The Five-Minute Window: What the Lead-Response Research Actually Says

If you have read anything about responding to sales leads quickly, you have met the five-minute rule. Respond within five minutes, the story goes, and your odds transform. It appears in a thousand blog posts, usually attached to a confident-sounding multiplier.

The underlying research is real, and it is worth understanding properly — partly because it supports a genuinely useful conclusion, and partly because a lot of what gets quoted alongside it is fabricated. Here is what actually exists.

Study one: the 42-hour average

In March 2011, Harvard Business Review published findings from James Oldroyd, Kristina McElheran and David Elkington, who audited how quickly companies responded to inbound web enquiries. The sample was substantial: 2,241 U.S. companies.

Among firms that responded at all within thirty days, the average time to first response was 42 hours.

The performance difference by speed was stark. Firms attempting contact within an hour of the enquiry were nearly seven times more likely to qualify the lead than those attempting contact even an hour later, and more than sixty times more likely than those who waited twenty-four hours or longer.

Study two: where the five minutes comes from

The specific five-minute threshold traces to earlier work by Oldroyd, conducted with InsideSales.com across a dataset of more than 15,000 leads. That analysis found that the odds of qualifying a lead were roughly 21 times higher when contact was attempted at five minutes rather than at thirty minutes.

That is the origin of the 21× figure. It is real, it is traceable, and it is narrower than almost everyone quoting it realises.

Three caveats that get dropped

It measures qualification, not revenue. "Qualifying" a lead means reaching a decision-maker and establishing that a real opportunity exists. It is not the same as signing a contract. No part of this research measured close rates or revenue, and anyone telling you speed makes you 21 times more likely to win the job is extending the finding past what it says.

The data is old. The lead-response work dates to 2007 and the HBR audit to 2011. Consumer behaviour around phones has changed considerably since. The direction of the effect is almost certainly still right — more recent replications point the same way — but the exact multipliers should be treated as illustrative.

It was not about contractors. The samples were broad B2B and consumer web enquiries, not luxury home services. The mechanism transfers well — a homeowner shopping three pool builders behaves a lot like anyone else comparing vendors — but nobody has run this study on your industry.

Two numbers you will see everywhere that we won't use

This is the part most articles skip, and it's the reason we wrote this one.

"78% of customers buy from the first business that responds"

This appears constantly, usually credited to a "Lead Connect survey." There is no published methodology and no original report. Researchers who have gone looking for it have not found it. It is industry folklore that acquired a decimal point somewhere along the way, and we won't repeat it.

You don't need it, either. The HBR qualification data makes the same point with actual evidence behind it.

"The average contractor responds in 42 minutes"

This one is interesting, because we suspect we know how it happened. The real, sourced figure is 42 hours, from the HBR audit. Somewhere in the chain of blogs citing blogs, "hours" became "minutes," and the corrupted version now circulates as a contractor-specific statistic with no study behind it.

We used the 42-minute version ourselves in a social post before checking it. We were wrong, and the graphic that accompanied that post had the correct figure — 42 hours — sitting right there in it.

So what should a contractor actually take from this?

Strip out the folklore and the defensible conclusion is still strong, and still useful:

• Response speed materially affects whether an enquiry becomes a conversation. That is well evidenced across large samples.
• The decay is front-loaded. The difference between five minutes and thirty is larger than the difference between two hours and three. Early minutes carry disproportionate weight.
• Most businesses are dramatically slower than they think. A 42-hour average across 2,241 companies is not a story about a few bad performers. It's the middle of the distribution.
• Nobody has measured what this is worth on a $150,000 backyard. The direction is clear. The magnitude for your specific business is not, and anyone quoting you a precise figure is inventing it.

How to measure your own, in about a week

Rather than adopt anyone's benchmark, measure yourself. It takes almost no effort and the number is usually a surprise.

• Log every inbound enquiry for seven days. Calls, web forms, DMs, texts. A note on your phone is enough — you don't need software for this.
• Record two timestamps for each: when it arrived, and when a human from your business first made contact. Not when you saw it. When you responded.
• Count the ones that never got a response at all. These are the ones that hurt, and they're easy to forget precisely because nothing happened.
• Take the median, not the average. One enquiry you answered in thirty seconds will drag an average down and hide the real pattern.

Two things usually fall out of this. The first is that the median is much longer than the owner guessed — typically hours, not minutes. The second is that the misses cluster: evenings, weekends, and the middle of the working day when everyone is on site. That clustering is the useful part, because it tells you exactly which window is worth automating first.

The practical version

For a contractor, "respond in five minutes" is not a discipline problem. It's an availability problem. You cannot answer in five minutes while you are running a crew, and you shouldn't try — the work is the work.

Which leaves automating the first touch: an instant text when a call is missed, and a voice agent that answers when the office is closed. Not to replace you. To hold the line for the four minutes before you can get there.

You can test that on our demo line, or book thirty minutes and we'll work out what it's worth against your actual numbers rather than a borrowed statistic.