THE SHORT ANSWER
The widely quoted speed-to-lead research is genuine but comes from business-to-business software sales in the early 2010s, so the dramatic multipliers travel badly to home improvement. What survives translation is the direction and the shape: contact probability falls sharply in the first hour, hardest inside the first ten minutes. The sensible reading is a modelled band for your own trade plus a split test you run yourself, not a borrowed statistic.
Two pieces of research underpin almost every claim made about response time. One is a 2011 Harvard Business Review analysis of thousands of enquiries, which reported that firms responding within an hour were far more likely to qualify a prospect than those responding an hour later. The other is the Lead Response Management work, which found contact odds collapsing between the five-minute and thirty-minute marks.
Both are real studies. Both are also fifteen years old, drawn largely from software and financial services, and repeated so often that the numbers have drifted in the retelling. Anyone selling you a response-time product will quote the biggest multiplier they can find. It is worth separating what the evidence supports from what has been decorated onto it.
The numbers, at a glance
What the original research covered: business-to-business web enquiries, mostly software and financial services, sampled in 2007 to 2011
The finding that travels well: contact probability declines steeply and non-linearly, with the sharpest drop inside the first ten minutes
The finding that travels badly: the headline multipliers, because qualification in a software pipeline is not the same event as booking a survey
Modelled home-improvement equivalent: a five-minute response reaches roughly twice as many homeowners on first contact as a sixty-minute response, not seven times as many
Why the multipliers shrink when you move to home improvement
Three structural differences compress the effect. Home-improvement enquiries carry a physical job attached to a specific address, so intent is more durable than an interest in a software trial. The buyer usually expects to speak to two or three trades, so being second is not fatal in the way it can be in a single-vendor purchase. And the audience answers mobile phones at a much higher rate than office-hours business buyers ever did.
The consequence is that the curve keeps its shape and loses its steepness. Fast response still wins materially more conversations, but a competent business that calls in twenty minutes is not competing at a seven-to-one disadvantage against one that calls in three. Selling the extreme version of the statistic tends to backfire, because contractors test it, see a smaller effect, and then discount the whole idea.
What is genuinely measured versus what is modelled
Measured, robustly: that contact rates fall with elapsed time, across every sector anyone has studied.
Measured, in specific sectors: that the decline is steepest in the first thirty minutes rather than spread evenly across the day.
Modelled, not measured: every band on this hub expressed as a percentage for a named trade, including ours.
Neither, and widely repeated anyway: precise claims about a single minute of delay costing a fixed percentage of revenue.
This distinction matters commercially. A modelled band is a planning tool you can argue with. A borrowed percentage presented as fact is something a supplier uses to make you feel bad about a process problem they do not have to solve.
Running the test yourself, which takes about six weeks
Randomly split incoming enquiries into two response groups. Group A gets called immediately. Group B gets called at a fixed delay, say ninety minutes, with everything else held constant: same script, same caller, same number of attempts, same follow-up. Log contact rate on first attempt, appointments booked, and jobs signed.
Sixty enquiries per group is enough to see a real difference in contact rate, because contact is a high-frequency event. Jobs signed will not reach significance at that volume and you should not pretend otherwise. Read the contact rate as evidence, treat the revenue difference as directional, and re-run it annually rather than trusting a one-off.
The finding most contractors miss entirely
In practice the difference between businesses is rarely five minutes against sixty. It is five minutes against never. A substantial share of purchased enquiries in this sector receive one unanswered call and no second attempt, which means the argument about optimal response time is being had by firms who have already solved a problem their competitors have not noticed.
If you are choosing where to spend attention, get every lead contacted at least four times before you spend anything getting the first attempt from twelve minutes down to four. The first fix is worth several times the second, and it costs nothing but a rule.
Reading any speed-to-lead claim critically
Ask which sector and which year the underlying sample came from.
Ask what event was counted: a connected call, a qualified prospect or a signed job.
Discount any multiplier quoted without a sample size attached to it.
Replace the borrowed figure with a modelled band for your own trade and job value.
Run a two-group split on your own enquiries before changing how you staff evenings.
How these figures were built
These figures are a benchmark model, not a survey. They combine an industry base range observed across Western European home-improvement campaigns with a country multiplier for local auction pressure. Treat them as a band to negotiate against, not a quote.
Want leads like this in your pipeline?
Flock runs the campaigns, screens the enquiries and hands you only the ones that match your service area, job size and capacity. You pay per lead, not per month.
Book a 15-minute fit check | See lead package pricing
Related answers
Frequently asked questions
So is speed to lead overrated?
It is over-quantified rather than overrated. The direction is well supported and the practical advice holds. What is unreliable is the specific multiplier, which comes from a different sector and a different decade and gets inflated each time it is repeated in a sales deck.
How much difference does five versus twenty minutes really make?
Modelled for home improvement, a modest one: perhaps five to ten percentage points of first-attempt contact. The large gaps appear between minutes and hours, and the largest of all between one attempt and several. Optimise in that order, because twenty minutes is not a failure and two days is.
Do these findings apply to enquiries I generate myself?
Yes, and the mechanism is identical, because it is about the homeowner rather than the source. The one difference is that self-generated enquiries often arrive with weaker intent signals, so the value of reaching them while the page is still open is if anything higher.
What single number should I track?
Median minutes from enquiry timestamp to first connected call, reviewed monthly. Medians resist the distortion caused by a handful of leads that sat for three days, which is exactly the distortion an average will hide from you. Pair it with the share of enquiries that were reached at all.
NEED A CLEARER PLAN?
Let’s turn your next move into momentum.
Talk to us →