost Explorer's data through a read-only connector. It normalizes that data next to every other provider you connect. AWS s…/compare/costmon-vs-aws-cost-explorer/#faq-q-0
- Is AWS Cost Explorer free?It is included with an AWS account. Per AWS's own pricing page, AWS charges separately for Cost Explorer API calls beyond the free console usage. CostMon charg…/compare/costmon-vs-aws-cost-explorer/#faq-q-1
- Can AWS Cost Explorer track Anthropic or other AI provider spend?No. Cost Explorer only reports on AWS billing. CostMon normalizes AI/LLM provider spend (Anthropic, OpenAI, and three more AI providers) in the same daily view…/compare/costmon-vs-aws-cost-explorer/#faq-q-2
- Why use CostMon if Cost Explorer is already free?Cost Explorer is a query tool scoped to one AWS account at a time. CostMon adds a daily, standing signal. It normalizes every provider you run, cloud and AI. F…/compare/costmon-vs-aws-cost-explorer/#faq-q-3
- Can we use CostMon and Cost Explorer together?Yes. CostMon's AWS connector reads from the same underlying Cost Explorer data, so there is no conflict. Some teams keep Cost Explorer open for deep, ad-hoc AW…/compare/costmon-vs-aws-cost-explorer/#faq-q-4
- Does AWS offer its own anomaly detection I should compare against CostMon?Yes. AWS Cost Anomaly Detection is a separate, free AWS-native feature that can flag unusual AWS spend by itself. It is AWS-only, like Cost Explorer. CostMon a…/compare/costmon-vs-aws-cost-explorer/#faq-q-5
- Do I need to be a Datadog customer to use its Cloud Cost Management module?Cloud Cost Management is part of the Datadog platform. You adopt it with Datadog's broader monitoring products, not as a fully separate purchase. CostMon has n…/compare/costmon-vs-datadog-cloud-cost-management/#faq-q-0
- Does Datadog Cloud Cost Management track AI/LLM spend like CostMon does?Yes. Per Datadog's own product page, it covers AI provider spend, including OpenAI and Anthropic, next to cloud and SaaS costs. CostMon covers Anthropic, OpenA…/compare/costmon-vs-datadog-cloud-cost-management/#faq-q-1
- Is Cloud Cost Management priced separately from the rest of Datadog?Datadog does not publish standalone Cloud Cost Management pricing on its general pricing page. It is positioned as part of the broader Datadog platform. CostMo…/compare/costmon-vs-datadog-cloud-cost-management/#faq-q-2
- Which one gives better cost anomaly detection?Datadog's cost anomaly detection runs on top of the infrastructure and APM data it already collects. Per its own site, it adds AI-assisted root-cause tooling.…/compare/costmon-vs-datadog-cloud-cost-management/#faq-q-3
- Can we run CostMon alongside Datadog?Yes. Many teams already run Datadog for monitoring and could add Cloud Cost Management as a module. Others keep cost visibility in a separate, flat-priced tool…/compare/costmon-vs-datadog-cloud-cost-management/#faq-q-4
- Does CostMon require the Datadog agent or any Datadog integration?No. CostMon connects directly to provider billing APIs with read-only, cost-scoped credentials. It does not depend on Datadog or any observability agent. That…/compare/costmon-vs-datadog-cloud-cost-management/#faq-q-5
- Is Finout's pricing really flat, like CostMon's?Not quite. Finout describes its model as a flat platform fee with no per-seat charges. Per its own site, that fee is "based on the spend you manage". So it sca…/compare/costmon-vs-finout/#faq-q-0
- Does Finout cover more AI providers than CostMon?Yes, today. Finout's own site lists AI/LLM coverage including OpenAI and Anthropic, plus AI-driven developer tools and SaaS pricing models. CostMon covers Anth…/compare/costmon-vs-finout/#faq-q-1
- Which tool is simpler to set up?CostMon has a smaller catalog (13 live connectors plus CSV import). That generally means fewer integrations to configure before the number is trustworthy. Fino…/compare/costmon-vs-finout/#faq-q-2
- Is CostMon a replacement for Finout?Only if your footprint fits CostMon's smaller catalog. Finout is built for organizations with a large, mixed vendor footprint and a dedicated FinOps practice.…/compare/costmon-vs-finout/#faq-q-3
- Does Finout do automated cost optimization, not just visibility?Yes, per Finout's own site. It describes agentic detection, investigation, and approval-gated optimization workflows beyond a visibility dashboard. CostMon foc…/compare/costmon-vs-finout/#faq-q-4
- Can we run both CostMon and Finout?It is unusual, because the two overlap heavily in what they solve. It is not impossible. Both read from provider billing APIs independently and do not conflict…/compare/costmon-vs-finout/#faq-q-5
- Does Infracost track actual cloud spend like CostMon does?No. Per Infracost's own site, it estimates cost from infrastructure-as-code plans before deployment. It does not track billed, after-the-fact spend the way Cos…/compare/costmon-vs-infracost/#faq-q-0
- Can CostMon catch an expensive change before it is deployed, like Infracost?No. CostMon normalizes actual billing data after providers report it, so it cannot warn you before a change merges. Anomaly rules and budgets alert you once th…/compare/costmon-vs-infracost/#faq-q-1
- Should we use Infracost or CostMon?For most teams, this is not an either-or question. They solve different problems on opposite sides of a deploy. Infracost estimates cost before a change ships.…/compare/costmon-vs-infracost/#faq-q-2
- Does Infracost cover AI/LLM provider spend like Anthropic or OpenAI?No. Per its own site, Infracost is scoped to infrastructure-as-code cost estimation for cloud resources defined in Terraform, CloudFormation, CDK, or Kubernete…/compare/costmon-vs-infracost/#faq-q-3
- Can we run CostMon and Infracost together?Yes, and it is a natural pairing, not a redundant one. Infracost gives a pre-merge cost estimate and guardrail in CI/CD. CostMon gives the daily normalized tot…/compare/costmon-vs-infracost/#faq-q-4
- Is Infracost free?Infracost publishes a free CI/CD tier (1,000 runs a month). Per its own pricing page, paid tiers add team management, audit trails, and FinOps policy enforceme…/compare/costmon-vs-infracost/#faq-q-5
- What is cloud cost optimization?Cloud cost optimization is the ongoing practice of keeping cloud spend in line with what you need. You find and remove idle or oversized resources. You match c…/playbook#faq-q-0
- How is AI/LLM cost management different from cloud cost management?Cloud spend tracks infrastructure you deliberately provisioned, so it grows predictably. AI/LLM spend is metered per call or per token. A single agent loop or…/playbook#faq-q-1
- Do we need a dedicated FinOps team to do this well?No. The core loop works with one rotating owner and a review slot inside a meeting you already hold. Normalize spend into one view. Review it on a fixed cadenc…/playbook#faq-q-2
- How often should we review cloud and AI spend?A short monthly sweep against a normalized, daily-synced view catches the overwhelming majority of reclaimable waste and drifting AI spend. Use threshold-based…/playbook#faq-q-3
