Every sales floor has heard it: call a web lead within five minutes or watch it die. The rule is quoted like physics. It is not physics. It is a chain of four studies, each with a specific sample, a specific funder, and a specific thing it measured. This autopsy grades each one, claim by claim, with our evidence labels: Observed (we read the primary document ourselves), Corroborated (confirmed by at least one independent source), Derived (computed from disclosed inputs), and Inferred (our judgment from incomplete evidence). The point is not that speed is irrelevant. The point is that you should know exactly what a number can and cannot carry before you build a staffing plan on it.
What did the 2007 lead response study actually measure?
The original artifact is the InsideSales.com/MIT Lead Response Management Study, presented October 16, 2007, authored by Dave Elkington, chief executive of InsideSales.com, and James Oldroyd, then at MIT's Sloan School [Observed]. The data: three years of call records from six companies, covering more than 15,000 leads and more than 100,000 call attempts, all drawn from InsideSales.com's own dialer platform [Observed].
The headline findings are about odds of contact and odds of qualification, not about closing. The study reports that the odds of contacting a lead drop 100 times when the first call happens at 30 minutes instead of 5 minutes, and the odds of qualifying drop 21 times [Observed]. The study itself does not measure close ratios or revenue [Observed]. The funder sells software whose value proposition is faster dialing [Observed]. That does not make the numbers false, but it makes them a sales asset first and a research finding second [Inferred].
What it can support: within these six companies, earlier calls reached more people. What it cannot support: "calling in five minutes makes you 100 times more likely to close," which is how the number is routinely quoted today, for example on vendor statistics pages that cite 100x and 21x with no attribution at all (accessed November 9, 2026) [Observed].
What did the 2011 Harvard Business Review audit find?
The strongest document in the family is "The Short Life of Online Sales Leads," published in Harvard Business Review in 2011 by James Oldroyd, Kristina McElheran, and David Elkington, archived at BYU ScholarsArchive [Observed]. The researchers sent a web-generated test lead to each of 2,241 US companies and timed the response [Corroborated, via the primary archive and the independent reconstruction at CIENCE's The 42-Hour Problem, accessed November 9, 2026].
The distribution: 37% of companies responded within an hour, 16% within 1 to 24 hours, 24% took longer than a day, and 23% never responded at all [Corroborated]. The much-quoted 42-hour average response time is a mean computed only among companies that responded within 30 days, so it excludes the 23% who never answered; the real picture is worse than the famous number [Derived]. A companion analysis of 1.25 million leads across 42 companies found that firms attempting contact within an hour were about 7 times as likely to qualify the lead as firms that waited a second hour, and more than 60 times as likely as firms waiting 24 hours or more [Corroborated]. Note what is missing: the paper never mentions five minutes, and it reports likelihood ratios without a baseline qualification rate, so the figures cannot be converted into revenue [Corroborated]. One author ran the company that sells response software [Observed].
Where did the 391% figure come from, and what is its baseline?
In 2012, lead-management vendor Leads360 (later Velocify) published "The Ultimate Contact Strategy," analyzed in trade press at the time, drawing on nearly 3.5 million leads. Chief Marketer covered it on November 27, 2012, under a URL promising a 391% conversion lift from calling within one minute [Observed]. Here is the problem: the surviving coverage does not state the baseline the 391% improvement is measured against, and does not define conversion [Observed]. A 391% lift on an undisclosed base is an applause line, not a planning input [Inferred]. The funder, again, sold speed-to-lead software [Observed].
What did the VanillaSoft and Telfer study find that contradicts the rule?
The most interesting entry is the lead conversion study by VanillaSoft with the Telfer School of Management at the University of Ottawa, using roughly 130 million interactions with about 45 million leads, drawn from more than 220 million call interactions at 550 companies between 2005 and 2017 [Observed]. Its reported finding: win outcomes were three times more likely when first contact landed between 10 and 60 minutes after the inquiry than in the first 10 minutes or after one hour [Observed]. The authors say they do not yet know why, and the document explicitly labels itself preliminary, with a full study promised for fall 2018 that we could not confirm was ever published [Observed]. So the one study in this family that examined wins rather than contacts points away from the five-minute rule, and it is preliminary, vendor-funded, and unreplicated [Derived]. Treat it as a caution flag on the rule, not a new rule [Inferred].
How bad is the gap between what companies believe and what they do?
InsideSales.com's own later audit work, reported by ThinkAdvisor in May 2013, tested more than 700 member companies of the AA-ISP: about 25% believed they responded within five minutes, 5% actually did, the most common number of actual contact attempts was zero, and while over 90% believed their first touch was a phone call, 67% actually led with email [Observed]. Whatever you think of the odds ratios, the measured gap between self-report and behavior is the most practically useful finding in the entire literature [Inferred].
What is the verdict on each study?
| Study | Year | Sample | Funder | What it measured | What it cannot support | Verdict |
|---|---|---|---|---|---|---|
| InsideSales/MIT LRM study | 2007 | 6 companies, 15,000+ leads, 100,000+ calls | InsideSales.com | Odds of contact and qualification by response time | Close rates, revenue, anything beyond 6 firms on one dialer | Real but narrow. Quote only as contact odds. |
| HBR, The Short Life of Online Sales Leads | 2011 | 2,241 audited companies; companion set of 1.25M leads at 42 firms | Academic authors plus an InsideSales executive | Response latency; qualification likelihood ratios | A five-minute threshold; revenue effects | Best in class for latency. Silent on five minutes. |
| Leads360, via Chief Marketer | 2012 | ~3.5M leads | Leads360 (Velocify) | A conversion lift with no stated baseline | Any quantitative planning claim | Unusable as published. No baseline, no definition. |
| VanillaSoft/Telfer | 2017-2018 | ~45M leads, 550 companies | VanillaSoft | Win likelihood by first-response window | A final conclusion; it is self-labeled preliminary | Intriguing contradiction. Awaiting the full study that never surfaced. |
| InsideSales audit, via ThinkAdvisor | 2013 | 700+ AA-ISP member companies | InsideSales.com | Self-report versus tested behavior | Industry-wide averages | The useful one. Perception and behavior diverge badly. |
So should you still respond fast?
Yes, and here is the honest chain of reasoning. Leads go cold within hours rather than days in every dataset anyone has published, including the audit showing 23% of companies never respond at all [Corroborated]. Whether the cliff sits at minute 5 or minute 45 is genuinely unsettled, and the only wins-based study says the first ten minutes may not be magic [Observed]. What is settled is that most organizations are nowhere near any of these thresholds, respond far slower than they believe, and frequently never respond [Corroborated]. Fix the never-respond problem before you litigate the minutes [Inferred]. Measure your own response-time distribution from your own CRM timestamps, because your number is Observed and every number above is somebody else's.