How Ace's Predictive Lead Scores Work

The six serving model families

The serving suite defines six predictive model families in BigQuery ML, plus a full-book fallback conversion fit for contacts outside the richer scoring path. Each family answers one ranking or outcome question:

The features feeding these models are engineered from your webhook stream — engagement velocity, inbound/outbound direction, response latency, channel behavior, deal context, IDX website activity — with strict as-of discipline: a training example only sees data that existed before its cutoff. That single rule is what separates a real predictive model from one that quietly "predicts" the past.

The loop that keeps the numbers honest

Most "AI lead scores" in real estate are static formulas. Ace's are predictions that get graded. The full loop:

1. Predict. Eligible contacts receive outputs from conversion, response, churn, appointment, buyer, and seller models; the serving path records supported predictions.
2. Label — hourly. The system checks each recorded prediction against what actually happened: did the contact respond or book an appointment within 24 hours? Within 7 days? Or go cold? Closed-won outcomes feed a separate, longer-horizon track.
3. Recalibrate — scheduled. Labeled outcomes can re-fit the mapping from raw model output to observed outcome rates. That calibrated probability is an internal input; the published Win Score remains a within-account percentile rank.
4. Serve — entitlement-aware. Every connected eligible account is scheduled nightly for Ace Win Score and Ace Churn Risk, whether free or on a paid seat. Trove can schedule additional batched refreshes after meaningful new activity.
5. Evaluate and promote. The labeled corpus grows as new leads arrive and deals close. Models are evaluated on a held-out slice and retrained when due; a candidate must pass promotion guardrails before replacing the deployed model.

For the deeper technical story — the data warehouse, the leakage audits, and why we publish what's still dark — read the predictive scoring & BI flywheel deep-dive.

What the numbers mean in practice

conversionProb → Ace Win Score

A 0–100 within-account percentile rank of calibrated win likelihood. Use it as a primary sort key: 90 means the contact ranks above roughly 90% of the account's scored book. It does not mean a 90% close probability, and score magnitudes should not be compared as absolute probabilities.

churnProb → Ace Churn Risk

The probability an engaged contact goes cold, bucketed Low / Medium / High with conservative cutpoints — "High" is deliberately rare so it stays actionable rather than becoming alert noise.

expectedGci → Ace Expected GCI (est.)

Win probability × your account's real median commission per closed deal (from your own closed-deal history in the warehouse). Always an estimate, always labeled. It converts probability into dollars at stake, which is how team leads should rank follow-up.

responseProb and appointment propensity

These two mostly work behind the scenes — ordering the daily queue, timing reach-outs, and powering on-demand scoring when you ask Ace about a single contact. Appointment propensity answers the question agents actually care about mid-funnel: "who's likely to get on my calendar this week?" Appointments themselves are tracked as first-class outcomes — set, held, no-show — which is also what makes the model's own report card honest.

buyer and seller propensity → Ace Buyer Tier and Ace Seller Tier

These models are rankers. Ace publishes the relative tier (Very High / High / Moderate / Low / None), not the raw model output as a close, buy, or sell probability. That keeps the agent-facing label useful without claiming absolute certainty the training design does not support.

The scores drive the experience — not just the fields

The FUB custom fields are the exported half. Inside the product, the same consolidated model suite (Win Score, Churn Risk, Opportunity, Seller intent, Expected GCI (est.)) runs the day-to-day experience:

• The contact launchpad. Open a contact in the embedded assistant and it leads with the model's read — a headline ("at risk of going cold", "one of your strongest opportunities"), the key signals, and quick actions that match.
• What's Next suggestions. Ace's suggested next moves are drafted against the model signals, so a high-churn contact gets a save play and an active seller gets a listing conversation — not a generic follow-up.
• The embed Intelligence dashboard. Your book ranked by the same scores, with a "Save first" list: engaged contacts the model flags as most likely to go cold, strongest first.
• Team dashboards. Team Cockpit aggregates the models per agent — average Win Score of each agent's book, churn exposure, modeled GCI (est.) — next to production and response speed.

So if a number in a FUB Smart List and a suggestion inside Ace ever seem to agree suspiciously well — that's because they're the same number.

What the scores can't do (read this part)

Frequently asked questions

• What is a calibrated lead score?

  Calibration describes how an internal probability lines up with observed outcomes. Ace Win Score is the account-relative output agents see: 30 means roughly the 30th percentile within that account, not a 30% close probability.

• How often do models retrain?

  The scheduler evaluates model age and drift, retrains when due, and promotes only candidates that pass guardrails. Every connected eligible account is scheduled nightly for Ace Win Score and Ace Churn Risk, whether free or on a paid seat.

• What data is used — and is my data shared?

  Models score your contacts using your account's own CRM activity. Cross-tenant benchmarks (like You vs Market) are a separate, anonymized system that only publishes cohort statistics backed by 10 or more teams — see Revenue Guard & You vs Market.

• Why did my contact's Win Score change overnight?

  A scheduled sync, a debounced activity-triggered rescore, a changed account distribution, or a model/calibration update can change the rank. Check Ace Last Analyzed where available; admins should also verify pipeline freshness rather than assuming a recent webhook means a completed write.

• Where do the scores show up?

  In the FUB custom fields (field guide), the contact launchpad and What's Next suggestions inside the embedded assistant, the embed Intelligence dashboard, the ranked daily queue, the admin dashboards (including per-agent aggregates in Team Cockpit), and via the MCP connector tools your AI assistant can call.