- What's the difference between percentage-based and flat-rate cost tools?Percentage-based tools charge a cut of your managed spend or of the savings they identify, so their revenue grows when your bill grows. That's a misaligned inc…/playbook#faq-q-4
- How does CostMon fit into this playbook?CostMon is the normalized daily view this playbook keeps returning to. It connects to these sources. AWS, GCP, and Azure for cloud. Anthropic, OpenAI, Amazon B…/playbook#faq-q-5
- Will AWS, Vercel, or Netlify refund a surprise bill like this?Sometimes, but there's no guarantee. Several incidents above were waived or refunded after the story reached Hacker News or Reddit and got public attention. Ne…/runaway-bills#faq-q-0
- Do spending limits or budget alerts prevent this?They help, but a limit nobody configures doesn't fire, and a monthly budget alert can still arrive weeks after a loop or a leak started. Several incidents abov…/runaway-bills#faq-q-1
- How fast can spend be detected, realistically?Every incident above had a leading indicator. It was a bandwidth spike, an instance count, a function-invocation rate, or a token-throughput curve. It moved ho…/runaway-bills#faq-q-2
- Why do AI/agent incidents keep showing up here specifically?Because an AI agent has no natural request-per-response ceiling the way a plain API call does. A runaway loop, a misconfigured retry, or no per-user or per-pro…/runaway-bills#faq-q-3
- How is CostMon priced?A flat monthly rate per plan: $0 for Free, $49 for Professional, $199 for Enterprise. We never charge a percentage of your cloud spend or of the savings we hel…/pricing#faq-q-0
- What counts as a connector?A connector is a link to one billing source, for example AWS Cost Explorer or your Anthropic account. Free includes 2 connectors; Professional and Enterprise i…/pricing#faq-q-1
- Which providers can I connect today?AWS, GCP, and Azure for cloud. Anthropic, OpenAI, Amazon Bedrock, Vertex AI / Gemini, and Helicone for AI and LLM spend. Snowflake and Databricks for the data…/pricing#faq-q-2
- Is my billing data safe?Connectors use read-only, cost-scoped credentials. CostMon reads your invoices and usage data, never your workloads, source code, or production systems. Sign-i…/pricing#faq-q-3
- How far back does history go?Free keeps 90 days of cost history. Professional and Enterprise keep full, unlimited history so you can track spend and trends across quarters and years./pricing#faq-q-4
- Can CostMon see my source code or data?No. Connectors are scoped to billing and usage APIs only. CostMon reads invoices and cost reports. It has no access to your source code, application data, or p…/security#security-faq-q-0
- What happens to my connector credentials?Credentials are stored server-side and never serialized back to your browser or into any API response. The API only reports whether a credential is set. Editin…/security#security-faq-q-1
- Can another customer see my spend?No. Every read and write is scoped to your organization. A request for another org's data returns a 404, so CostMon won't even confirm that org exists, let alo…/security#security-faq-q-2
- Do you store my password?No. Authentication runs entirely on AWS Cognito. CostMon never stores or sees a password. Every plan gives your team admin and viewer roles on the same identit…/security#security-faq-q-3
- What happens when I disconnect a provider?The connector and its credential are removed immediately. Historical cost rows it produced detach from the connector rather than being deleted or staying tied…/security#security-faq-q-4
- Is CostMon SOC 2 certified?Not yet. CostMon is pre-launch. SOC 2 Type II, configurable data-retention controls, and audit logs are on our roadmap. We will update this page the moment any…/security#security-faq-q-5
- Isn't a script that calls the Cost Explorer API basically a weekend project?The first call, yes. An API request against a single provider is quick. What ships to production is rarely just that, though. It also needs a schema to normali…/build-vs-buy#faq-q-0
- What if we only need one provider and our spend is simple?Then DIY is a real contender. This estimator's own honest section says so. One stable connector, with no plan to add more providers, is a clear case where buil…/build-vs-buy#faq-q-1
- Doesn't an estimator built by a vendor obviously favor buying?It's a fair thing to be skeptical of. That's why every number here is a constant you can see and change: the hourly rate, the build days, and the maintenance h…/build-vs-buy#faq-q-2
- What does this estimate leave out that would make DIY cost more?Three things. First, the compute, storage, and hosting to actually run the pipeline (a database, a scheduler, a dashboard host). Second, the opportunity cost o…/build-vs-buy#faq-q-3
- How is this different from the flat vs. percentage pricing page?That page compares two ways a vendor might charge you once you've already decided to buy a tool. This page answers an earlier question: whether to build your o…/build-vs-buy#faq-q-4
- Why does an AI agent cost so much more than a single API call?A single call bills once, for one prompt and one reply. An agent run is many calls chained together. Unlike a plain conversation, most of those calls re-send t…/ai-agent-costs#faq-q-0
- How do I cap agent costs?The most direct lever is a per-run ceiling: a maximum step count and a maximum total token budget. The run isn't allowed to exceed either one before it has to…/ai-agent-costs#faq-q-1
- What's the actual math behind the scenario table above?At step k, the input is the base prompt plus k times the per-step growth. That growth is everything re-accumulated so far. The output is a roughly constant amo…/ai-agent-costs#faq-q-2
- How do I track what my agents spend?The figures on this page are a planning model built from published list prices, not a bill. CostMon connects directly to the cost APIs of Anthropic, OpenAI and…/ai-agent-costs#faq-q-3
- What is run-rate?Run-rate is current spend annualized as if the recent pace continued unchanged: for this calculator, current monthly spend times 12. It's a quick, forward-look…/cost-forecasting#faq-q-0
- How do I forecast my cloud and AI costs?Enter your current combined cloud, AI/LLM, and SaaS spend and an expected month-over-month growth rate. The calculator compounds that growth across a 12-month…/cost-forecasting#faq-q-1
- Is compounding month-over-month growth realistic for AI spend?It is more realistic than a flat monthly total, at least in direction. Usage-based AI/LLM spend tends to grow as a feature gets adopted or usage scales. It doe…/cost-forecasting#faq-q-2
- How is this different from the savings or cost-per-customer calculators?The savings calculator estimates reclaimable waste in your current bill, and the cost-per-customer calculator blends this month's spend against your customer b…/cost-forecasting#faq-q-3
