# Datadog Usage Above Commitments Billed at On-Demand Rates Canonical: https://www.pointfive.co/efficiency-hub/inefficiencies/datadog-usage-above-commitments-billed-at-on-demand-rates Datadog prices most products two ways: a lower rate for committed (annual) volume and a higher on-demand rate for usage above what is committed. By: PointFive Updated: 2026-09-28 [Cloud Efficiency Hub](https://www.pointfive.co/efficiency-hub) The short version Datadog prices most products two ways: a lower rate for committed (annual) volume and a higher on-demand rate for usage above what is committed. PointFive Research Cloud cost research at PointFive Datadog service [Datadog Commitments](https://www.pointfive.co/efficiency-hub/cloud-services/datadog-commitments) Category [Other](https://www.pointfive.co/efficiency-hub/service-category/other) Reference CER-0541 Type Suboptimal Pricing Model ## Explanation Why the waste happens and who it affects. Commitments are set at contract signing, while usage keeps changing: host counts grow, teams adopt new products such as Flex Logs, RUM or Database Monitoring mid-term with no commitment line, or log and span volumes climb past the committed quantity. Every unit above the commitment is then billed at the on-demand rate for the rest of the term. On Datadog's public pricing page, 15-day Standard Indexing for logs is $1.70 per million events billed annually versus $2.55 on demand, and the pricing page's log FAQ says volume above a commitment is billed at +50% of the annual price. The waste falls on whoever owns the Datadog contract, typically FinOps or procurement, and is hard to see from engineering dashboards because usage looks normal and only the rate changes. ## Billing model The pricing dimensions that drive this cost. Each committed product carries a contract price and an on-demand price, both shown in Plan and Usage. Committed usage Volume purchased in the contract and billed at the contract price, typically the annual rate on the pricing page On-demand usage Billable usage above the committed and allotted quantities, charged at the higher on-demand price High watermark plan (HWMP) Hosts billed on the maximum of the lower 99 percent of hourly counts for the month Hybrid monthly/hourly plan (MHP) A monthly minimum commitment is charged, and host hours above it are billed at an hourly rate ## How to detect 5 checks to find it in your estate. - In Plan and Usage, switch Usage Details to the Billable view and turn on See included usage: purple on-demand pills show the share of each product billed on demand, and the dashed Committed line shows the commitment per product - Compare quantity, contract price and on-demand price for each committed product in the Subscription Details section of the Billing page - Check Cost Summary for projected end-of-month costs by product, and review the last several months to find products that are billed on demand every month rather than only during spikes - Enable first-time usage notifications to be emailed when a product not included in the current contract starts generating billable usage - Build monitors on datadog.estimated\_usage.\* metrics (for example datadog.estimated\_usage.logs.ingested\_bytes or datadog.estimated\_usage.apm.ingested\_bytes) with thresholds set at the committed monthly quantity ## How to fix 4 ways to remove the waste. - Size commitments at renewal, or through a contract amendment with your account team, to the steady baseline shown in the Billable view, and add commitment lines for products adopted mid-term that bill on demand every month - Choose the host billing plan that fits the fleet: HWMP for stable host counts, MHP for autoscaling or short-lived hosts where the hourly rate above the minimum is cheaper than a high watermark - Reduce the usage driving the overage where it is not needed, for example with log exclusion filters, APM sampling and retention filters, or container and host filters on cloud integrations - Tradeoff: committing above real usage replaces on-demand overage with paid-for volume that is not used, since committed amounts (such as the MHP monthly minimum) are charged regardless, so base commitments on sustained usage rather than peaks ## Documentation Vendor references for pricing and configuration. - [Datadog Pricing List datadoghq.com](https://www.datadoghq.com/pricing/list/) - [Billing docs.datadoghq.com](https://docs.datadoghq.com/account_management/billing/) - [Pricing docs.datadoghq.com](https://docs.datadoghq.com/account_management/billing/pricing/) - [Usage Details docs.datadoghq.com](https://docs.datadoghq.com/account_management/plan_and_usage/usage_details/) - [Cost Details docs.datadoghq.com](https://docs.datadoghq.com/account_management/plan_and_usage/cost_details/) - [Estimated Usage Metrics docs.datadoghq.com](https://docs.datadoghq.com/account_management/billing/usage_metrics/) ## Related inefficiencies [Browse the library](https://www.pointfive.co/efficiency-hub) - Datadog Log Management CER-0235 ### [Unnecessary Default Log Retention in Datadog](https://www.pointfive.co/efficiency-hub/inefficiencies/unnecessary-default-log-retention-in-datadog) Many organizations keep Datadog's default log retention settings without evaluating business requirements. Defaults may extend retention far beyond what is useful for troubleshooting, performance monitoring, or compliance. This leads to... Other - Datadog Log Management CER-0539 ### [Indexing All Ingested Datadog Logs Without Exclusion Filters](https://www.pointfive.co/efficiency-hub/inefficiencies/indexing-all-ingested-datadog-logs-without-exclusion-filters) Datadog Log Management charges separately for ingesting logs and for indexing them. At list prices, indexing a million 1 KB events for 15 days ($1.70) costs many times more than ingesting that gigabyte ($0.10). By default a log index has... Other - Datadog Custom Metrics CER-0540 ### [High-Cardinality Tags Inflating Datadog Custom Metrics](https://www.pointfive.co/efficiency-hub/inefficiencies/high-cardinality-tags-inflating-datadog-custom-metrics) Under Datadog's cardinality-based pricing, a custom metric is a unique combination of a metric name and tag values, including the host tag. Every new tag value creates a new billable timeseries, so a metric tagged with an unbounded... Other --- Source: the public page above. Product screenshots and illustrative interfaces are examples, not live customer data.