How the estimate is built
- Add up your monthly leads from the four sources.
- Today: leads times the share that book gives appointments. Appointments times the share that close gives deals or jobs. Deals times the average value gives revenue. A year is twelve months.
- Speed effect: we assume the share of leads that book falls by a fixed percentage each time the wait for a first reply doubles. Replies faster than 5 minutes count as 5, because 5 minutes is the fastest interval the studies below measure. The multiplier is the appointment rate at your target time divided by the rate at your current time. You can replace it with your own number.
- At the target reply time, the share that book is today's share times the multiplier, held at 100% at most. The close rate and the value of a deal stay as you entered them.
- The smaller and larger rows scale the lift (the multiplier minus 1) by one half and by two, with the same 100% limit.
The share of the 5-minute appointment rate that is kept when the first reply takes a given time is (1 − d) ^ log2(minutes ÷ 5), where d is the percentage lost per doubling.
Worked example
100 leads a month, 10% book, 25% of appointments close, $10,000 a deal, first reply at 4 hours (240 minutes) moving to 5 minutes, 5% lost per doubling. 240 minutes is 48 times 5, about 5.6 doublings, so a 4-hour reply keeps about 75% of the 5-minute appointment rate (0.95 to the power 5.6). The multiplier is 1 divided by 0.751, or 1.33. The share that book goes from 10% to 13.3%: appointments from 10.0 to 13.3 a month, deals from 2.5 to 3.3, and revenue from $25,000 to about $33,293 a month. That is about $8,293 more a month, or $99,516 a year. The calculator's arithmetic is checked against figures like these by a Node script in the site's code repository (scripts/verify-speed-to-lead.mjs).
The default is an assumption, and a cautious one
None of the studies below measures what reply time does to booked appointments for a real estate or contractor lead. So the size of the effect is yours to set. We start it at 5% lost per doubling: a reply within the hour keeps about 83% of the 5-minute appointment rate, and a reply the next day about 66%. We chose that to understate rather than overstate. It is far below the headline figures in most of the studies in the next section, which measure other things in other settings. It is not derived from any of them, and we cannot show it is right for your market.
What the research says, and where each study stops
Measured: three years of lead and call data from six companies: over 15,000 leads and over 100,000 call attempts. Outcomes were whether the first call reached a person (a contact) and whether the lead became qualified, meaning willing to enter the sales process, which at some companies meant setting an appointment.
Result: the odds of contacting a lead called within 5 minutes versus 30 minutes were 100 times higher, and the odds of qualifying it 21 times higher.
Limits: these are odds for a first call, not appointment or close rates (the report says the study did not address close ratios). The data came from the vendor's own system, and the report does not name the companies or industries. The widely repeated "21 times more likely to convert" is this qualification figure restated, and the study did not measure conversion. A survey of 495 companies in the same report found much smaller effects: each step slower to the first call attempt (its steps run from 0 to 5 minutes up to a week or more) went with a 4.3% drop in the share of leads qualified and a drop of nearly 2% in close rates, and the report does not say whether those are percentage points or relative changes.
Measured: an audit in which 2,241 US companies each received a web-generated test lead and were timed, plus a separate study of 1.25 million leads received by 29 consumer-facing and 13 business-to-business companies. There, qualifying a lead meant having a meaningful conversation with a key decision maker.
Result: 37% of the audited companies replied within an hour, 16% within one to 24 hours, 24% took longer than 24 hours and 23% never replied; the average reply time among those that answered within 30 days was 42 hours. Companies that tried to contact a lead within an hour were nearly 7 times as likely to qualify it as those that tried an hour later, and more than 60 times as likely as those that waited 24 hours or longer.
Limits: qualification here is a conversation, not a booked appointment or a sale. The text we read does not report results by industry, so nothing in it is specific to real estate or contractors. A co-author was the CEO of InsideSales.com when it was published, the vendor behind the 2007 and 2021 studies on this page. HBR's page shows only the introduction; we confirmed these figures against a full-text copy of the article.
Measured: three years of data from over 400 companies: 5.7 million marketing leads and 55 million sales activities.
Result: the infographic states that conversion rates are 8 times higher within 5 minutes than at 6 minutes or more.
Limits: a vendor study of inbound marketing leads, framed around business-to-business sales. The infographic does not define "conversion", and the page we opened links only to the infographic, not to a full method.
Measured: more than 11.6 million message exchanges between buyers and sellers on Fiverr, a marketplace where employers contact freelancers for services such as graphic design, copy editing and web development, before deciding whether to hire. Also experiments: three main ones with more than 3,600 participants and five supplemental ones with another 5,000.
Result: in the Fiverr data, a one-hour delay was associated with a 46% lower likelihood of being hired and a full-day delay with a roughly 90% lower likelihood. The effect persisted when participants could see other information, including ratings and the content of the reply. In the experiments, faster replies lost their appeal when recipients believed they were automatic or AI-generated.
Limits: freelance services, not property or home services, and observational marketplace data. The journal page was not open to us, so we confirmed these figures from the authors' universities' announcements: UC San Diego and Vanderbilt Owen.
How fast businesses actually reply: two 2026 audits
Workato, March 2026: demo requests filled in at 114 business-to-business companies. None called within 5 minutes and one sent a personalised email within 5 minutes; the average email reply took 11 hours 54 minutes and the average call 14 hours 29 minutes, and 31% called at all. Roof AI, 2026: website inquiries sent to the top 74 US real estate brokerages. 41% never replied and 9% replied within 5 minutes.
Limits: both measure reply speed only, not what it did to bookings or sales, and both publishers sell products that speed up replies. The Roof AI page gives little detail on its method. We use them only as context for why measuring your own reply time is worth doing, not in the math.
Left out on purpose. Some figures circulate widely on this topic, for example "78% of buyers choose the first agent who responds" and a "391%" conversion lift for calling within a minute. We could not trace them to a primary source we could open, so they do not appear on this page.
Check it against your own numbers
Export the last 90 days of leads from your CRM or portal, with the time of the first reply and whether an appointment was booked. Group them by first-reply time (under 5 minutes, up to an hour, up to a day, longer) and compare booking rates. Then set the assumption above to match, or enter your own multiplier. Leads that get fast replies may differ in quality or source from the rest, so treat a gap as a strong clue rather than proof.