Lead qualification on WhatsApp
The first reply happens in seconds, whether or not anyone is free.
A Meta lead decays in minutes. By the time a portal lead has been round-robined to an agent, called twice and left to ring out, the buyer has already spoken to three other brokers. PropTech AI takes the first conversation off the critical path entirely — it opens on WhatsApp the moment the lead lands, establishes budget, configuration, corridor and timeline, and hands you a scored lead with the reasoning attached.
- Channel
- WhatsApp — Meta Cloud API, verified business number
- Score
- 0–100 across five weighted dimensions
- Labels
- HOT · WARM · COLD · DEAD · FRAUD
- Response time
- Seconds from lead arrival
- Voice
- Your brand, or an independent advisory
- Data
- Your own workspace — never pooled
The conversation
Not a form in a chat bubble. A real exchange on WhatsApp Cloud API from your verified business number, in the buyer's language and register.
Answers immediately
Inbound from Meta ads, portals, your website or a channel partner referral is engaged within seconds, at any hour.
Asks like a broker, not a form
Budget, configuration, corridor, possession timeline and financing all come out of conversation rather than a questionnaire.
Replies grounded in your inventory
The drafter answers from your own project data. It may search public record for possession dates, RERA status and zoning — it is mechanically blocked from sourcing your price or availability anywhere else.
Knows what it does not know
Fee model, brokerage terms, developer relationships and any verdict on whether a price is fair get deferred to a human, by rule.
The score
Five dimensions, weighted to 100, with a written reason on each. Business rules sit above the model as deterministic gates, so the label is auditable rather than a vibe.
HOT, WARM, COLD, DEAD or FRAUD
One label, one number, one explanation — enough to decide whether this is worth a call right now.
Hard gates the model cannot talk its way past
HOT needs both a 70 total and genuine intent. An unaffordable budget caps the label regardless of how good the conversation was.
Borderline cases come to you
A lead sitting between two labels is flagged for human validation with the reason, instead of being silently resolved to the safer one.
Fraud is quarantine, not a low score
Scored separately, excluded from the 100, and requiring corroborating signals before it acts.
What happens after the score
Qualification is only useful if it changes what happens next.
Matched inventory, both directions
Units ranked for a lead and buyers ranked for a unit come from one shared scorer, so the two lists never disagree.
Follow-ups that fire themselves
A queue drains on a schedule, so the second and third touch happen without anyone remembering to make them.
Dead leads get one more chance
A single reactivation touch at 30 days. Nurture is nearly free and a false negative on a multi-crore sale is not.
Outcomes feed back
What actually happened on the call returns to the system, and scoring is checked against it rather than tuned to flatter itself.
The workspace
A dashboard for the person managing the pipeline, not a settings screen.
Priority calls first
The day opens on who to call, with the full conversation and score breakdown one tap away.
Pipeline as a board
Drag a lead between stages; the conversation, matches and follow-ups travel with it.
Inventory and demand
Your project catalogue with matched buyers per unit and a demand view across corridors.
Every reply auditable
Each AI message is stored with the exact context it was generated from, then re-checked offline so a defect gets found and fixed rather than repeated.
See it against your own pipeline
The demo is not a slide deck. Bring a week of real leads and we will show you what the scoring does with them.