Your website converts fine. Your response time is where the leads die

Almost every conversion optimisation project stops at the form submission, as though the job ends when someone presses send. For a lot of businesses the largest single leak in the funnel is the gap between that moment and the moment a human replies — and unlike most CRO work, closing it costs nothing in traffic, ad spend or design.

Where the famous numbers actually come from

Before the advice, the provenance — because the statistics in this field are quoted constantly and attributed wrongly almost as often, and a misattributed number is a number you cannot defend in a meeting.

Two corrections worth making, because they affect how much weight the numbers can carry.

The 21× figure is not Harvard's. It belongs to the Oldroyd study out of MIT Sloan, conducted with InsideSales.com. A large share of pages ranking for this topic credit it to Harvard Business Review, conflating it with HBR's separate and later audit. If you are citing it in a board paper, cite MIT.

The 42-hour figure is from 2011. It appears across 2026 statistics round-ups presented as current data, usually without a date. It may well still be directionally true — more recent mystery-shopper work finds a majority of companies never replying at all — but a fifteen-year-old average is not a 2026 benchmark, and treating it as one is exactly the kind of unsourced repetition that makes this whole topic feel like folklore. Check the vintage of any response-time statistic before you build a business case on it.

What the research actually shows, and what it does not

The core finding survives the scrutiny: the odds of reaching and qualifying an inbound lead fall steeply within the first hour, and most of the damage happens in the first thirty minutes. That is a robust, repeatedly reproduced pattern, and it makes intuitive sense — someone who filled in your form thirty seconds ago is at their desk, thinking about the problem, and has probably just submitted the same form to two competitors.

What the research does not establish is that response speed alone causes the improvement. Companies that answer in five minutes tend to differ in other ways: they are better staffed, they have defined ownership of inbound enquiries, they have a CRM that works. Some of the measured gap belongs to those things rather than to the clock. The honest reading is that response time is a strong lever and a reliable symptom, and improving it usually requires fixing the process problems that were also costing you deals.

The decay curve

The shape matters more than any individual row. This is not a gentle linear decline where being twice as fast is twice as good. The expensive losses are concentrated in the first half hour, which means the operationally useful target is not "respond faster" — it is "cross the five-minute line, or accept that you are in the slow group".

What it is worth, and the honest version of that number

Take a business receiving 60 enquiries a month at an average deal value of $3,000, currently responding within about a day.

Now the caveat that the rest of the internet omits. That $36,000 gap is the benchmark difference between fast and slow companies, not the amount you will gain by setting an alarm. Because the comparison is correlational, a realistic expectation from response time alone is a fraction of it. Even a quarter of that gap, though, is $9,000 a month from a change that requires no additional traffic, no additional ad spend and no redesign — which is a better return than almost anything else on a typical CRO backlog.

Run the same table with your own numbers before doing anything else. If you are a low-volume, high-value business, the case is usually overwhelming. If you sell a $40 product to 4,000 people a month, it is probably not your priority at all.

How fast is fast enough?

"Five minutes" is the headline, but it is not the right target for every business. The real question is how long your buyer waits before contacting someone else.

The row that surprises people is the last B2B one. At high deal values, an instant auto-response from a chatbot can actively damage credibility. Speed is a proxy for attentiveness, and at that end of the market a considered same-day reply from a named person demonstrates it better than an instant generic one.

The delay is structural, not lazy

Businesses that respond in 42 hours are rarely ignoring their leads. Something in the path is broken, and it is nearly always one of these:

• The notification goes somewhere nobody watches. A shared info@ inbox, or a form plugin emailing an address that forwards to a filtered folder.
• Nobody owns inbound. When a lead belongs to everyone, the first fifteen minutes belong to nobody.
• The form submits into a void. No CRM, no task, no assignment — just an email that has to be noticed, read and acted on by a person who is doing something else.
• Enquiries arrive outside business hours. A meaningful share of web enquiries arrive in the evening or at weekends, and a Monday-morning reply to a Friday-night enquiry is a 60-hour response no matter how sharp the team is on weekdays.
• Deliverability quietly fails. Notifications land in spam, or the form's mail configuration breaks silently after a hosting change. This one is common and invisible — the enquiries simply stop, and it gets blamed on the market.

That last point deserves a five-minute check today: submit your own contact form and time how long it takes for the notification to reach a human being's screen. A surprising number of businesses discover the answer is "never". If you find that your forms have been failing, the wider question of why a site is not producing leads often answers itself.

Six changes that cut response time

In rough order of effort against return. The first three are website changes; the rest are process.

