Cost per lead is the wrong metric, and it is making projects worse

The metric rewards the wrong behaviour

Cost per lead is the dominant reported metric in Indian real estate marketing, and it has a structural flaw: it can be improved almost arbitrarily by degrading lead quality.

Broaden the audience, remove qualifying questions from the form, bid into cheaper placements, run a generic offer creative rather than a specific one — and cost per lead falls. Lead volume rises. The dashboard improves. The sales team spends its day calling numbers that do not connect, and site visits are flat.

The reverse error is equally damaging and less visible. A campaign producing a small number of expensive but well-qualified enquiries looks like a failure on a cost-per-lead report while being the most profitable line in the plan. Developers routinely switch off the campaigns that were working.

The metric ladder

The useful framing is a ladder of metrics, each closer to revenue than the last, with the honest observation that the further down you measure, the more infrastructure it requires and the longer the feedback loop.

Why cost per qualified site visit is the practical answer

Cost per registration is the honest number, but it lags by several months — too slow to optimise a live campaign against. Cost per booking is better but needs volume to stabilise, and in a project selling eight units a month the sample is thin.

Cost per qualified site visit sits at the point where the metric is both meaningful and fast enough to act on. A site visit requires the prospect to invest travel time, which filters out most noise. Volume is usually sufficient for weekly decisions. And it correlates well with bookings, so improving it tends to improve the number that matters.

The prerequisite is a consistent definition of 'qualified' that does not drift — which in practice means call-quality audits against a published rubric, because without them a sales team under pressure will qualify anything.

Attribution has to reach the CRM

None of this is possible if attribution stops at the ad platform. Platform reporting tells you which campaign generated form fills. It cannot tell you which campaign generated bookings, because the booking happens in the CRM weeks later.

Closing that loop means capturing source, campaign, ad set, creative and keyword at lead level into the CRM, preserving it through every stage transition, de-duplicating across portals, forms and calls, and feeding qualified-visit and booking events back to the platforms as conversion signals.

It is unglamorous work and it is the single highest-return investment in a real estate marketing operation. After sixty to ninety days it produces something most developers have never had: a ranked list of audiences and creatives by cost per booking, and the confidence to switch off the ones generating volume and no revenue.

• Lead-level source and campaign capture into the CRM
• Call tracking with recorded-call quality scoring
• De-duplication across portals, web forms and phone
• Offline conversion upload back to the ad platforms
• Monthly reconciliation of spend to bookings and registrations

What to do if you cannot measure that far

Some developers genuinely cannot close the loop yet — the CRM does not support it, or the sales operation is not disciplined enough for the data to be trustworthy.

In that situation the honest position is not to fall back on cost per lead but to use the best available proxy and to say so. Cost per contactable lead, measured by actually sampling whether numbers connect, is crude and far better than nothing. Site visit volume against spend, even without lead-level attribution, gives directional guidance.

What should not happen is a campaign optimised to cost per lead while everyone involved knows the metric is misleading. That is how a developer ends up with twelve thousand leads and forty bookings.