A lead form submits at 9:14pm on a Thursday. It lands in a shared inbox that nobody owns overnight. A rep opens it at 9:40 the next morning, calls twice on Friday, gets voicemail, and marks the record dead on Monday. Nothing in that chain was anyone's fault, and the lead still cost you money. Most speed to lead statistics you will see quoted at that point come from a single study, and it is older than most of the teams quoting it. Knowing which study, and what it actually measured, changes what you do about Thursday night.
Where the speed to lead statistics actually come from
Two figures carry the entire category. Call a web lead within five minutes and you are roughly 100 times more likely to make contact, and roughly 21 times more likely to qualify that lead, than if you wait 30 minutes. Both numbers come from the same place: a 2007 analysis run by InsideSales.com with a researcher from MIT, built on about three years of data covering more than 15,000 leads and over 100,000 call attempts.
Saying that out loud makes the claim stronger rather than weaker. It is one dataset, drawn from one vendor's customer base, which means inside sales teams already running dialers against inbound web leads. The mechanism it measured, how likely you are to catch a human on the phone shortly after they raised their hand, has aged better than the calendar suggests. The framing around it has not.
Two limits matter more than the age. The qualification number is a qualification number: the study measured odds of contact and odds of qualifying, and its own methodology states it did not address close ratios. Anyone selling you "21 times more likely to close" has swapped one word and inflated the claim by an unknown amount. The second limit is provenance on the companion stats. The widely repeated line about buyers purchasing from whoever responds first has no traceable original study behind it. The familiar averages, 47 hours to first response and single-digit shares of companies replying inside five minutes, trace to an audit published inside a paywalled 2011 article and are not readable on the live page today. None of that belongs in a board deck.
The honest version is narrower and more useful. There is decent evidence that fast dials raise contact and qualification rates on inbound web leads. There is no current public benchmark for what an average company's response time is. There is no credible revenue multiplier at all.
Slow follow-up raises your cost per lead before the sales call
Here is the part most operators miss. Response time also shows up upstream, inside the ad auction.
Google's Local Services documentation tells advertisers it is highly recommended that you respond to leads as soon as they come in, and that a consistently fast response time could improve your ad's ranking on search and increase your ability to receive leads. The same page notes that if you have 2 or more message leads in the last 90 days, estimated messaging response times may appear in your ads, ranging from a few minutes to one day.
Read that as a media buyer rather than a sales manager. Your follow-up speed can move ranking, and on that surface it can be printed on the ad itself as a label the buyer reads before they ever contact you. A slow team pays on both sides of the transaction: weaker contact rates on the leads it buys, and weaker economics on buying them. That is a different conversation from the one most teams are having, because it is owned by whoever runs media, not by whoever runs the phones.
Audit the routing before you audit the reps
Both major platforms locate the fix in the plumbing. Google's lead form guidance is blunt about it: for rapid follow-ups, it recommends setting up a webhook integration or an integration with Zapier to route leads directly into your CRM system. Meta's advertiser guidance points the same direction, steering advertisers away from manually downloading leads as a CSV out of Ads Manager and toward a live CRM connection, on the grounds that the manual route slows follow-up down.
When response time is bad, the instinct is to hold a meeting about urgency. The delay is usually sitting in a handoff nobody has looked at since the campaign launched.
Five questions worth answering this week
- Where does a lead physically land the second it is submitted, source by source? Name the system, not the person.
- How long does each hop take: form to CRM, CRM to assignment, assignment to a notification on somebody's phone?
- Who owns the hours between 6pm and 8am, and what happens when that person is asleep? A queue is not an owner.
- When are your lead campaigns scheduled to run? Meta tells advertisers to run lead campaigns during the hours their organisation can actually connect. Leads landing at 9pm and getting called on Tuesday is a media buying decision the owner controls, not a staffing problem.
- What is your real first-touch time, measured from the form timestamp rather than from the moment a rep opened the record?
Then set a standard and hold the team to it. Five minutes on inbound forms during working hours, with an automated acknowledgement outside them, is the operating standard we recommend. That is our number, chosen because it is short enough to force the routing to be automatic. No study in this post prescribes it.
A model you can run on your own numbers
Assumptions first, and every one of them is yours to replace. Assume $20,000 a month in paid lead generation at a $50 cost per lead, which is 400 leads. Assume your team currently reaches 30% of them, so 120 contacted. Assume 25% of contacted leads book a qualified appointment, so 30 appointments. Divide spend by appointments and you are at roughly $667 per qualified appointment.
Now change one input and hold everything else still. Keep spend at $20,000, CPL at $50 and the contact-to-appointment rate at 25%, and move contact rate from 30% to 40%. That is 160 contacted and 40 appointments, or $500 per qualified appointment. A ten point move in contact rate took 25% out of your cost per appointment without touching creative or targeting.
This is a model, not a result. Nothing here tells you that tightening response time moves your contact rate ten points, and on a low-volume, high-consideration offer it may move it far less. The reason to run the arithmetic is to size the prize before you spend a quarter rebuilding routing. If it produces a rounding error on your numbers, go spend that quarter on the offer instead. If the media looks fine to you and the conversion machinery behind it was never really built, that is a different diagnosis, and it is the one taken apart in why lead-gen campaigns stall at $10k a month.
After-hours leads and the automated first touch
The 2025 and 2026 answer to unworked and after-hours leads is an automated first touch, and the loudest evidence for it comes from vendors with an obvious interest in selling you one. Take it as direction of travel rather than proof. Salesforce, writing about its own internal deployment, reports that in four months, agents contacted 130,000 leads and created 3,200 opportunities working leads that were previously untouched. On those two figures, that is an opportunity rate of about 2.5% on a pool that was otherwise generating nothing at all. Meta now offers lead nurturing inside Messenger, Instagram and WhatsApp, which puts the same idea in the platform where the lead already is.
One caution on how these get sold. Twilio's 2025 consumer research found that 88% of consumers are more likely to buy when engagement is personalized in real time, while only 44% of brands say they are executing at that level, and that 71% of consumers abandon purchases when experiences fall flat. Those numbers describe relevance and real-time personalization. They are not a measurement of how fast buyers expect a reply, and you will see them cited as though they were.
If you want one thing to do with this, measure your real first-touch time from the form timestamp for two weeks, split by source and by hour of day. Most teams discover their problem is concentrated in one source and one six-hour window, which makes it a routing and dayparting fix rather than a hiring decision. The speed numbers everyone quotes are older and narrower than the confidence around them suggests, and the underlying behaviour they describe is still the cheapest thing in your funnel to improve. If you would rather have someone own that end to end alongside the media, that is the work we do, and you can apply here.