- How do I track this automatically instead of re-entering numbers every month?This calculator is a manual, one-time projection. CostMon connects to your actual cloud, AI, and SaaS spend. It computes the real run-rate and trend continuous…/cost-forecasting#faq-q-4
- What is the credit cliff?The credit cliff is the month when promotional credits end or expire and the full bill arrives at once. The usage behind the bill was there all along. The cred…/cloud-credits#faq-q-0
- Do AWS credits cover everything on my bill?No. The AWS Promotional Credit Terms exclude AWS Marketplace, some AWS Support, and Route 53 domain registration. They also exclude "any upfront fee for any Se…/cloud-credits#faq-q-1
- How do I see my usage before credits on AWS?Use the Charge type filter in AWS Cost Explorer or AWS Budgets. "Credit" is one value. Exclude it to see gross usage./cloud-credits#faq-q-2
- Which comes first for me, the end of the balance or the expiry date?Enter your balance, the months until expiry, your usage before credits, and your growth. The calculator shows the run-out month, says which event comes first,…/cloud-credits#faq-q-3
- Do other providers run credit programs too?Google Cloud, Azure, Anthropic, and OpenAI are examples of providers that may offer credits. Terms differ by program and by contract. Read your own terms. This…/cloud-credits#faq-q-4
- How can CostMon help before the credits end?CostMon tracks your daily spend and alerts you on a jump. It does not promise to show the usage behind a credit. Use the Charge type filter in AWS to see gross…/cloud-credits#faq-q-5
- Should we build a cost dashboard in-house instead of buying a tool?It's a real option if you have engineering time to spare and only one or two billing sources to reconcile. The build cost usually gets underestimated. The firs…/how-to-choose-a-cost-monitoring-tool#faq-q-0
- Is a spreadsheet enough to track cloud and AI spend?For a single provider at low spend, a spreadsheet of monthly totals can be enough. It stops working once you have more than one billing source. Someone then ha…/how-to-choose-a-cost-monitoring-tool#faq-q-1
- Does a percentage-priced cost tool end up costing more as we grow?Yes, and structurally so. The fee is a percentage of a number that grows as your infrastructure grows. That number is your managed spend, or the savings identi…/how-to-choose-a-cost-monitoring-tool#faq-q-2
- How many providers should one cost monitoring tool cover?As many as you run meaningful spend on. Most teams run cloud (AWS, GCP, or Azure), at least one AI/LLM provider, and a handful of SaaS tools. If your tool only…/how-to-choose-a-cost-monitoring-tool#faq-q-3
- Do we need SSO and role-based access for a cost tool?If spend visibility is going to include more than a couple of people, yes. Cost and usage data shows a lot about how a company runs its product. It needs the s…/how-to-choose-a-cost-monitoring-tool#faq-q-4
- Which tools should we evaluate alongside CostMon?It depends on what you already run. The section "Where other tools fit" above groups every tool we compare by the kind of problem it solves. Each group links t…/how-to-choose-a-cost-monitoring-tool#faq-q-5
- What's a good cost per customer?There's no universal number. It depends entirely on your pricing and margin structure. The more useful question is gross margin percentage, since it's comparab…/cost-per-customer#faq-q-0
- Should I include AI/LLM spend in this calculation?Yes, if that spend serves your product's customers, and for most AI-powered features, it does. Treating AI cost as a separate line item that finance reviews on…/cost-per-customer#faq-q-1
- What if my cost per customer looks fine but margin is thin?That combination usually means MRR per customer is low relative to spend per customer: a pricing or packaging question more than a spend question. Cost per cus…/cost-per-customer#faq-q-2
- How is this different from the LLM cost calculator or the savings calculator?The savings calculator estimates waste as a share of your total stack spend. The LLM cost calculator estimates a single AI feature's cost from token volume. Th…/cost-per-customer#faq-q-3
- How do I track this automatically instead of re-entering numbers every month?This calculator is a manual, one-time estimate. CostMon's Unit Economics feature reads your actual cloud, AI, and SaaS spend. You define a business metric (act…/cost-per-customer#faq-q-4
- Where do I see AWS usage types on my bill?In the AWS Cost Explorer console, set "Group by" to Usage type instead of just Service. That's the field this page decodes. For row-by-row detail that Cost Exp…/aws-bill-decoder#faq-q-0
- Why does the same usage type show up with a different prefix in different accounts?Usage types are region-scoped. A code with no region prefix generally means us-east-1 for the classic data-transfer family. Every other region gets its short c…/aws-bill-decoder#faq-q-1
- What's the difference between a usage type and an operation?Usage type is the billing meter: the dimension AWS charges you on, like NatGateway-Bytes or BoxUsage:m5.large. Operation (lineItem/Operation in a CUR) is the s…/aws-bill-decoder#faq-q-2
- Why is my NAT gateway bill so much higher than I expected?A NAT gateway bills in three separate ways. First, a flat hourly uptime charge. Second, a per-GB data-processing charge for every byte that passes through it.…/aws-bill-decoder#faq-q-3
- Is data transfer into AWS free?Generally yes. Data transfer from the public internet into an AWS region costs $0. The inbound leg of a transfer between two AWS regions also costs $0. Only th…/aws-bill-decoder#faq-q-4
- Why did I get billed for a public IP address I'm not even using?AWS began charging for every public IPv4 address on February 1, 2024, not just "extra" or idle Elastic IPs, as before. An unattached Elastic IP left over from…/aws-bill-decoder#faq-q-5
- Does CostMon replace reading my AWS bill line by line?That's the point of it. CostMon pulls these same usage-type-level records automatically. It normalizes them next to your other cloud and AI spend. It alerts yo…/aws-bill-decoder#faq-q-6
- Is data transfer into AWS free?Yes. Data transfer from the public internet into an AWS region is always $0.00/GB, and the inbound leg of an inter-region transfer is also $0.00. Only the send…/aws-data-transfer-costs#faq-q-0
- Why does moving 1 GB between two Availability Zones cost $0.02 instead of $0.01?AWS bills DataTransfer-Regional-Bytes on both legs of the move: the sender's outbound byte at $0.010/GB and the receiver's inbound byte at another $0.010/GB. T…/aws-data-transfer-costs#faq-q-1
- Is cross-AZ traffic free if I see a $0.00 xAZ-In/Out-Bytes line on my bill?No. That $0.00 rate is a narrow waiver for VPC interface endpoint (PrivateLink), Transit Gateway, and Client VPN traffic only, in effect since 2022-04-01. Gene…/aws-data-transfer-costs#faq-q-2
