Stopwatch with a glowing hand near the top of the dial, a picture of speed to lead

Speed to Lead: Why 5 Minutes Is the Number

Speed to lead is the gap between someone raising their hand for your business and a human from your business making contact, and the number worth running on is five minutes. That figure traces to a 2007 study of web leads which found the odds of reaching a person fall roughly 100 times between a five minute callback and a thirty minute one, and the odds of a real conversation fall about 21 times. The finding holds up and is worth acting on. It is also from 2007, it never measured whether anyone bought anything, and most pages quoting it have never opened it.

Key Takeaways

  • The five-minute number comes from a 2007 study by Dr. James Oldroyd and InsideSales.com: six companies, over fifteen thousand web leads, over one hundred thousand call attempts.
  • Between five and thirty minutes, the odds of reaching a live person drop about 100 times and the odds of a real conversation about 21 times. Neither is a close rate.
  • Harvard Business Review audited 2,241 US companies in 2011: a 42 hour average response, and 23% never responded at all.
  • Your clock starts when the customer hits submit, not when you see the notification.
  • The 78% statistic, that buyers hire whoever responds first, has no study behind it that we could find.

If a web form came in while you were out on a service call, you saw it at lunch, and your callback went to voicemail, you already know what slow speed to lead feels like. This page covers where the five-minute rule comes from, what it proves and what it never measured, and how to hit it from the truck. I have spent more than fifteen years in search, and I keep score in booked calls, not clicks.

What does speed to lead mean?

Speed to lead, also called lead response time, is one measurement: how long from a prospect signalling interest to a person at your company making contact, timed from the moment they reach out rather than the moment you notice. The signal can be a form, a chat, a text, or a missed call. The metric does not care which.

In our AI gap report analysis of 17 local businesses across 1,352 AI answers, we found 5.5% of answers named only one business, leaving no runner-up at all.

Here is where most owners measure it wrong. They start the clock when they see the lead. On the surface that makes sense, because the notification is the first moment you could have done anything about it. But the customer started the clock when they hit submit. Everything in between counts, and in a small service business that gap is where the minutes disappear: the form emails an inbox, the inbox lives on a phone in a truck, and the phone is face down on the seat until lunch. The customer waited two hours. The owner would say the response was instant, because from the moment he saw it, it was.

To get a real number, take your last twenty leads and note the timestamp on the submission and on your first outbound contact. Use the median, not the average: one Saturday night lead answered on Monday wrecks an average. Then split business hours from after hours, since those are two different problems. If you are not certain the form on your website even delivers reliably, start there. My rule is that a lead only counts once the form's endpoint confirms it arrived, so send yourself a test submission and see where it actually lands. A short conversation is usually enough to sort out what the numbers turn up.

One honest limit: we cannot always read a client's call data directly, and for one client the counts come from a call-log export they send us. The tap-to-call count in your website analytics is not a call count either, because it misses everyone who dials from Google's results or types the number in; in one comparison, taps fell from 38 to 32 while tracked calls rose from 38 to 50 over roughly the same period. If missed calls are part of your speed to lead problem, count from the call log.

Where does the five-minute rule actually come from?

The five-minute rule has one origin, and it is worth naming properly, because almost nobody does.

In October 2007, Dave Elkington of InsideSales.com and Dr. James Oldroyd, then a faculty fellow at MIT's Sloan School of Management, presented a study called Lead Response Management at MarketingSherpa's Business-to-Business Demand Generation Summit. They examined three years of data across six companies that generate and respond to web leads, covering more than fifteen thousand leads and more than one hundred thousand call attempts.

Two definitions matter before the numbers do. A contact meant a call that connected with a live person and lasted a defined number of seconds. A qualification meant the point where a lead was willing to enter the sales process, usually agreeing to an appointment. Hold that distinction, because the numbers everyone repeats are about those two things and nothing else.

The headline finding, in the report's own terms: the odds of contacting a lead called at five minutes versus thirty minutes drop 100 times, and the odds of qualifying a lead over that same gap drop 21 times.

And the limitation, which the report states about itself: it did not address close ratios. So nobody can honestly tell you the five-minute rule makes people buy. What the data supports is narrower and still valuable. Answering fast makes people answer, and makes them willing to talk. What happens in that conversation is on you.

The age is the other thing to say out loud. This is 2007 data, drawn from business-to-business web leads on one vendor's calling platform. It is directionally sound, it is the best documented finding on the subject, and anyone presenting it as a current 2026 measurement of your industry is guessing. We would rather hand you a nineteen-year-old number you can check than a fresh one you cannot. What it means for one specific business is worth a conversation.

How quickly should you respond to a lead?

