Skip to main content

Precomputed Metrics

Importing Precomputed Metrics​

You can import precomputed metrics in three ways.

image

Importing Precomputing Metrics from your Data Warehouse​

Statsig integrates natively with cloud data warehouses such as Snowflake, BigQuery, Redshift to ingest any of your existing metrics for computing experiment results. See Data Warehouse Ingestion to get started.

Logging Precomputed Metrics using HTTP API​

You can also ingest these metrics using Statsig's HTTP API. As shown below, the log_custom_metric API call requires an Unit Identifier (e.g. user_id, stable_id) and the corresponding ID type. The API call should also include either a metric_value or (if it's a ratio metric) provide a numerator and denominator of the metric.

curl \
  “https://events.statsigapi.net/v1/log_custom_metric” \
  --header “statsig-api-key: <YOUR-SDK-KEY>” \
  --header “Content-Type: application/json” \
  --request POST \
  --data “{“metrics": [{"user_id": "1237", "metric_name": "test_metric", "id_type": "user_id", "metric_value": 90}, {"user_id": "4568", "metric_name": "ratio", "id_type": "stable_id", "numerator": 3, "denominator": 15}]}”

Response: {"success":true}

Statsig does not automatically process these metrics until you mark them as ready, as it's possible you might land data out of order. Once you are finished loading data for a period, you mark the data as ready by hitting the mark_data_ready API:

curl --location --request POST ‘https://api.statsig.com/v1/mark_data_ready’ \
--header ‘statsig-api-key: {your statsig server secret}’ \
--header ‘Content-Type: application/json’ \
--data-raw ‘{
    “timestamp”: 1647975283,
    “type”: “metrics”
}

The timestamp provided should:

  • Be a unix timestamp
  • Represent the latest point in time for which all metrics before have been uploaded.

Any future calls to this API with an earlier timestamp will be invalid. Statsig doesn't guarantee correct behavior if you record the same metric with an earlier timestamp with another API call.

Statsig processes metrics as a full day in the PST timezone and wait until a full day is marked as ready before processing that day.

Uploading Data to Azure Blob​

You can also import data into Statsig by uploading your data into a Statsig-owned Azure Blob container. To learn more, hit up the Statsig team on the Statsig Slack channel.

Debugging Precomputed Metrics​

Statsig creates a metric detail page for all precomputed metrics that you import from your data warehouse or ingest using the HTTP API. These metric detail pages take a few hours to generate post-import or ingestion. The fastest way to start seeing and debugging your precomputed metrics is via the Metrics Logstream in the Metrics Catalog tab within Metrics.

Screen Shot 2022-07-13 at 4 22 46 PM

The Metrics Stream will surface all ingested, precomputed metrics in real-time as they are ingested, enabling you to check metric name, metric value, unit identifier, ID type, and ingestion date.

Tip

Customers can trip up on ensuring that their precomputed metrics have the right ID type. Pay extra attention to this column!

Statsig also offers the ability to include test metrics, tagged with isTest. You can toggle these on/ off while debugging in the Metrics Stream. The isTest flag is currently available for precomputed metrics ingested via Statsig's APIs. Support for this flag via our integrations with Snowflake, BigQuery, and Redshift is coming soon.

To mark a batch of metrics as test metrics the isTest parameter needs to be set to true as part of the data for the request as seen below.

curl \
  “https://events.statsigapi.net/v1/log_custom_metric” \
  --header “statsig-api-key: <YOUR-SDK-KEY>” \
  --header “Content-Type: application/json” \
  --request POST \
  --data “{"isTest": true, “metrics": [{"user_id": "1237", "metric_name": "test_metric", "id_type": "user_id", "metric_value": 90}, {"user_id": "4568", "metric_name": "ratio", "id_type": "stable_id", "numerator": 3, "denominator": 15}]}”

292825183_585502089655173_8192569048240569580_n

Finally, the Metrics Stream only appears if you're actively ingesting precomputed metrics. If you're not seeing it appear at the bottom of your Metrics Catalog, Statsig likely is not receiving your precomputed metrics due to a connection issue or an invalid schema.