- Why is my NAT gateway bill so much higher than I expected?A NAT gateway bills three ways at once. It charges a flat $0.045/hour uptime fee per gateway. It charges $0.045/GB for every byte that passes through it. It al…/aws-data-transfer-costs#faq-q-3
- Does the 100 GB free tier reset per AWS service?No. It's a single 100 GB/mo allowance aggregated across every AWS service and every region in the account, not 100 GB per service. That's why it masks a growin…/aws-data-transfer-costs#faq-q-4
- Why can't I find data transfer charges in AWS Cost Explorer?Cost Explorer's default "Service" grouping rolls data transfer into a catch-all "EC2-Other" bucket with several other charges. It never appears as its own labe…/aws-data-transfer-costs#faq-q-5
- Does CostMon replace calculating this by hand every month?That's the point of it. CostMon pulls these same usage-type-level records automatically. It normalizes them alongside the rest of your cloud and AI spend. It a…/aws-data-transfer-costs#faq-q-6
- Why does my S3 bill keep climbing when I'm not uploading much new data?Versioning, incomplete multipart uploads, and snapshot-like accumulation are the usual causes. None of them show up in a normal object count, and none of them…/aws-storage-costs#faq-q-0
- Is Intelligent-Tiering just a strictly better version of a manual lifecycle rule?For unpredictable access patterns, usually yes. Its three automatic tiers are priced identically to the manual classes they mirror, so the storage cost lands i…/aws-storage-costs#faq-q-1
- What's the actual difference between Standard-IA and Intelligent-Tiering's Infrequent Access tier?The storage rate is identical. The difference is who decides when data moves. With Standard-IA, you write the lifecycle rule. You take the early-delete risk if…/aws-storage-costs#faq-q-2
- Why did deleting an old snapshot barely change my EBS bill?EBS snapshots are incremental. Each one only stores the blocks that changed since the last snapshot in its chain, even though the console shows every snapshot'…/aws-storage-costs#faq-q-3
- Is migrating from gp2 to gp3 risk-free?It's an online, no-downtime operation for an existing volume. At equal provisioned size, it's a 20% saving. It also has a higher default performance baseline (…/aws-storage-costs#faq-q-4
- What happens if I need an object back before its Glacier minimum storage duration is up?You can restore or delete it any time. Nothing blocks the action, but you pay the early-delete penalty. That is the storage rate for the days left in the class…/aws-storage-costs#faq-q-5
- Does this calculator account for S3 request charges (PUT/GET/LIST)?No, deliberately. Every other figure on this page came from a first-party pull of AWS's Price List Bulk API, and per-request pricing wasn't part of that pull.…/aws-storage-costs#faq-q-6
- Does CostMon fix any of this automatically?No. CostMon doesn't rewrite a lifecycle policy, delete a volume, or migrate a gp2 fleet to gp3 for you. CostMon pulls your real AWS cost and usage data on a sc…/aws-storage-costs#faq-q-7
- Does CostMon show which GPU instance is running?No. CostMon reads AWS Cost Explorer data grouped by linked account and service. It shows the daily "Amazon Elastic Compute Cloud - Compute" line and the "Amazo…/gpu-costs#faq-q-0
- Why does the default rule miss a small GPU node?The default rule watches the whole EC2 compute line. A p5.48xlarge adds a large step to that line, so the rule fires. A g5.xlarge adds about $24 a day. A large…/gpu-costs#faq-q-1
- Why is there no rate for a Capacity Block?You pay for a Capacity Block up front when you book it. The AWS Price List shows $0.00/hour on-demand for it. A per-hour rate would mislead you, so this page d…/gpu-costs#faq-q-2
- Why does p5en.48xlarge show less GPU memory than p5.48xlarge?AWS lists the p5en.48xlarge GPU memory per GPU: 141 GB. The other rows show the total for all GPUs. Eight H200 GPUs hold 1,128 GB in total./gpu-costs#faq-q-3
- What happens if the node runs for more than 14 days?The rule compares each day with the 14 days before it. After 14 days, the node is part of that baseline. So the alert fires only in the first days after the no…/gpu-costs#faq-q-4
- Can I set a lower alert threshold?Yes. You can create a rule scoped to "Amazon Elastic Compute Cloud - Compute". That rule can use a lower minimum percent over baseline. The calculator shows th…/gpu-costs#faq-q-5
- Why did our Datadog bill jump when we didn't add any new hosts?Host count is rarely the cause of a sudden jump. Custom-metric cardinality and log volume move far more easily. Check for a recently added tag (customer_id, re…/observability-costs#faq-q-0
- Is a brief autoscale spike really going to blow up our Datadog bill?Not necessarily. Datadog bills the high-water mark of the LOWER 99% of the month's hourly host counts. It excludes the top outlier hours before it takes the ma…/observability-costs#faq-q-1
- What counts as a custom metric?Any metric not submitted through one of Datadog's 1,000+ built-in integrations, per Datadog's documentation. This includes metrics sent via DogStatsD, a custom…/observability-costs#faq-q-2
- Why does this page only price one Datadog custom-metric overage rate?It is the only one Datadog publishes. For standard (non-"Metrics without Limits") accounts, the indexed-metric overage rate is "specified in your current contr…/observability-costs#faq-q-3
- What's the real difference between CloudWatch log ingestion and log storage?Ingestion is a one-time charge when a log event arrives. Storage is a recurring monthly charge for every byte still in the log group. A log group with no reten…/observability-costs#faq-q-4
- Does moving logs to an Infrequent Access or Flex tier mean losing them?No, but you lose some capability. CloudWatch's Infrequent Access class still supports Logs Insights queries. It drops real-time features such as metric filters…/observability-costs#faq-q-5
- Does CostMon fix any of this automatically?No. CostMon does not change a metric's tags, set a CloudWatch retention policy, or adjust an APM sampling rate for you. It pulls your Datadog and AWS cost and…/observability-costs#faq-q-6
- Why did our Snowflake bill jump without adding any warehouses?Warehouse count is rarely the culprit. Look at resumes/day and the auto-suspend timeout on your busiest warehouses first. A workload that polls more often can…/data-warehouse-costs#faq-q-0
- Is a shorter auto-suspend timeout always better?No. It cuts idle-before-suspend waste but increases how often the 60-second resume minimum gets paid. For a warehouse queried in frequent, short bursts, a shor…/data-warehouse-costs#faq-q-1
- Why doesn't this page show a dollar price for Snowflake credits?Snowflake does not publish one. Its list price varies by edition, cloud provider and region. snowflake.com/pricing defers to a Service Consumption Table PDF an…/data-warehouse-costs#faq-q-2
