Only your own numbers can answer that. Enter your monthly lead volume, deal value, and the conversion rate you achieve at your current response time, then either enter your measured rate at five minutes or pick an explicit what-if uplift. The tool prices the gap. It publishes no response-time multiplier, because we cannot verify one.
This model prices the gap between two conversion rates at two response times. You supply both rates, or you supply one and explicitly choose a what-if uplift. It is a model of the numbers you entered, and it deliberately ships no default response-time multiplier of its own.
Because we could not verify one well enough to publish it, and publishing an unverifiable number as though it were established is exactly the practice this site has been cleaned of. There are famous figures in this category that circulate widely, quoted at second and third hand, usually without a study version, a sample description, an industry, or a date. Some of them may well be directionally right. None of them are things we can stand behind as a published constant that then propagates into your business case and out of it into your board pack. So the tool asks you for the number instead, in one of two honest modes. If you have measured your conversion at different response times, enter both rates and the model prices the real gap. If you have not, choose an explicit what-if uplift, which the tool labels on screen as an assumption you selected rather than a fact it supplied. This makes the tool slightly harder to use and considerably more defensible. If you take the output to a decision-maker, every input in it is either measured by you or chosen by you, and the share link records which.
The mechanism is simple and does not require a statistic to explain. A person who has just submitted a form or called a number is, at that moment, actively thinking about the problem. They are at their desk, the tab is open, the context is loaded. An hour later they are in a meeting. A day later they have contacted two competitors and one of them has already booked a visit. There is also a race component. In most competitive categories the buyer contacts several suppliers, and the first meaningful response often frames the entire comparison: the first supplier to reply sets the terms the others get measured against, and frequently books the appointment before the others have replied at all. Neither of those effects needs a multiplier to be believable, and neither tells you how large the effect is in your business. That is what the measurement in the next section is for.
Pull the last ninety days of inbound leads from your CRM with two fields: time from lead creation to first outbound contact attempt, and whether the lead converted. Bucket by response time: under five minutes, five to thirty minutes, thirty to sixty, one to four hours, four to twenty-four hours, and over a day. Compute conversion within each bucket. That table is the honest version of every statistic in this category, and it is specific to your business rather than to an unnamed aggregate. Read it carefully before believing it, because the obvious confound is real. Leads that get answered in under five minutes are often the ones that arrive during business hours from people who called rather than emailed, and those may convert better for reasons that have nothing to do with speed. If your fast bucket is mostly phone and your slow bucket is mostly form fills, you are comparing channels, not response times. The clean version is a test rather than an analysis: for one month, respond to every second lead as fast as you can and handle the rest normally. That randomises the confounds away, and after a month you have a number nobody can argue with.
For inbound calls, yes, and that is the easier half. For form fills, it depends entirely on whether responding is somebody's actual job during the hours leads arrive. The three structural blockers are consistent. Leads arrive outside working hours, and nobody is there. Leads arrive during working hours but the person who responds is doing something else that cannot be interrupted. And leads arrive into a system nobody is watching, so the clock starts long before anyone knows there is a lead. Each has a different fix and only the third is a software problem. Notification routing solves the third cheaply. The second is a staffing and priority question. The first is where always-on answering earns its cost, because it is the only one where nothing else works. Before buying anything, run this model with your after-hours lead share. If most of your slow responses are out of hours, the fix is narrow. If they are spread through the working day, no answering system will fix a prioritisation problem.
Invert the question. Instead of asking how much faster responses are worth, ask how small an improvement would still cover the cost. The tool computes that break-even directly: your monthly cost of change divided by your current monthly revenue from these leads, expressed as a percentage uplift. If the answer is that you need a 2% relative improvement in conversion to break even, that is a very different decision from needing 40%, and it is a decision you can make without a published multiplier at all. This is usually the number worth taking to whoever approves the spend, because it removes the argument about the size of the effect and replaces it with a question about plausibility. Almost nobody will argue that responding in five minutes rather than an hour is worth less than a 2% relative improvement. Plenty of people will argue about whether it is worth 40%.
Because the widely-quoted figures circulate without a checkable version, sample or date, and this site does not republish statistics it cannot stand behind. You enter your own measured rates, or you choose an explicit what-if that the tool labels as your assumption.
Export ninety days of leads with created-at and first-contact-at timestamps plus a converted flag, bucket by response time, and compute conversion in each bucket. Check the channel mix per bucket before believing the difference.
A first human or agent contact attempt: a call, a text, a personal reply. An automated acknowledgement email is not a response, and counting it as one is the most common way a reported median gets flattered.
Median. Response-time distributions have long tails, and a handful of leads answered a week late will make a mean meaningless.