Real estate has an unusual failure profile. Lead generation is typically well funded and reasonably effective: portals, ads, referrals, hoardings all produce enquiries. Site visits happen. And then a substantial share of visitors who were genuinely interested simply stop being contacted, and buy somewhere else.
The ninety days after a site visit is where bookings are won and lost, and it is the least systematised part of most brokerages' operations.
Why the post-visit window is so leaky
Four structural reasons, none of which is about effort.
The cycle is long and the memory is short
A property decision takes weeks or months. During that time the prospect talks to family, arranges finance, compares three other projects and goes quiet for a fortnight. A rep carrying thirty such prospects at different stages cannot hold thirty timelines in their head, and the ones who go quiet are precisely the ones who drop off the mental list.
The context lives with the rep, not the record
The site visit generates the most valuable information in the entire cycle: which unit they liked, what the objection was, who the decision-maker actually is, what their finance situation looks like. In most operations this is in the rep's head and a two-word note. When that rep is unavailable. Or leaves. The context goes with them and the next conversation starts from zero.
The work happens away from a desk
Site visits are on site. If updating the CRM requires returning to an office, it happens hours later, in summary form, or not at all. Every real estate CRM project that fails, fails here.
Qualification data is industry-specific
Budget, preferred locality, configuration, possession timeline, finance status. A CRM without these as real fields forces reps into free-text notes, which cannot be filtered. So "show me everyone with a 90 lakh budget looking in the western suburbs" is not answerable.
Building the post-visit system
Capture the visit as a structured event
The site visit should be an activity on the lead timeline with an outcome, not a gap between two calls. That means recording which project and unit was shown, the outcome, the objection and the next step: at the visit, from a phone.
Once visits are structured events, site-visit-to-booking conversion becomes a measurable funnel stage. Most brokerages cannot currently state that number, which means they cannot tell whether their problem is getting visits or converting them. Two entirely different problems with different fixes.
Make the custom fields real fields
Budget, locality, project, configuration, possession timeline, finance status. As structured fields these become filters, and filters become working lists: everyone above a budget threshold in a specific micro-market who visited in the last thirty days and has not been contacted this week. That list is the actual daily work, and it is unbuildable from free-text notes.
Enforce cadence, because ninety days will not manage itself
The mechanisms are the same as any long-cycle sale, and they matter more here because the cycle is longer:
- Scheduled follow-ups tied to the visit outcome, set at the visit rather than later.
- Call sequences for prospects who go quiet, so re-contact is a queue rather than a decision.
- Outcome tags that set the next action automatically. A callback request schedules itself.
- Colour indicators so an overdue follow-up is visible without opening the record.
- SLA timers on new enquiries, where speed is decisive.
Keep the timeline, not just the note
A chronological view of everything: calls, visits, WhatsApp exchanges, quotations, notes. Means anyone picking up the lead can reconstruct a two-month history before dialling. In a sector with meaningful rep turnover, this is the difference between inheriting a prospect and inheriting a phone number.
Why mobile-first is the whole thing
Everything above depends on the rep being able to do it from the site. If the app is a cut-down version of the web product, reps will not use it, and the system fails at the point of data entry.
What full parity has to include:
- Lead updates from the phone. Outcome, objection, next step, recorded at the visit while it is accurate.
- Click-to-call with automatic logging. The rep gets faster; the record happens as a side effect rather than as extra work.
- Quotation generation and sharing on site. Sending a priced quotation over WhatsApp during the visit, logged against the lead, closes faster than promising to email it.
- Offline capture with sync. Under-construction sites, basements and peripheral locations have no signal. If the app needs connectivity, the record does not exist for exactly the visits you most want to track.
That last point is the one most often underestimated. Offline capability is not an edge case in this sector; it is Tuesday.
WhatsApp is the channel, so log it
Property conversations happen on WhatsApp: floor plans, price lists, photographs, negotiation. Fighting that is pointless.
The problem is not the channel; it is that the conversation lives on a rep's personal phone. When they leave, the negotiation history leaves too. Sending WhatsApp from inside the CRM keeps the channel and the record: the exchange sits on the lead timeline alongside calls and visits, and it survives staff turnover.
What to measure
Five numbers, in order of usefulness:
- Site-visit-to-booking conversion. The number most brokerages cannot state, and the one that tells you whether your problem is upstream or downstream.
- Leads with no activity in seven days. The clearest leading indicator of post-visit leakage.
- Average attempts after the site visit. If it is one or two, your cadence is not running.
- Time to first contact on new enquiries. Median and 90th percentile.
- Source-wise cost per booking. Not cost per lead. Portals that generate volume and no bookings are the most common source of wasted spend in this sector, and only booking-level attribution reveals them.
For multi-office brokerages and developers
One additional layer. If you run several offices or project teams, branch-wise management makes them comparable on identical definitions. And owner-wise analysis exposes the variance within each one.
The finding is usually the same: overall conversion is a blend of a few reps who follow up relentlessly and several who do not. That is a coaching problem you can only see when the follow-up data is structured, and it is considerably cheaper to fix than buying more leads: which is the instinctive response to a conversion problem, and usually the wrong one.