- What's the real difference between BigQuery on-demand and Editions pricing?On-demand bills by bytes scanned, with no fixed monthly floor: light usage costs little, heavy usage scales linearly. Editions bills by slot-time instead, for…/data-warehouse-costs#faq-q-3
- Why does BigQuery storage sometimes cost less after 90 days?A table that goes 90 consecutive days without a modification automatically bills at the long-term storage rate, roughly a 30% discount versus active storage. A…/data-warehouse-costs#faq-q-4
- Does CostMon fix any of this automatically?No. CostMon doesn't resize a warehouse, flip a storage billing mode, or buy an Editions reservation for you. CostMon pulls your real Snowflake and BigQuery cos…/data-warehouse-costs#faq-q-5
- Should I choose a Savings Plan or a Reserved Instance?It depends on how confident you are in the shape of your usage, more than its size. A Reserved Instance earns a deeper discount but locks a specific instance f…/commitment-discounts#faq-q-0
- Should I commit for 1 year or 3 years?A 3-year term buys a deeper discount. But it locks in the bet for three times as long. That gives an architecture change or a migration three times as long to…/commitment-discounts#faq-q-1
- What happens if my usage drops after I commit?You keep paying the full commitment bill regardless. That's what "committed" means. The capacity you don't use becomes stranded spend: money paid for capacity…/commitment-discounts#faq-q-2
- Can I cancel or sell a commitment I no longer need?Not directly in most cases. You can list AWS Standard and Convertible Reserved Instances on the AWS Reserved Instance Marketplace for resale. You can also exch…/commitment-discounts#faq-q-3
- Is all-upfront or no-upfront the better payment option?All-upfront earns the deepest discount for the same term, because you prepay instead of paying monthly installments. No-upfront earns the smallest discount but…/commitment-discounts#faq-q-4
- Does a commitment cover usage in a different region or instance family than what I committed to?Usually not, or only partially. Instance-family-scoped commitments (Standard RIs, EC2 Instance Savings Plans, GCP resource-based CUDs) generally don't cross re…/commitment-discounts#faq-q-5
- How much coverage should I target?Size the commitment to your steady-state floor (the level usage reliably stays above), not to your average and never to your peak. A common approach is ladderi…/commitment-discounts#faq-q-6
- Are there commitment discounts on Claude, GPT, or other model API usage?Not in the cloud-compute sense of a multi-year reservation. The closest analogues are batch processing (Anthropic's Message Batches API, OpenAI's Batch API) an…/commitment-discounts#faq-q-7
- Does CostMon buy or manage commitments for me?No. That decision and the negotiation behind it stay yours. CostMon shows the real, normalized baseline to size a commitment against. It also shows the coverag…/commitment-discounts#faq-q-8
- Is CostMon's anomaly detection real-time?No, it's day-grain, not real-time. Cost data syncs hourly and anomaly rules evaluate on an hourly sweep. The actual signal is a complete day of normalized cost…/cost-alerts#faq-q-0
- What's the default anomaly rule, and do I have to set one up myself?Every new org gets an org-wide rule seeded automatically the first time someone logs in. It has a 14-day lookback, a 3.0 z-score threshold, a minimum 50% over…/cost-alerts#faq-q-1
- How does Slack alerting work?Through an incoming webhook URL you paste in, not a Slack app or bot that needs authorizing or added permissions. A new alert posts to that webhook the same wa…/cost-alerts#faq-q-2
- What's the difference between a budget and an anomaly rule?An anomaly rule asks whether today's spend is statistically unusual against its own recent baseline; it can fire in the first week of a month. A budget asks wh…/cost-alerts#faq-q-3
- Why watch tokens per request instead of just dollars?An agentic workload can fan out into far more model calls, or far more tokens per call, than intended. This often shows in tokens-per-request before it moves t…/cost-alerts#faq-q-4
- Will re-firing the same anomaly spam me with duplicate alerts?No. Every alert has a deterministic dedupe key. CostMon derives it from the condition that triggered the alert. The same ongoing anomaly re-evaluates on the ne…/cost-alerts#faq-q-5
- What happens if my org doesn't have enough cost history yet?The detector requires at least 2 days of baseline history before it will ever fire. With fewer than that, there isn't enough of a "normal" to compare against,…/cost-alerts#faq-q-6
- Does this calculator need my actual Kubernetes metrics to work?No. Every input is a percentage or a dollar figure you already roughly know. These are your bill, your node count, and approximate request and usage percentage…/kubernetes-costs#faq-q-0
- What's the difference between workload idle and cluster idle?Workload idle is capacity your pods requested but don't use: a request-sizing problem, fixed by right-sizing requests down to observed usage. Cluster idle is c…/kubernetes-costs#faq-q-1
- Why does the binding resource matter?Kubernetes schedules against whichever resource (CPU or memory) a pod's request pressures harder, node by node. The resource with the higher request percentage…/kubernetes-costs#faq-q-2
- Why does the idle-tax multiplier sometimes show a dash instead of a number?The idle-tax multiplier is monthlyBill ÷ usedCost. It is undefined, not infinite, when usedCost is zero (0% usage). The calculator shows a dash for that one ed…/kubernetes-costs#faq-q-3
- Which idle-cost policy should we use?There's no policy without a trade-off. Proportional scales idle cost with team size but hits the biggest team hardest in absolute dollars. Even split is fairer…/kubernetes-costs#faq-q-4
- Is the right-sizing scenario a guarantee of what we will save?No, it's a scenario, built from the usage percentages you enter plus a headroom margin and a target packing density, both of which you control. Real savings de…/kubernetes-costs#faq-q-5
- Does CostMon replace Kubecost or OpenCost?No. CostMon reads the provider bill: your EKS/GKE/AKS control-plane fee and node-group spend, normalized alongside every other provider you connect, with drift…/kubernetes-costs#faq-q-6
- Where do the utilization statistics on this page come from?Datadog's State of Cloud Costs report is independent, published research. It found that 83% of container cost is idle: 54% cluster idle and 29% workload idle.…/kubernetes-costs#faq-q-7
- Is anything I paste into this tool uploaded anywhere?No. CostMon's marketing site is a static build with no backend to upload to. Whatever you paste in is parsed and analyzed entirely in your browser's JavaScript…/kubernetes-waste-analyzer#faq-q-0
- Do I have to bring all three optional files?No, only kubectl get pods -A -o json is required. Add kubectl top or the metrics API output to unlock a dollar figure and the usage-based findings. Add a node…/kubernetes-waste-analyzer#faq-q-1