Within five minutes. Asking how fast you should respond to a lead gets one number. The more useful question is where on the curve the expensive part sits: not the first day, the first ten minutes.

Between five and ten minutes, the 2007 study found the odds of reaching a live person fall by about five times. Five minutes to thirty is where the floor drops out. Across the whole first hour, contact odds fall by more than ten times, then the curve flattens, because most of the damage is done.

Gap in response time What the data found Source and year
5 minutes vs 10 minutes Odds of reaching a live person fall about 5 times; odds of qualifying fall about 4 times Lead Response Management study, 2007
5 minutes vs 30 minutes Odds of reaching a live person fall about 100 times; odds of qualifying fall about 21 times Lead Response Management study, 2007
Across the first hour Odds of reaching someone fall by more than 10 times; odds of qualifying by more than 6 times Lead Response Management study, 2007
Within 1 hour vs one hour later Nearly 7 times as likely to qualify the lead Harvard Business Review, March 2011
Within 1 hour vs 24 hours or more More than 60 times as likely to qualify the lead Harvard Business Review, March 2011
After 20 hours Each additional call attempt starts to hurt the odds of making contact Lead Response Management study, 2007

Read that table honestly. It is assembled from two studies four years apart, with different samples; the 2011 rows come from a separate dataset of 1.25 million leads across 29 consumer-facing and 13 business-to-business US companies, where qualifying meant a meaningful conversation with a key decision maker. It is not one clean curve from one experiment, and anyone drawing it as a single smooth line is drawing, not measuring.

How slow is everyone else, really?

Slow: in the best-known audit, the average reply took 42 hours and almost a quarter of businesses never replied at all. In March 2011, Harvard Business Review published the audit that should be pinned above every service business owner's desk. The researchers sent a web-generated test lead to 2,241 US companies and measured what happened. Thirty-seven percent responded within an hour. Sixteen percent responded somewhere between one and 24 hours. Twenty-four percent took longer than a day. And 23% never responded at all. Among the companies that did respond within 30 days, the average response time was 42 hours.

Read that last group again. Nearly one in four businesses paid to generate a lead, received it, and never answered it.

That is the field you are actually competing in. You are not up against a wall of five-minute operators. A same-day callback already puts you in the better third of the businesses that study looked at, and a five-minute one puts you somewhere most of your competitors have never been.

One honest gap: nobody has published a credible, methodology-backed version of that audit for the trades. Plenty of pages will tell you that only 12% of contractors answer within five minutes. None of them attach a dataset, so we are not repeating the figure. Knowing where you sit in that spread takes an afternoon, and we are glad to help you read the result.

Which speed to lead statistics should you not trust?

Three numbers circulate constantly on this topic. One is unsourced and two are mangled.

"78% of customers buy from the company that responds first." This is the most repeated speed to lead statistic on the internet. We went looking for the study behind it in September 2026 and could not find one: no report, no sample size, no methodology, no publication date. Every citation leads to another blog citing another blog, usually crediting an unnamed survey nobody links. It may be true. It is not evidence.

"The average contractor responds in 42 minutes." The 42 is real. The unit is not. Harvard Business Review reported a 42 hour average, across 2,241 audited companies of every kind, in 2011. Somewhere along the way the hours became minutes and the general population became contractors. Same digits, different planet.

"Leads contacted within five minutes are 21 times more likely to convert." Both halves of this are wrong. The 21 times figure is about qualification, not conversion, and the 100 times figure that usually gets attached to the thirty-minute mark is about contact. The pairing gets flipped constantly, and the word "convert" gets bolted on to a study that says in plain language that it did not measure close rates.

Here is the rule worth keeping past this page. If a statistic arrives without a study name, a year, and a sample size, treat it as a slogan. That test costs nothing and it will disqualify most of what you read about marketing.

What does a five-minute response look like when you are on a roof?

It looks like two steps sized for a day spent on jobs: a fast acknowledgement, then the real conversation. The honest objection first: you are in an attic, under a sink, or on I-65 with a trailer. Nobody becomes five-minute responsive through willpower, and advice that assumes a sales team at desks is advice for a business that is not yours.

So split the job in two. The five-minute clock belongs to acknowledgement, not to the conversation. What the customer needs inside five minutes is proof a human has their request and a real time when they will hear back. The actual conversation, the one where you ask what is wrong with the unit and when you can get out there, can happen at minute forty and still win the job. If you can only make one thing fast, I would make it the acknowledgement.

An experienced salesman named Mike taught me why that first small step matters, back when I sold furniture. Mike had shoppers lie on their backs under a couch to count the springs, and pull the cushions apart to see how it was built. Some people took naps on the beds. At the end he would simply ask, "We love this couch. Do you love this bed?" They said yes, and they bought, every time: tens of thousands of dollars of furniture. You only count springs under a couch when part of you has already decided.

