Most small businesses have a rough idea of how quickly they reply to enquiries, and most of those ideas are too generous. Lead response time — the gap between an enquiry arriving and your first genuinely useful reply — is one of the few numbers in a small business that decides revenue without ever appearing on a report. It is not really a customer service metric. It is a sales metric, and in a WhatsApp-first market like Singapore it is often the one that settles who gets the job.
This post is about the metric itself: what it means, why it compounds, how to measure it honestly, and what to do once you see the number. The tactical version — the mechanics of answering WhatsApp enquiries in seconds — is covered in our companion piece on replying to WhatsApp leads in 30 seconds. This one takes the wider view, across every channel an enquiry can arrive on.
What lead response time actually means
Most definitions are too loose to be useful. Count an automatic "thanks, we'll get back to you" as a reply and your numbers will look excellent while your close rate stays flat. The only definition worth measuring is the gap between the enquiry landing and your first useful reply.
- The clock starts when the enquiry arrives, not when you happen to notice it.
- The clock stops at the first reply that moves the customer forward: a price, an available slot, or a straight yes or no.
- An automatic acknowledgement does not stop the clock, because it answers nothing the customer wanted to know.
Defined that way, your lead response time is usually worse than you expect — which is what happens when the person answering enquiries is also the person doing the work.
Why lead response time is a race, not a service standard
Here is the part that changes how you think about it. The customer enquiring with you is almost never enquiring only with you. They have four tabs open, or the same message pasted into four WhatsApp threads. That is ordinary behaviour when comparing property agents, contractors, clinics or tuition centres. Nobody sends one message and waits.
That makes response time a race rather than a standard, and the race compounds. Whoever replies first asks the qualifying questions, frames what a good job looks like, and often books the slot before anyone else has answered. By the time the second business replies with a better price, the customer has committed, and re-opening a settled decision takes effort most people will not spend. The better business loses before it has said a word.
Response time does not reward the best business in the market. It rewards the one that was available at the moment the customer decided to ask.
Larger sales teams call this speed to lead and staff around it, with rotas and duty phones. A small business cannot, which is why the metric matters more here, not less.
What the research says, and what it does not
The most cited work here is the 2011 Harvard Business Review article "The Short Life of Online Sales Leads" by James Oldroyd, Kristina McElheran and David Elkington. Roughly, they found that firms attempting contact within an hour of an enquiry were around seven times more likely to reach a meaningful conversation with a decision-maker than those that waited an hour longer, and dramatically more likely than those that waited a day.
Treat that as direction rather than a promise. It studied mostly US business-to-business firms, the magnitudes are approximate, and your market is not theirs. What travels is the shape of the curve, not the multiplier: the decay is steep early, then flattens. Five minutes versus an hour matters far more than one day versus two.
Why most small businesses measure lead response time wrong
Almost nobody measures it, and those who believe they do are measuring memory. Memory favours the replies you were proud of. You remember answering someone from the car park in ninety seconds. You do not remember the enquiry that arrived at nine on a Sunday and got a reply at ten on Monday, because that customer never complained. They booked someone else, and silence leaves no trace.
The second mistake is measuring one channel and assuming the rest match. WhatsApp feels fast because your phone buzzes. Meanwhile the web form lands in an email account you open twice a week, the marketplace enquiry sits behind a muted notification, and the missed call at lunchtime is never returned. Your real lead response time is set by your worst channel, not your best.
How to measure it with what you already have
You do not need software. You need one ordinary week, a spreadsheet, and the willingness to count the enquiries you would rather not.
- List every enquiry from one normal week across every channel: WhatsApp, Telegram, missed calls, web forms, email, marketplace and directory enquiries.
- Record two timestamps for each — when it arrived, and when you sent the first reply that actually answered something.
- Take the median rather than the average, because one three-day reply distorts an average and hides the typical case.
- Then sort the list slowest first and study the worst ten per cent, because that is where the lost jobs are.
Nearly every business that does this finds the same pattern. The median looks respectable: a few minutes, in working hours, on the channel they watch. The tail is where the damage sits — evenings, weekends, public holidays, and the hours spent mid-job with both hands full. For home services, that is most of the week.
The tail is also the honest answer to what fast has to mean here. Where people message businesses the way they message friends, minutes read as normal and hours read as indifference. You are judged against whoever answered second-fastest.
Fast is not the same as useful
It is possible to fix your response time and change nothing. An instant reply that does not answer the question buys little, because the customer keeps shopping while waiting for the real answer. The test is simple: after your first reply, does the customer have any reason to stop looking?
- A useful first reply answers the actual question, with a number or a date wherever you can give one.
- If you cannot quote yet, it narrows the gap with one specific question rather than a general request for details.
- Where possible it offers a next step the customer can take now, such as a time they can confirm — which is why an online booking flow beats a faster acknowledgement.
A ladder of fixes, ordered by effort
There is no single switch that makes you respond to leads faster, only a sequence. Work down it in order: each rung is cheaper than the one below, and each makes the next easier.
- Consolidate your channels into one inbox, so answering no longer depends on which app you opened.
- Set honest expectations: publish hours you can actually keep, instead of implying you are always available.
- Prepare written answers to the five questions you get every week: price, availability, coverage area, how long the job takes, and what is included.
- Cover the gaps rather than the whole day, using WhatsApp automation for the evenings, weekends and mid-job hours where your tail lives.
- Then automate the answer itself, so the first reply is useful and not merely prompt.
Most owners jump straight to the last rung and are disappointed, because an assistant with nothing to answer from can only stall politely. The middle rungs are what make the last one work: the difference between an AI receptionist that quotes your prices and one that promises somebody will be in touch.
What stays human, and where microcrew fits
Not everything should be answered in thirty seconds by a machine. Complaints, bad news, negotiation beyond your published range, and anything with real emotional weight belong to you. So does any enquiry where the honest answer is that this is not quite what you do. A setup worth trusting knows the edge of its own competence and hands over rather than improvising. Test that handoff first.
microcrew is built for the part below that line. It replies to WhatsApp and Telegram enquiries in about thirty seconds, around the clock, in the customer's language, answering from your own prices, policies and documents in Google Drive. It books into Google Calendar, captures leads from allow-listed Gmail senders so email enquiries land beside chat ones, follows up leads who went quiet — and can call them back in your own cloned voice — and shows enquiries, conversations and bookings in one place. NRIC and card numbers are redacted at ingestion, AI processing runs on Google Vertex AI in Singapore under a no-training agreement, and retention is bounded.
Plans are SGD $99, $199 and $399 a month, and access is invite-only for now. If your worst ten per cent is costing you work, request early access.