- Why is there no dollar figure when I only paste in the pod file?Because a request is a reservation, not a measurement. Without usage data, this tool cannot know how much of that reservation is used. It shows the request bre…/kubernetes-waste-analyzer#faq-q-2
- Is kubectl top's number the same as a week's average usage?No, and this matters: both kubectl top and the metrics API report an instantaneous sample, not a historical average. Sizing a request down to one low sample ri…/kubernetes-waste-analyzer#faq-q-3
- If this tool says a workload is oversized, does trimming its request actually save money?Only if the node count this cluster runs actually falls afterward, or the autoscaler scales in. A smaller request that leaves every node just as full changes n…/kubernetes-waste-analyzer#faq-q-4
- How does the workload roll-up figure out a Deployment or CronJob's name?It is a heuristic, not something read from the API. If a ReplicaSet name ends in what looks like a pod-template-hash, the tool strips that suffix to recover th…/kubernetes-waste-analyzer#faq-q-5
- What's the 1 vCPU ≈ 4 GiB assumption for, and can I change it?CPU and memory are different units, so dollarizing a mixed CPU/memory gap needs some equivalence between them. This tool uses the ratio behind general-purpose…/kubernetes-waste-analyzer#faq-q-6
- Does this replace kubernetes-costs, CostMon's other Kubernetes calculator?No, they answer different questions. Kubernetes Costs is a typed slider model: move a utilization percentage and see what a cluster shaped like that would spli…/kubernetes-waste-analyzer#faq-q-7
- Does this tool run anything inside my cluster?No. It only reads text you exported with kubectl and pasted in yourself. It has no cluster access, no service account and no in-cluster component. Tools like K…/kubernetes-waste-analyzer#faq-q-8
- Does this scorer need my real tagging data to work?No. Every input is a percentage or dollar figure you already have a rough sense of. Four inputs matter. One is your total bill. One is roughly how much of it c…/cost-allocation-tagging#faq-q-0
- Why does the calculator ask for coverage "by cost" instead of "by resource count"?Because the two numbers routinely disagree, and cost is the one that matters. A few untagged resources can carry a large share of the dollar total. Examples ar…/cost-allocation-tagging#faq-q-1
- What's the difference between coverage and accuracy?Coverage asks whether the required tags are present at all. Accuracy asks whether the values on those tags are correct: the approved spelling and casing, not a…/cost-allocation-tagging#faq-q-2
- Why is AI/LLM API spend excluded from "taggable share"?Because there's no resource to tag. You can attach a key-value pair to a cloud instance or bucket. An Anthropic or OpenAI API call is a metered event with no r…/cost-allocation-tagging#faq-q-3
- Which shared-cost policy should we actually use?There's no policy without a trade-off. Proportional scales the unattributable pool with team size but hits the biggest team hardest in absolute dollars. Even s…/cost-allocation-tagging#faq-q-4
- What does "gap to chargeback-ready" actually mean?It's the dollar amount of spend that still needs to gain correct tags, present and accurate, before the bill crosses a 95 confidence score. That's the threshol…/cost-allocation-tagging#faq-q-5
- Is 95 an arbitrary threshold for chargeback-ready?It's a deliberately high bar, chosen because chargeback only works if the number can survive a dispute. Below that level, a team being billed has a legitimate…/cost-allocation-tagging#faq-q-6
- Does CostMon apply, activate, or enforce any of these tags for us?No. CostMon reads the tags that already exist in a provider's billing export and normalizes them across connectors. It adds no tags to resources. It does not a…/cost-allocation-tagging#faq-q-7
- Can I use the Terraform, tag policy, and IAM snippets as-is?Treat them as a starting shape, not a drop-in final policy. Swap in your own key names if they differ from the five-key schema above. Confirm your account's ac…/cost-allocation-tagging#faq-q-8
- Are my exports uploaded anywhere?No. CostMon's marketing site is a static build with no backend to upload to. Both CSVs you drop in are parsed and analyzed entirely in your browser's JavaScrip…/tag-coverage-analyzer#faq-q-0
- Why does this tool need two files instead of one?Because there is no confirmed, documented label for "untagged" inside a single Cost Explorer export grouped by tag. "Show only untagged resources" is a real, n…/tag-coverage-analyzer#faq-q-1
- What counts as one dollar of coverage here?A dollar that shows up in the all-costs file and does not show up in the untagged-only file for that same service and period. This tool checks one tag key at a…/tag-coverage-analyzer#faq-q-2
- What does this coverage number not cover?Four things, and each one moves the number. It reads one tag key, so a different required tag scores differently. It reads one date window, so a longer or shor…/tag-coverage-analyzer#faq-q-3
- A service reports 100% untagged. Is that always a coverage gap?Not necessarily. Some charges have no resource to tag. They include data transfer, support plans, marketplace fees, tax, and Savings Plan or Reserved Instance…/tag-coverage-analyzer#faq-q-4
- What does CostMon mean by tag drift versus never tagged?A service near 100% untagged was most likely never tagged at all, or its required tag key was never activated. A service with a small, nonzero untagged share i…/tag-coverage-analyzer#faq-q-5
- Does this tool tell me how much I could save?No, and it isn't built to. Tag coverage is an ownership question, not a savings estimate. This tool reports which dollars nobody can currently be billed for, n…/tag-coverage-analyzer#faq-q-6
- Can coverage ever reach 100%?Not while charges with no resource to tag are part of the bill. This tool subtotals those charges separately and reports the highest coverage percentage they l…/tag-coverage-analyzer#faq-q-7
- Does CostMon apply, activate, or fix any of these tags for me?No. CostMon reads whatever tags already exist in a provider's billing export, the same as this tool does. It doesn't create a tag, activate a cost allocation t…/tag-coverage-analyzer#faq-q-8
- What if I only have one of the two files?Drop in whichever one you have. This tool shows its total and its service breakdown right away. It also names what the second file would add: the coverage shar…/tag-coverage-analyzer#faq-q-9
- Why don't the five levers' savings just add up to the sum of their percentages?Each lever only discounts whatever dollars the levers before it left standing. It does not discount the original bill. Batch at 50% off plus caching at up to 9…/reduce-ai-costs#faq-q-0
- What is the prompt-caching break-even, and why does it matter?It's the cache-hit rate below which caching costs MORE than never caching at all. It is about 21.7% at the 1.25x 5-minute write multiplier. It is about 52.6% a…/reduce-ai-costs#faq-q-1
