An enquiry lands at 4:40pm on a Friday through the website form. It goes to a shared inbox that three people can see and nobody owns. On Monday it is under eleven newer emails. On Tuesday somebody calls, and the prospect says they have already spoken to two other suppliers and one of them has sent a quote.
Nothing went wrong in that story. No one was lazy, no system crashed. The lead simply had no owner and no clock. That is what first response time measures, and it is usually the cheapest leak in the business to close.
Why first response time is worth fixing before anything else
Most sales improvement projects are slow. Better pitch decks, better pricing, better objection handling, better collateral. All of it takes months and the results are hard to attribute.
First response time is different. It is an operational fix, not a skill fix. You are not asking anyone to sell better. You are asking that an enquiry be picked up by a named person inside a defined window. The change is mechanical, it can be in place in a week, and the effect shows up in the same month.
It also compounds. A lead called within the hour is warm and remembers filling in the form. A lead called on day four has to be re-sold on the idea that they were interested in the first place. Every downstream conversation gets harder, and the three pipeline stages where deals quietly die get more crowded.
And the failure is invisible. Nobody files a complaint about the call they never got. The revenue never appears, so nobody misses it.
Define what you are actually measuring
Before the playbook, agree on the definition. Most teams measure something flattering and then wonder why the numbers do not match the customer’s experience.
First response time is the elapsed time between the enquiry arriving in your business and a human attempting genuine two way contact with that specific person. Note the three parts of that.
- Arriving in your business, not arriving in the CRM. If a WhatsApp enquiry sat on a salesperson’s phone for six hours before it was entered, those six hours count.
- Attempting contact, not connecting. A ringing phone that nobody picks up is a genuine attempt and the clock stops there. But it only counts if the attempt is logged.
- Two way contact. An automated acknowledgement email is not a response. It is polite, and it buys you a little goodwill, but it does not stop the clock.
Decide once whether your clock runs on wall time or working hours. Both are defensible. Working hours is more honest about what a nine person team can promise, and it stops the Friday evening enquiries from wrecking every report. Whatever you choose, write it down and never quietly switch when the number looks bad.
The five-step playbook
1. One queue, all fifteen sources
Enquiries do not arrive in one place. Website form, IndiaMART, JustDial, a Google Ads landing page, a WhatsApp Business number, a trade exhibition register, an inbound call to reception, a referral forwarded by a director, a LinkedIn message. Every one of those is a lead source, and every one of them has its own hiding place.
You cannot put a clock on something you cannot see. Step one is to route all of them into a single queue before anyone tries to improve response speed. In a CRM with fifteen lead sources configured, that queue is the CRM itself. Until then it can be a single spreadsheet tab, as long as it is genuinely single.
The part people skip: reception calls and director referrals. Those are usually the highest intent enquiries in the building and they are almost never logged, because a human already spoke to the person. Log them anyway. If they are not in the queue, they are not in the number.
2. Assign an owner at the moment of arrival
A shared inbox has no owner. A queue reviewed at the 10am standup has no owner for the sixteen hours before the standup.
Every enquiry needs a named person attached to it within minutes of arrival, and that assignment has to happen without a human deciding it. Round robin works. Assignment by source works. Assignment by territory or product line works. Pick the rule that matches how your team is already organised and automate it.
The rule matters less than the fact that it is a rule. What you are removing is the pause where three people each assume one of the other two has it.
3. Set one window, in writing
One number for the whole team. Not a target per person, not a target per source.
Most SMB sales teams can hold a first response time of one working hour for digital enquiries. Some can do fifteen minutes. A field sales team where everyone is at client sites until 6pm might honestly manage three working hours, and three hours that is actually met beats one hour that is aspirational.
Write it on a wall. Say it in the Monday meeting. The window only works if every person can state it from memory.
4. Make the first touch a call, with a written ladder
Call first. An email or a WhatsApp message is easier to send and much easier to ignore. Voice gets you information a form never will: budget rough size, timeline, whether they are comparing three vendors, whether they are the decision maker.
Then write down what happens when the call does not connect, because most of them will not.
- Call attempt one within the window.
- If unanswered, a WhatsApp message within five minutes referencing the enquiry, not a generic template.
- Call attempt two the same day, at a different hour.
- Call attempt three the next working day.
- After three unanswered attempts, move the lead to a nurture cadence rather than leaving it open forever.
Five steps, no judgement calls. That is the point. Judgement calls are what produce a lead that has been in Contacted for nineteen days.
This is also where an automated first touch earns its place. BizvoraOne includes an AI voice agent that calls new leads shortly after they arrive, which covers the 9pm enquiry and the lunch hour gap. Treat it as the thing that stops the clock and gathers basic qualification, not as a replacement for the human call that follows.
5. Log the attempt where the next person can see it
An unlogged call did not happen. Not morally, but practically: the next person to touch the lead cannot see it, so they either duplicate it or assume it was done.
The log needs four fields and no more. Time of attempt, channel, outcome, next action with a date. If logging takes longer than thirty seconds, people will stop doing it under pressure, and pressure is exactly when the log matters. If your history currently lives across a spreadsheet, a phone and somebody’s notebook, moving that history into one place without losing it is the prerequisite for everything above.
Measure it honestly, not flatteringly
Here is where most teams fool themselves.
Use the median, not the average. Take five leads with response times of 4 minutes, 6 minutes, 9 minutes, 12 minutes and 31 hours. The average is over six hours, which makes the team look terrible. The median is 9 minutes, which makes them look excellent. Neither tells you the truth on its own. The median tells you what a typical enquiry experiences. Report it alongside the count of breaches.
Count the breaches, name them, read them. The single most useful weekly number is: how many enquiries missed the window, and what were they. Five named leads you can actually inspect is worth more than any percentage. Read them and you will find the pattern. It is nearly always one source, one shift, or one person’s day off.
Count the leads that were never contacted at all. These are the worst ones and they are usually excluded from the report, because a lead with no first touch has no first response time to average. Report them as a separate line item. Zero is the only acceptable figure.
Measure the ninetieth percentile too. The median tells you about the normal case. The ninetieth percentile tells you about the case that quietly cost you a deal.
Do not measure per person until the process works. Publish the team number for the first month. Individual league tables introduced too early make people close leads fast to protect their stats.
Where this does not apply
Be honest about the limits. If your enquiries are tenders with a fixed submission date, speed of first response is close to irrelevant and completeness of response is everything. If you sell one large project a quarter and every lead comes from a personal relationship, a formal window is bureaucracy you do not need.
High volume, multi source, multi salesperson inbound flow is where a first response time discipline pays. That describes most manufacturing, real estate, IT services, education and healthcare businesses with a marketing spend. It does not describe everybody.
What to do first
Do not build the system this week. Measure the current state instead.
Take the last thirty enquiries from every source you can find. For each one, write down the arrival timestamp and the timestamp of the first genuine human attempt. Where you cannot find either, write “unknown” and count those separately, because the count of unknowns is itself the finding.
Work out the median, the ninetieth percentile, and the number never contacted at all. That takes an afternoon.
Then pick one window, one assignment rule, and one place to log. Run it for two weeks and read the breaches by name. You will fix more with that reading than with any new tool.
