What drives a lead's score
The score is built from signals, not a single guess. Three different streams feed it, and it is worth knowing which of your actions actually register.
The three streams that feed a score
- What a buyer does on your branded site: viewing listings, running searches, favoriting, sharing, using the affordability calculator, and submitting a form (find-an-agent, showing request, open-house RSVP).
- How the lead arrived: hand-added, CSV imported, referred, or captured at an open house. Each carries its own starting weight.
- What you log: calls, texts, emails, meetings, and showings you record on the lead, plus outcomes like qualified, offer made, or closed. This is the heaviest stream in practice.
Logging your activity is not admin busywork. Agent-logged effort carries the largest weights in the model, so a lead you are genuinely working will not look hot until you record the work.
What does not move the score today
Opens and clicks on Cultivate email do feed the score, gently: an open is worth 1 and a click 3, both fading fast. A click also records which listing was clicked, so the timeline names the house. Only Cultivate sends are tracked this way, so opens and clicks on a saved-search alert or an ordinary one-off email change nothing. A lead who reads everything you send and never replies will drift up slightly, not enough on its own to change their band.
The same is true of inbound texts on their own. A reply registers when it lands as a logged activity marked reached, which is what credits the exchange. The raw inbound message by itself does not.
Is it one AI call?
No. The base score is a plain weighted sum of signals over a rolling window, which is why it can be explained line by line. REHL AI adds a small bounded nudge on top and can never swing the number.
Does browsing before someone signs up count?
Yes, once they create an account on your site and become a lead. See the pre-account browsing article for the conditions.
Need more help? Contact support.