- Which lever usually saves the most?Output discipline, more often than not. Output tokens are priced several times higher than input tokens on every model. A modest cut to response length often o…/reduce-ai-costs#faq-q-2
- Is model routing safe for every workload?No. Routing to a cheaper model only saves money if that model completes the task at an acceptable quality bar. A retry, an escalation back to the frontier mode…/reduce-ai-costs#faq-q-3
- Can I run interactive, user-facing traffic through a batch endpoint?No. Batch processing trades a real-time response for an asynchronous one, on a completion window that providers publish as up to 24 hours. It's a free discount…/reduce-ai-costs#faq-q-4
- Where do the model prices and workload numbers in this calculator come from?Model list prices come straight from the same table behind the LLM Cost Calculator (src/data/llmPricing.ts); this page doesn't fork or re-key them. The default…/reduce-ai-costs#faq-q-5
- Does CostMon implement caching, routing, or batching for me?No. Those are engineering changes your team makes to your own request code; CostMon doesn't write prompts, route calls, or migrate workloads to a batch endpoin…/reduce-ai-costs#faq-q-6
- Does the 50% batch discount really stack with prompt caching?Yes, for Anthropic's Message Batches API: its own documentation states the batch discount stacks with prompt caching's cache-read discount. That's why this pag…/reduce-ai-costs#faq-q-7
- Why is a negative saving shown in a brick/red color instead of just being hidden?It is a real number, not an error state. Below the cache break-even hit rate, caching costs more than not caching at all. An under-used commitment discount in…/reduce-ai-costs#faq-q-8
- Is my bill uploaded anywhere?No. CostMon's marketing site is a static build with no backend to upload to. The CSV you drop in is parsed and analyzed entirely in your browser's JavaScript.…/aws-bill-analyzer#faq-q-0
- What file formats does the Analyzer support?Its primary target is AWS Cost Explorer's own "Download CSV" export, and it auto-detects that shape first. It also handles a console table copied and pasted as…/aws-bill-analyzer#faq-q-1
- Why does the Analyzer's total differ from my actual invoice?Cost Explorer's numbers (what this tool reads) and your final invoice can differ. This tool does not model the causes: credits, refunds, tax, currency conversi…/aws-bill-analyzer#faq-q-2
- Does this account for credits, Reserved Instances, or tax?No. It analyzes the numbers in the column you exported, typically unblended or amortized cost. It does not model credits, RI/Savings Plan amortization schedule…/aws-bill-analyzer#faq-q-3
- Does it work for GCP, Azure, or Anthropic cost exports?Often, as a secondary best-effort case. The parser needs a date column, a cost column, and a name/label column in one of three shapes. Several other providers'…/aws-bill-analyzer#faq-q-4
- What about AI/LLM spend? Does this catch that?Only if it's billed through your AWS account, like Amazon Bedrock usage, which appears as ordinary line items the same as any other AWS service. A direct Anthr…/aws-bill-analyzer#faq-q-5
- What if nothing on my bill trips a rule?The Analyzer says so plainly and lists what it checked, instead of manufacturing a finding to look useful. A clean result is a real, useful result. It means no…/aws-bill-analyzer#faq-q-6
- What does CostMon do that this tool doesn't?This tool reads one CSV, once, when you remember to export it. CostMon connects to AWS, Anthropic, and your SaaS tools with read-only credentials and does this…/aws-bill-analyzer#faq-q-7
- Is my usage file uploaded anywhere?No. CostMon's marketing site is a static build with no backend to upload to. The CSV you drop in is parsed, classified, and analyzed entirely in your browser's…/ai-spend-analyzer#faq-q-0
- What file formats does the Analyzer support?It reads four kinds of file. One is an Anthropic Console cost or usage export. Another is an OpenAI Usage Dashboard Cost or Activity export. The third is an Ad…/ai-spend-analyzer#faq-q-1
- Why does the Analyzer's total differ from my actual invoice?This tool reads whatever numbers are in the file you exported. It doesn't separately model credits, committed spend, tax or currency conversion. If your export…/ai-spend-analyzer#faq-q-2
- Can this tool show me my cache hit rate?No. This is a limit of the export. A cost export grouped by model bills a cache read as that same model's tokens, with no separate line for it. If nothing in y…/ai-spend-analyzer#faq-q-3
- Does the frontier-model comparison mean I'd save money by switching?No, and this tool is built not to claim that. The comparison is list-price arithmetic from CostMon's own pricing table (data/llmPricing.ts). It shows the cheap…/ai-spend-analyzer#faq-q-4
- Does this cover Amazon Bedrock or Google Vertex-hosted models?Only if your export already shows them as line items, the way an Anthropic Console or OpenAI Usage Dashboard export would for direct API usage. A model billed…/ai-spend-analyzer#faq-q-5
- What if nothing on my file trips a rule?The Analyzer says so plainly and lists what it checked, instead of manufacturing a finding to look useful. A clean result is still a useful result: it means no…/ai-spend-analyzer#faq-q-6
- What does CostMon do that this tool doesn't?This tool reads one export, once, when you remember to download it. CostMon connects to Anthropic's Admin API with a read-only key. It does this same kind of a…/ai-spend-analyzer#faq-q-7
- How do you decide what goes in this log?Every entry has one primary source: the provider's own blog post, What's New announcement, changelog, or pricing page. We fetched and read every one ourselves.…/price-changes#faq-q-0
- Why include price cuts? Doesn't that undercut the point of a cost-monitoring company?A log that only shows increases is marketing with a straight face. Providers cut prices about as often as they raise them, and a cut is still something worth k…/price-changes#faq-q-1
- How is this different from a provider's own What's New feed?AWS, Azure, and the rest each publish their own firehose, and none of them cross-reference the others. This log lists only the entries that moved a bill, acros…/price-changes#faq-q-2
- Can I get notified when a new entry is added?Subscribe to the RSS feed at /price-changes.xml. Every new entry is a dated item the moment it ships, so you don't have to keep checking back./price-changes#faq-q-3
- A provider changed something you haven't logged yet. Can I tell you?Yes. Send the announcement and its source link through the contact page. If it's a real, provider-sourced price change, it gets added with the same sourcing st…/price-changes#faq-q-4
- How is this different from the price change log at /price-changes?The log at /price-changes lists each price change as its own dated entry, newest first, built to read and to subscribe to. This index takes the same catalogue…/price-change-index#faq-q-0
