Overview

Metric Name pricing bills custom metrics based on the number of unique metric names you submit and the volume of datapoints those names produce. It replaces the cardinality-based Custom Metrics billing model for organizations that opt in.

Note: This page applies if your contract uses Metric Name pricing SKUs. These SKUs are mutually exclusive with the Timeseries (cardinality) pricing SKUs. If your contract uses Timeseries pricing instead, see Custom Metrics billing.

Datadog distinguishes ingested datapoints from indexed datapoints. Ingested datapoints are every metric datapoint your services send. Indexed datapoints are the points that remain queryable. By default, every ingested datapoint is also indexed. Metrics without Limits™ optionally reduces your indexed volume by excluding tags you don’t need, so you’re billed only for the datapoints that remain queryable. Indexed volume is always less than or equal to ingested volume.

SKUs

Metric Name pricing introduces three SKUs:

These SKUs are mutually incompatible with the Timeseries (cardinality) pricing SKUs.

Pricing structure

Both metric names and indexed datapoints are priced with marginal, volume-based discounting across 15 tiers. Higher-volume tiers have lower per-unit rates. As your usage increases, the portion that crosses into each higher tier is billed at that tier’s rate. Previously billed usage is never repriced.

Per-unit rates vary by contract type:

Specific rates and tier boundaries are defined in your contract. Contact Sales or your Customer Success Manager for details.

Monthly and hourly usage

Metric Name pricing usage is tracked hourly and billed either hourly or monthly, depending on your contract. Each unique metric name is counted once per month, in the hour it first submits indexed datapoints. Every billed metric name includes 10 million indexed datapoints. Datapoints beyond that allowance accumulate in a single monthly overage bucket across all of your metrics.

At the end of the billing period, the total monthly bill is the sum of:

• The tiered metric names charge for the month’s unique metric names
• The tiered indexed datapoints charge for the month’s overage datapoints

Monthly counters reset for the next billing cycle.

Hourly billing produces the same total as monthly billing. Hourly processing affects when tier boundaries are crossed, not the final charge.

Example

Suppose, over a single month, your hourly usage looks like:

At the end of the month:

• The 556 metric names are billed across the marginal tiers they fall into.
• The 58 million overage datapoints are billed across the marginal tiers they fall into.

Ingestion behavior

Under Metric Name pricing, every metric datapoint your services send to Datadog counts toward ingestion, independent of Metrics without Limits™ configuration. By default, all ingested datapoints are indexed. Configuring Metrics without Limits™ reduces your indexed volume.

Your free ingestion allowance covers ingested datapoints up to five times your indexed volume. You’re charged for ingested datapoints only above that threshold in a given month.

Note: This is a change from the cardinality-based model, in which only metrics configured with Metrics without Limits™ contribute to ingested volume.

Distribution metrics

For Distribution metrics, a multiplier applies to both ingested and indexed datapoints, regardless of whether the metric is configured with Metrics without Limits™. The multiplier is five times by default (one each for the count, sum, min, max, and avg aggregations Datadog generates). When percentile aggregations (p50, p75, p90, p95, p99) are enabled, the multiplier is ten times.

• For unconfigured Distribution metrics, ingested and indexed volumes are equal after the multiplier is applied.
• For configured Distribution metrics, the multiplier applies to ingested volume, while indexed volume can be lower depending on tag configuration and indexing rules.

Example

A Distribution metric that submits 100 datapoints with default aggregations is billed as 500 datapoints (100 × 5). With percentile aggregations enabled, the same submissions are billed as 1,000 datapoints (100 × 10).

Historical Metric Ingestion

Historical Metric Ingestion usage is calculated based on ingestion time, not the metric’s original timestamp. Each HMI datapoint contributes to both ingested and indexed volume.

Committed adoption

You can commit to a usage volume independently for metric names and for datapoints. Each commitment is priced at the marginal pricing tier the committed volume falls into. The full committed volume is billed at that tier’s rate, and marginal discounting does not apply within the committed range. Usage above the committed amount follows the standard marginal schedule, starting at the tier boundary immediately above the committed volume.

For example, if you commit to 15,000 metric names that fall within Tier 6, all 15,000 are billed at the Tier 6 rate. If your actual usage reaches 75,000 metric names, the additional 60,000 follow the marginal schedule from Tier 6 through Tier 8.

Usage Attribution

For the initial phase of Metric Name pricing, Usage Attribution is supported for indexed datapoints. Metric names cost is attributed in proportion to each tag’s share of the total datapoints. A tag-defined group that contributes 20% of the datapoints is attributed 20% of the metric names cost.

Usage Attribution for ingestion is not supported in the initial phase.

Custom and standard metrics

Metric Name pricing uses the same definition of custom and standard metrics as cardinality pricing. For what is considered a custom metric, see Custom Metrics.

Troubleshooting

For technical questions, contact Datadog support.

For billing questions, contact your Customer Success Manager.

Further reading

Additional helpful documentation, links, and articles:

Custom Metrics billing (cardinality pricing)DOCUMENTATION Learn more about custom metricsDOCUMENTATION Metrics without Limits™DOCUMENTATION Best practices for custom metrics governanceGUIDE