The short answer
Inside a minute if you can staff it, inside five if you cannot. A lead you never reach converts at zero no matter how good the rest of your process is, and contact rate sits first in the chain that produces cost per booking, so it multiplies every rate after it. On the worked funnel from our cost-per-lead article, moving contact rate from 60% to 85% and changing nothing else cuts cost per booking 29%, from ₹2,16,450 to ₹1,52,788. No new creative, no new targeting, no change in cost per lead.
Why the first term is the one to attack
Cost per booking is cost per lead divided by the product of the five rates after the lead: contacted, qualified, site visit booked, showed up, closed. Because they multiply, a proportional gain anywhere is worth the same as a proportional gain anywhere else. Contact rate is not special in the arithmetic. It is special in practice, for two reasons.
- It is usually furthest from its ceiling. Qualification and close rates are constrained by the market and your product. Contact rate is constrained by whether someone dialled, which is a staffing and routing decision you control outright.
- It is the cheapest to move. Nobody has to get better at selling. Somebody has to pick up the phone sooner, and a system has to make sure the lead reaches them.
| Before | After | |
|---|---|---|
| Contact rate | 60% | 85% |
| Lead to booking | 0.693% | 0.982% |
| Leads per booking | 144 | 102 |
| Media per booking at ₹1,500 CPL | ₹2,16,450 | ₹1,52,788 |
A 29% cut in the cost of a booking from one rate, with the media plan untouched. And because contact rate is bounded at 100%, you can see exactly how much headroom is left. If you are at 60%, you know there is 40 points of it.
What actually decays in those minutes
Three things, and it is worth being precise about them, because the mechanism tells you what to build.
Memory and frame. They did not fill one form. They filled several in one sitting, because that is how anyone shortlists. Whoever calls first is a continuation of what the buyer was already doing. Whoever calls on Tuesday is an interruption from a company they have to be reminded about.
Availability. The moment of maximum intent is the moment they were already on their phone looking at property. That window closes on its own. Twenty minutes later they are driving, in a meeting, or back at work.
Competition. Broad targeting means every developer in the segment is reaching a heavily overlapping audience. You are not competing for the buyer's attention in the abstract. You are competing with four other callers on the same afternoon.
You will find plenty of quoted multiples for how much better a one-minute callback performs. We are not going to repeat a number we cannot source to your account. The direction is not in dispute, the magnitude is account-specific, and it is measurable from data you already hold.
Measure your own decay curve
Your call logs carry a timestamp and your CRM carries a lead creation time. The difference between them is the only speed-to-lead study that should set your policy. Bucket every lead from the last quarter and fill this in.
| Time to first dial | Leads | Contact rate | Site visit booked |
|---|---|---|---|
| Under 1 minute | |||
| 1 to 5 minutes | |||
| 5 to 30 minutes | |||
| 30 to 120 minutes | |||
| 2 hours to same day | |||
| Next day or later | |||
| Never dialled | 0% | 0% |
Two things usually fall out of that table the first time anyone builds it. The never-dialled row is larger than management believed, and the drop between the first two buckets and the rest is steeper than anyone expected. Both are worth more than another round of creative testing.
The arithmetic of a second dial attempt
One dial is not a contact attempt. If a single dial connects with probability p and you treat attempts within a short window as roughly independent, n attempts connect with 1 minus (1 minus p) to the power n.
Connect probability after n attempts = 1 - (1 - p)^n| Attempts | Cumulative connect |
|---|---|
| 1 | 45.0% |
| 2 | 69.8% |
| 3 | 83.4% |
| 4 | 90.8% |
The independence assumption is doing real work there and it is not quite true, because someone who never answers a first call is a different kind of person from someone who happened to miss one. So treat the table as the shape rather than the forecast: the second and third attempts carry most of the recoverable contact rate, and after the fourth you are mostly annoying people. Measure your own multi-attempt curve the same way you measured the decay curve.
What makes the number achievable
A service level is a staffing decision, not an aspiration written on a whiteboard. Five rules make the difference between a policy and a number that quietly drifts.
- Decide who owns the hours your leads actually arrive in. Property gets browsed in the evening and at weekends. An SLA that only holds between 10am and 6pm on weekdays is an SLA for a minority of your leads.
- Automate the first touch, humanise the second. An instant WhatsApp carrying the project's approval status and a booking link holds the frame while a person gets to the phone. It is not a substitute for the call.
- Route by source, not round robin. The person who knows the project should get the lead from that project's campaign, or the first two minutes are spent on internal lookup.
- Define attempts and a window before you define outcomes. How many dials, over how long, before a lead is marked unreachable. Without that rule, unreachable means whatever the busiest telecaller decides it means.
- Report time to first dial weekly, on the same dashboard as cost per lead. A rate nobody reports is a rate that decays back to where it started within a quarter.
Questions people ask
- How quickly should you call a real estate lead?
- Inside a minute where staffing allows, inside five otherwise. The reason is arithmetic rather than etiquette: contact rate is the first of the five rates between a lead and a booking, and because those rates multiply, it scales everything after it. Raising contact rate from 60% to 85% with no other change cut cost per booking by 29% in our worked example.
- How many times should you call a lead before giving up?
- Three to four attempts within a defined window captures most of the recoverable contact rate. At a 45% single-dial connect rate, treating attempts as roughly independent, cumulative connect goes 45%, 70%, 83%, 91% across four attempts. That independence assumption makes it an optimistic ceiling, so measure your own curve, but the shape holds: the second and third attempts do most of the work.
- How do I measure speed to lead without buying a tool?
- Take lead creation time from your CRM and first dial time from your call logs, bucket every lead from the last quarter by the difference, then compute contact rate and site-visit-booked rate for each bucket. That gives you your own decay curve, which is the only one that should set your policy. The never-dialled row is usually the finding.
- Does an automated WhatsApp reply count as speed to lead?
- It helps and it does not substitute. An instant automated message carrying approval status and a booking link holds the buyer's attention while a person gets to the phone, which is valuable because the buyer is shortlisting several projects at once. Contact rate in the funnel should still be measured on human contact.