Credits are the currency used to purchase different data points and actions in Unify. Most actions require 1-2 credits, but the value varies by provider.

We consider the security of customer data our top priority. We maintain SOC 2 Type II compliance, perform real-time threat monitoring, and regularly contract external security audits. Unify also supports SSO authentication. For more information, reach out to security@unifygtm.com.

Email enrichment averages 1 credit per record (in rare cases up to 10). Phone enrichment averages 4 credits per record (in rare cases up to 20). Unify also offers a range of signals from dozens of vendors. These signals vary significantly by source, from a 1 credit average for things like web intent or technographics to a significantly higher rate for more complex runs or niche sources.

Yes. On any paid plan, you can add credits at any time as a one-time top-up. If you have multiple seats, top-ups go into a shared pool available to everyone on your team. They don't auto-renew, and unused credits carry over for up to 12 months.

We evaluate every data vendor before making them available on our platform, assessing both quality and cost. Think of it less like a black-box data layer and more like a vetted wholesale marketplace: we source from best-in-class specialized providers, negotiate pricing on your behalf, and are transparent about exactly which vendors power your results. You can see the underlying providers in your workflow, so you're trusting names you already know.

Unify enforces a hard cap: when your seat credits and account pool are both exhausted, credit-consuming actions pause rather than charge. You'll see an in-product prompt to top up or upgrade, and you're back to full access the moment credits are added or your cycle resets. Any automation already in flight will finish its current step before stopping.

It depends on what you're asking Unify to do. A simple list-building search is fairly low. Enriching a list with phone and email or writing a personalized email sequence will be more moderate. Deep agentic workflows that layer in multiple signals, news alerts, and multi-step research can run quite high. Each prompt can vary significantly based on the sources and steps involved. The more there are, the more that will be used.