Changelog

You can now generate personas from your own first-party data. Once you've connected a data source and Minerva has processed it, open the new Personas page and click Generate: Minerva combines your records with our enrichment properties to find distinctive, high-performing groups in your audience.

Each persona is statistically validated and a real pattern in your audience. Minerva profiles it against US demographic averages and shows how many more people like it exist across the full data lake, so you know who they are and how many more you could reach.

Personas are real segments: activate one directly, or click Find more like this to search for net-new customers similar to the persona, excluding anyone already in your audience.

You can now create API keys that belong only to you instead of your whole organization. Personal keys sit alongside organization keys in API settings and work exactly the same way in your requests, so nothing in your code changes.

Live now for everyone!

You can now use address data to match people through our Enrich API. Records that include an address match 20% more often than those without, so if you have address data sitting unused in your records, start including it in your requests to /v2/enrich and /v2/resolve. See the Enrich input guide for accepted formats. Still on a V1 endpoint? Migrate to V2 first to take advantage of address lookup.

Both changes are live now. If there's an input we don't support yet that would help you match more records, tell us!

Bulk Enrich now has a faster, clearer workflow. Upload a CSV or Excel workbook, choose the right sheet, preview your data, and review its field mappings. Automatic mapping is more accurate, with clearer warnings when information is missing or incomplete.

You can now match records using LinkedIn URLs, giving you another way to identify people when email or phone data is unavailable.

If match rates are lower than expected, the new match rate investigation agent explains likely causes and recommends what to fix next.

Activate as much of a segment as you actually need. You now set the size for each activation, anywhere from a 100-record minimum to your whole segment (up to 300,000 per activation).

Slices are sampled at random, so they remain representative of your segment. And only net-new records count toward billing.

Activations launched with support for Meta, HubSpot, Salesforce, Lob, and CSV export. We've now added six more destinations: Google Ads, LinkedIn Ads, TikTok Ads, X Ads, Klaviyo, and Customer.io. For these new destinations, connect once through Integrations and they'll be available on every segment.

Whatever you activate, whether it's a bulk enrichment from your own input records or a person search, we enrich the records first, so you're always working with our highest-quality data. On activation, we export just the right columns for your destination to boost platform match rates and help you reach high-value contacts.

More destinations are on the way. Missing one you need? Tell us and we'll bump it up the list.

We've rebuilt how API keys work on Minerva. Until now you had a single personal key, with no expiry, no usage visibility, and no way to separate keys per use case.

The new keys fix that: create as many as you need, set an optional expiry date, see when each was last used, and revoke at any time. The full secret is only visible once, at creation, so store it somewhere safe.

For now, new keys are owned by your organization, so any member can use them. User-level keys will return later.

Manage your keys in API settings. To start building, check out the API docs, or use the Minerva SDK if you're working in Python. Your legacy personal key still works, but will be deprecated in the upcoming months, so we recommend switching to the new keys now.

The Minerva SDK is one client for the whole Data API: identity resolution, person enrichment, LinkedIn lookup, email validation, country inference, and usage tallies.

Install it with pip install minerva-sdk, set your API key as MINERVA_API_KEY, and you're ready to go. The SDK is in early access: the interface is stable, and we're hardening the edges ahead of the public release. Feedback welcome.

Find the package on PyPI, or in the official SDK docs.