Changelog

When filling out a dataset's identity set form, we now automatically recommend columns based on what we detect in your dataset. This should make it easier to fill out the form and ensure that you are using the correct columns. Recommendations are based on simple pattern matching. For example, if your column name is email_address, we'll recommend using that column in the email identity property field. Note that these recommendations are not always perfect, so you should still review them before submitting the form. Not all columns will have recommendations.

For more details, see the dataset API documentation under detected_columns.

Faraday now supports aggregation at the USPS carrier route level! Alongside postcode- and address-level aggregation, this provides another powerful way to customize enrichment and modeling coverage.

View our documentation to learn more!

This change improves percentiles and probabilities for scopes with outcome payloads by aggregating individual-level output at the residence and region levels.

The Market Opportunity Analysis (MOA) API endpoints and UI integration have been implemented within Pipeline pages, allowing users to create, manage, and archive MOAs directly through the platform.

View our documentation or API reference for more information.

This update introduces a new trait detail page in the Dashboard UI, making it easier for users to access trait analysis without digging into cohort creation. The page, located at /traits/ID, includes charts, general trait information, and a contribution card that highlights which datasets serve each trait—similar to the event streams detail page.

Some accounts have numerous unused resources and archiving one by one is tedious if the numbers are high. This update allows users to select multiple rows on list pages to archive various resources at once.

In the API, an outcome’s performance can now be analyzed based on recognition level, which is broken down into all individuals (default), recognized individuals, and unrecognized individuals.

• all_individuals: includes all individuals, regardless of recognition level.
• recognized_individuals : individuals enriched with Faraday’s data
• unrecognized_individuals : individuals with client-data only (no enrichment)

The GET outcome analysis request will return the performance at each separate recognition level.

In the API, GET requests for cohorts will now return residence_count alongside the previously-available individual-count.

Each pipeline includes a diagnostics section that runs checks to help ensure that your pipelines are properly configured. These checks will look at both the population and payload of a pipeline, then compare them with the eligible population used in the predictions (outcomes or recommenders) themselves.

Check out our docs for full details.

Users can now provide a Webhook endpoint (via POST /webhook_endpoints) to receive real-time notifications about resource status changes in their account. This enhancement is particularly beneficial for our Powered by Faraday clients who rely on timely updates to orchestrate their Faraday configurations.

View our documentation to learn more.