1. Put a calendar booking option next to the form. The fastest possible response is one that does not require a response. For considered services, letting someone book a slot removes the delay entirely for the segment most ready to buy. Keep the form for everyone else.
2. Send an immediate, honest acknowledgement. Not "we value your enquiry" — a real expectation: who will reply, and by when. This does not improve your actual speed, but it measurably reduces the chance the buyer keeps shopping while waiting, and it costs nothing.
3. Route form submissions to a person, not an inbox. Assign to a named owner with a notification they will actually see, and keep a copy in a CRM or task list so nothing depends on someone noticing an email.
4. Be honest about hours on the page. If you do not answer at 10pm, say when you do. Setting a correct expectation outperforms silence, and it stops out-of-hours enquiries counting as failures.
5. Set a written SLA and measure against it. Pick the target from the table above, write it down, and report median response time weekly. An unmeasured target is a preference.
6. Cut the fields that slow qualification rather than adding more. Long forms are often justified as pre-qualification, but they reduce volume and rarely speed up the human step. Ask for what you need to make the first call useful, not everything you would eventually like to know — the trade-offs are covered in more detail in how many fields a lead-gen form should have.

What to measure

Three numbers, reviewed weekly. Median time to first response, not mean — a handful of forgotten enquiries will drag a mean into meaninglessness while hiding a decent typical performance. Percentage answered within your SLA, which is the number to actually manage. And percentage never responded to, which is uncomfortable and usually higher than anyone expects.

Track those alongside conversion rate rather than instead of it. A site converting 4% of visitors into enquiries while 30% of those enquiries go unanswered is not a 4% conversion rate in any sense that matters to the business — it is 2.8%, and the missing 1.2% is more expensive than anything you could fix on the page, because you already paid to acquire it. This is the same gap between measured conversions and commercial outcomes that makes cost per qualified lead a more honest metric than cost per lead.

Frequently asked questions

What is speed to lead and why does it matter?

Speed to lead is the time between someone submitting an enquiry and a human responding to it. It matters because the odds of reaching and qualifying an inbound lead fall steeply within the first half hour. The MIT Sloan study by Dr James B. Oldroyd with InsideSales.com found that calling at 5 minutes rather than 30 minutes improved the odds of making contact by around 100 times and the odds of qualifying by around 21 times, based on three years of data across six companies, 15,000 leads and over 100,000 call attempts.

Is the 21x lead response statistic from Harvard?

No, and this is one of the most widely repeated attribution errors in marketing. The 21 times qualification figure and the 100 times contact figure come from the Lead Response Management study conducted by Dr James B. Oldroyd at MIT Sloan with InsideSales.com. Harvard Business Review published a separate and later audit of 2,241 companies, which is the source of the 42-hour average response time and the finding that a substantial share never respond at all. Cite MIT for the multipliers and HBR for the response-time averages.

Is the average lead response time really 42 hours?

That figure comes from a Harvard Business Review audit published in 2011, and it is frequently reproduced in 2026 statistics round-ups without a date, which makes it look current. It may still be directionally accurate, and recent mystery-shopper testing has found large shares of companies never replying at all, but a fifteen-year-old average should not be treated as a 2026 benchmark. Measure your own median rather than relying on a borrowed number.

Should every business aim to respond within five minutes?

No. Five minutes is the right target for urgent local services, where the buyer is working down a list until someone answers. For considered local services, under 30 minutes in business hours is usually enough, and for mid-value B2B, under an hour. For high-value or committee-led B2B purchases, a considered same-day reply from a named person outperforms an instant automated one, because at that deal size speed can read as carelessness rather than attentiveness.

How much revenue is slow lead response actually costing?

Work it out with your own numbers rather than a benchmark. For a business taking 60 enquiries a month at a 3,000 dollar average deal value, moving from next-day to under-five-minute response spans benchmark close rates of 12 percent and 32 percent, which is 7.2 deals against 19.2, or roughly a 36,000 dollar monthly difference. That gap is correlational, so the realistic gain from response time alone is a fraction of it, but even a quarter of it is 9,000 dollars a month with no extra traffic or ad spend.

Why is our lead response so slow when nobody is ignoring leads?

It is almost always structural. The five common causes are notifications going to a shared inbox nobody watches, no named owner for inbound enquiries, forms that submit into a void with no CRM or task created, enquiries arriving outside business hours, and form notification emails silently failing after a hosting or DNS change. Submit your own contact form and time how long it takes to reach a human screen; a surprising number of businesses find the answer is never.

Does an automated acknowledgement email count as responding?

Not for the purposes of the research, which measures time to human contact. An immediate acknowledgement is still worth sending, because setting a clear expectation about who will reply and by when reduces the chance the buyer keeps shopping while they wait. Make it specific rather than generic, and do not let it become a substitute for measuring your real median time to a human response.

The takeaway

Two things are worth doing this week, and neither needs a budget. Submit your own enquiry form and time how long it takes for a human being to respond, because a meaningful share of businesses discover the notification never arrives at all. Then run the close-rate arithmetic on your own lead volume and deal value, so you know whether this is your biggest available win or a distraction from something else. If it is a win, the fix is boring and structural: a named owner for inbound, a notification somewhere that gets seen, a written target, and a weekly median to manage against. And when you quote the research to make the case internally, get the attribution right. The multipliers are MIT's, the 42-hour average is Harvard's and dates from 2011, and being the one person in the room who knows the difference is worth more than the statistic.