Action creates commitment. A homeowner who texts back a good time to call, or picks a slot for the visit, has done something, and that small act tends to hold them while other companies are still dialing. Your five-minute reply is the moment to ask for it.

Three HVAC companies receive the same form at 10:40 on a Tuesday morning. The first replies at 10:43 with a name and a one hour callback window. The second calls at 4pm. The third calls Wednesday. By 4pm that homeowner has usually stopped shopping, and the 2007 curve says the gap between 10:43 and 4pm is where the job was decided, not the quality of anyone's quote.

The shape of the fix is not complicated, which is different from saying it is easy. Form submissions go to a phone that rings, not only to an inbox. An unanswered ring is the same clock; what a missed call costs is its own arithmetic. Whoever answers that phone can book a job without asking permission. The after-hours message states a real callback time instead of implying a service level you cannot hold. And the path from your site to that first contact is short enough to survive a bad signal, which is as much a website conversion rate question as an operations one. If that sounds like one more job on a full day, we can map it with you.

The form is part of the reply too. My rule for a landing page is a short form asking for name, email and phone, then only what the office needs to schedule, ending with a line that tells the customer a call is coming shortly, with one phone button beside it that shows the number. That line is a promise your five-minute process has to keep.

If you use call tracking numbers, the order of setup decides whether calls still reach you. We read the tracking script first to see which number it swaps, get the call-routing list before changing any number, and put the real business line on the site as the number to swap.

Does speed to lead matter?

Yes, but only as much as the answer behind it. Speed multiplies whatever your process already does. If the answer is good, five minutes makes it better. If the person answering has no information, no authority to book, and no idea what the customer filled out, then ninety seconds is just a faster way to sound unprepared. A competent callback at minute twelve beats a useless one at minute two, because what the customer remembers is the answer, not the timestamp.

If the leads themselves are wrong, speed makes you efficient at losing. Where your leads come from and what they cost is a separate question with separate math, and fixing response time will not rescue a bad source.

The 2007 research had a second half worth a mention: a survey of 495 companies across more than 40 industries found that each tier of delayed response lined up with roughly 4.3% fewer qualified leads, and each extra unproductive call attempt with about 5% fewer. That half was self-reported by sales and marketing managers rather than observed, so read it as a direction rather than a measurement.

Where I land: pick the hours you can genuinely cover and be five-minute fast inside them. Outside them, I would not promise that speed at all; I would set an after-hours message with a real callback time and keep it, because an honest wait holds up better than a reply you cannot deliver. If you would rather see how your setup looks from the outside first, get your audit.

Frequently asked questions

What is a good speed to lead benchmark for a small service business?

Acknowledge every lead inside five minutes during the hours you cover, and measure the median rather than the average so one weekend outlier does not hide the pattern. Split business hours from after hours. No credible benchmark specific to the trades has been published, so anyone quoting you a contractor-specific number is quoting marketing.

Is the five-minute rule still true in 2026?

Directionally yes, precisely unknown. The evidence is the 2007 Lead Response Management study, and nothing published since has matched its methodology in a form we could verify. Buyer behaviour has gotten faster since 2007, not slower, so the direction is safe. Treating the exact multipliers as current measurements is not.

Does a text message count, or do I have to call?

The 2007 study measured phone contact only, so the 100 times figure does not transfer to text. What it supports is that the customer needs a human response fast. A text naming a person and a real callback window is a legitimate acknowledgement, and it buys you time to have the real conversation properly.

Should I respond to leads at night and on weekends?

Cover the hours you genuinely can, and set an after-hours message that gives a real callback time instead of implying instant service. An overpromise at 11pm costs more than an honest wait. The 2007 findings about the best days and times to call describe outbound calling patterns, not inbound response, so do not stretch them.

How many times should I follow up if nobody answers the first call?

The verified data speaks to when you call, not how many times, and we will not invent a number. It does say that after about 20 hours each additional dial starts to hurt your odds of making contact, and that extra unproductive attempts track with lower qualification, though that second finding was self-reported. Persistence early beats persistence late.

One number to find this week

Take your last twenty form submissions and find the median gap between submission and your first real contact, business hours and after hours separate. Most owners have never looked at that number, and it is usually the most uncomfortable in the business.

It is also the cheapest marketing variable to change: it costs nothing to shorten and applies to leads you have already paid for. Treating response time as an economic input rather than an office habit is the work of a fractional CMO, a part-time marketing leader, which is what GetLocalLeads.AI's executive-level consulting is for. Book a call and bring the number.