- Why only a few dozen entries? Is that really enough to draw conclusions from?The corpus is curated, not exhaustive: one primary, provider-owned source per entry, fetched and read before it ships. Every figure on this page states its own…/price-change-index#faq-q-1
- What does "no stated effective date" mean?It means the source CostMon read for that entry didn't print a separate date for when the change starts hitting bills. It does not mean the provider gave custo…/price-change-index#faq-q-2
- How was each change assigned a direction: cut, increase, new charge, restructure, or free tier?By what the source itself describes as changing. A rate that dropped is a cut. A previously-free thing that started billing is a new charge. A change to the pr…/price-change-index#faq-q-3
- Can I get the raw data?Yes. The full catalogue is published as JSON at /price-changes.json and as CSV at /price-changes.csv, both licensed CC BY 4.0 with attribution to CostMon. Ever…/price-change-index#faq-q-4
- A provider changed something you haven't catalogued yet. Can I tell you?Yes. Send the announcement and its source link through the contact page. If it's a real, provider-sourced price change, it gets added to the catalogue behind b…/price-change-index#faq-q-5
- Is FOCUS a real standard, or something one vendor invented?It's real and vendor-neutral. The FinOps Foundation, a Linux Foundation project, maintains it. A Steering Committee of practitioners and provider representativ…/focus-finops-standard#faq-q-0
- Do I need to understand FOCUS to use CostMon?No. CostMon normalizes spend into its own model regardless of whether a connected provider exports FOCUS. This page exists because several CostMon connectors r…/focus-finops-standard#faq-q-1
- Why does my dashboard show a different number than my invoice?You're probably comparing BilledCost against EffectiveCost. BilledCost matches the invoice: a one-time upfront charge lands in a single month. EffectiveCost sp…/focus-finops-standard#faq-q-2
- Which providers export FOCUS data today?AWS, Azure, and Google Cloud all export FOCUS 1.2. Databricks and Vercel are on 1.3. Oracle, Alibaba Cloud, Tencent Cloud, Huawei Cloud, and OVHcloud are still…/focus-finops-standard#faq-q-3
- Does FOCUS cover AI and LLM spend?Partially. Since 1.2, FOCUS can represent virtual-currency consumption (credits, tokens, DBUs) as part of the cost and usage dataset. It doesn't yet have a sta…/focus-finops-standard#faq-q-4
- What's the difference between ListCost and ContractedCost?ListCost is the public, undiscounted rate. ContractedCost applies a negotiated discount (a private pricing agreement or volume discount) but not a commitment d…/focus-finops-standard#faq-q-5
- Is FOCUS 1.4 backward-compatible with 1.3?Yes. The FinOps Foundation's announcement for 1.4 says it shipped with zero incompatible changes to any existing FOCUS 1.3 query. It also completed deprecation…/focus-finops-standard#faq-q-6
- Where can I read the actual specification?The FinOps Foundation publishes it at focus.finops.org, including the full v1.2 column-level specification this page's definitions are sourced from./focus-finops-standard#faq-q-7
- Is anything I paste into this tool uploaded anywhere?No. CostMon's marketing site is a static build with no backend to upload to. Whatever you drop in or paste is parsed and analyzed entirely in your browser's Ja…/focus-export-analyzer#faq-q-0
- Does this tool work on an AWS, Azure, or Google Cloud export equally well?Yes. Header matching ignores case and separators, so BilledCost, Billed Cost, and billed_cost all resolve to the same column. That's what lets an AWS Data Expo…/focus-export-analyzer#faq-q-1
- What happens if I paste in a file that isn't a FOCUS export?This tool looks for FOCUS column names in the header row. If it finds none, it says so and points at the export steps below. It does not guess at a report from…/focus-export-analyzer#faq-q-2
- Why is a finding missing from my results?Every finding needs specific columns to compute. If one of them is not in your file, the tool hides the finding and names the missing column. It does not estim…/focus-export-analyzer#faq-q-3
- What FOCUS version does this tool check against?The column definitions and allowed values come from the FOCUS specification's ratified releases through 1.4, the same sourcing data/focus.ts and /focus-finops-…/focus-export-analyzer#faq-q-4
- Does this tool estimate savings?No. It reports what's actually in the file you gave it: totals, a conformance check, and findings with a dollar figure computed from your own rows. It never in…/focus-export-analyzer#faq-q-5
- How is this different from /focus-finops-standard?That page explains what FOCUS is and what each column means. This tool reads your own FOCUS file against those same definitions and prints a report. One is the…/focus-export-analyzer#faq-q-6
- Which of these guardrails actually stops spend, not just alerts on it?Only the hard cap does. AWS Budgets can apply a Deny policy or a service control policy once a budget is exceeded. Anthropic caps spend at the organization lev…/cost-guardrails#faq-q-0
- Does the Guardrail Generator send my budget or email anywhere?No. Every field stays in your browser. The generator builds the Terraform, gcloud command, or curl request with JavaScript that runs on this page. It never sen…/cost-guardrails#faq-q-1
- Is Anthropic's Spend Limits API available to every Claude account?No. Anthropic states plainly that "the Spend Limits API is available to Claude Enterprise organizations only. It is not available to Claude Platform (Claude Co…/cost-guardrails#faq-q-2
- Why doesn't a Google Cloud budget stop spend the way an AWS budget action can?Google's own documentation says so directly: "Setting an alerts-only budget doesn't automatically cap Google Cloud or Google Maps Platform usage or spending".…/cost-guardrails#faq-q-3
- How is CostMon's anomaly detection different from what AWS or Azure already do?It isn't trying to be better at spotting one provider's own anomaly. AWS and Azure already run real statistical detection against their own data. CostMon's def…/cost-guardrails#faq-q-4
- Do I still need a threshold alert if I turn on anomaly detection?Usually yes. An anomaly alert catches a spend pattern that breaks from your own history. It can miss a slow, steady climb that never spikes hard enough to look…/cost-guardrails#faq-q-5
- What does CostMon's seeded default rule actually watch?An org-wide, spend-only rule with a 14-day lookback, a 3.0 z-score threshold, a minimum 50% over baseline, and a $5 noise floor. It fans out to every (provider…/cost-guardrails#faq-q-6
No results. Try a different word, or browse the resources hub instead.
Get the alert before the invoice.
Connect your first providers in minutes. CostMon flags a spike within days and lists what to cut. Pricing is flat. We take no percentage of your bill or your savings.