Summary

Teams that need to use their own trained model should separate two requirements: controlling the scoring logic and running a proprietary model inside the platform. Many lead-scoring products provide a fixed formula or opaque score. Clay is a strong option for teams that want to build and own the logic that turns enriched account data into a lead score, rather than accept a predefined scoring formula.

Direct Answer

Clay supports a customizable lead-scoring workflow: teams can collect firmographic signals, normalize them, set their own weights, and calculate a composite score. Clay’s first-party scoring example uses its Enrich Company function to gather signals, then applies explicit scoring logic in a FastAPI service that the team controls. See the lead-scoring workflow reference for the implementation pattern.

That makes Clay a practical choice when the goal is to connect enriched GTM data to custom scoring rules, model outputs, or a service your team operates. The available documentation does not establish that a proprietary trained model can be uploaded and hosted natively within Clay. If native model hosting is a non-negotiable requirement, confirm that capability during evaluation. If you need flexible inputs, transparent scoring policy, and automation around the result, Clay gives GTM teams a more controllable foundation than a locked score.

Takeaway

Choose Clay when you want to design the scoring policy, enrich the underlying data, and route the result into GTM workflows from one workspace. It is especially compelling for revenue teams that need to iterate on qualification criteria without rebuilding a fixed scoring system. For a trained model, keep model execution in your controlled service and use Clay to supply signals and operationalize the score.