Speed to lead.
What the study found.
Someone fills in your form or messages your page, then keeps looking. Speed to lead is how long it takes your business to make a real reply. Many pages ranking for the phrase repeat the same large multipliers, often without a link to the study behind them. We went back to the original research document and read what it actually measured. It is older and narrower than the quotes suggest. This guide sets out what it found and what it did not, then gives you a checklist to time your own first reply this week.
Every figure about the research is cited to the 2007 study document or to Harvard Business Review, read on 14 October 2026. The HBR article sits behind a paywall, so we cite only the part we could read. The Spam Act point is general information, not legal advice. Updated 14 October 2026. 10-minute read. How we write these guides.
The short answer.
Speed to lead is the time between someone enquiring and your business making its first real reply. The best-known research is a 2007 study of web-form leads at six companies. It found the odds of reaching a lead by phone dropped 100 times when the first call came at thirty minutes instead of five[1]. It measured contact and qualification, not sales, so treat it as a strong signal, not a forecast.
What is speed to lead?
Speed to lead is a sales term for the gap between an enquiry arriving and the first real attempt to answer it. The clock starts when the person sends the form, the message or the email. It stops when someone from the business actually responds to what they asked. An automatic thank-you screen does not stop the clock, because nobody has answered anything yet.
It is easy never to measure it. The enquiry lands in an inbox, a Facebook page or a job system, and somebody gets to it when they can. Nobody sees that gap from the inside, which is why it is worth measuring. Why website enquiries never arrive covers the worse case, where the enquiry never reaches anyone at all.
What did the original study measure?
The two best-known multipliers, 100 times and 21 times, come from one document: the Lead Response Management study by InsideSales.com and Dr James Oldroyd, then at MIT's Sloan School of Management. It was presented at MarketingSherpa's Business-to-Business Demand Generation Summit on 16 October 2007.
| Question | What the study document says |
|---|---|
| The question it asked | When companies should call web-generated leads for the best contact and qualification rates[1] |
| The data | Three years of records from six companies that generate and respond to web leads: over fifteen thousand leads and over one hundred thousand call attempts[1] |
| Which leads | Leads captured through a web form and called at least once[1] |
| How they were answered | Phone calls by sales or lead qualification staff[1] |
| What counted as contact | A call that reached a live person and lasted a set length, which differed by company[1] |
| What counted as qualified | A lead willing to enter the sales process, in some cases by setting an appointment. Each company defined it its own way[1] |
| What it left out | Sales. The document says the study did not address close ratios[1] |
The copy we read is a research summary web page, not a journal paper. It does not name the six companies, say where they were or what they sold, or publish its calculations. It says the data came from the InsideSales.com system, software sold by one of its authors[1].
What did the study find?
Response time was the headline. The study found that the day of the week and the time of day made a statistically significant difference to contact and qualification, then said the effect of response time dwarfed them both[1].
- Within the first hour, the odds of a call reaching a lead fell by over 10 times, and the odds of qualifying a lead by over 6 times[1].
- Between five and ten minutes, the odds of contact fell 5 times, and the odds of qualifying fell 4 times[1].
- Calling at thirty minutes instead of five, the odds of contact dropped 100 times, and the odds of qualifying dropped 21 times[1].
- After 20 hours, the authors found that every extra call attempt actually hurt the chance of making contact to qualify a lead[1].
As a test, the study's authors asked a top lead provider for mortgage quotes. InsideSales.com's president received seven calls about that one enquiry. The first came 30 minutes later and the last came three days later. [1]
Read the list again with one word in mind: odds. The study reports how the odds of contact and qualification changed. It does not say a fast call books a job. A lead you ring in four minutes can still say no, and the document never measured how many leads went on to buy.
How far should you trust it?
It is useful evidence, and it is narrow. Before you build a decision on it, weigh these limits.
- Age. The findings were presented in October 2007[1], so the study is now nearly two decades old. The ways people enquire have changed since, and the study cannot tell you how that shifts the numbers.
- Channel. It timed phone calls to web-form leads[1]. It says nothing about texts, emails, chat or automated replies.
- Scale. Six companies, none of them named[1]. You cannot tell whether any of them looked like your business.
- Setting. It was presented at a business-to-business marketing summit[1], and nothing in it describes Australian customers or service trades.
- Outcome. It measured contact and qualification, and states that it did not address close ratios[1].
- Method. It reports how much the odds changed but does not publish the calculations, so the figures cannot be checked from the document itself.
James Oldroyd later wrote about online sales leads in Harvard Business Review with Kristina McElheran and David Elkington. Their article, The Short Life of Online Sales Leads, ran in March 2011, and its opening says their research shows most companies are not responding nearly fast enough to potential customers' online queries[2]. The rest sits behind HBR's paywall. We have not quoted figures from it that we could not read, and if you see one credited to that article, ask for the page it came from.
Figures that appear on other speed-to-lead pages but not in the study document are left out of this guide. A number with no link to the research behind it is a claim, not a finding.
Why does a fast reply matter?
The study does not explain why the odds fall so quickly, and we will not pretend it does. The practical reason is easy to see from the customer's side. A person who has just asked is still near their phone and still has the problem in mind. An hour later they may already be talking to someone else.
The study's own test shows that second risk. One mortgage enquiry produced seven calls over three days, which the authors took as a sign the lead had gone to several businesses[1]. When one enquiry reaches several businesses, the first to reply has the conversation to itself for a while.
You may not buy leads from a lead seller, but someone who fills in your form may still have asked other businesses the same question. That is the honest case for replying first. It buys a head start, and the study cannot tell you how big a head start it is for a business like yours.
How do you time your first reply?
Run this from a phone and an email address the business does not recognise. Most steps take a few minutes, and the result is your own number instead of somebody else's multiplier.
List every door
Write down every way an enquiry can reach you: the website form, social media messages, email and the phone, plus any lead ads you run.
Send a test through each one
Use a name you will recognise later, and note the exact minute you sent it.
Note when a person sees it
Not when it lands in a system. When a named person gets an alert they will actually look at. If nobody is alerted, write that down. It is the first thing to fix.
Note the first real reply
A reply that answers or asks about the enquiry, from a person or from wording the owner approved. An automatic thank-you page does not count.
Repeat after hours
Send one test on a weeknight and one on a weekend. Enquiries do not keep office hours, so check whether anyone is rostered to look then.
Put the times side by side
For each door, record when the test was sent, seen and answered. The longest gap is your real speed to lead, whatever the average says.
Find the cause of the longest gap
Look for a missing alert, an alert going to someone on the tools all day, or an inbox nobody owns.
Set a target you can keep
Pick a reply time for each door and each part of the day that the business can actually meet. Write down who owns it, and test again next month.
If you run Facebook lead ads, our guide to Meta lead ads shows how to send Meta a test lead and time where it lands.
How do you shorten it?
Start with the gap between the enquiry arriving and a person knowing about it, because nothing else can happen until someone knows.
Send alerts to a person
Route every door to a named person's phone, with a backup for when they are busy. A shared inbox nobody owns is where enquiries sit and wait.
Reply the way they asked
The study timed phone calls[1]. If someone enquired by message, a quick reply in the same thread may suit them better than a call. Ask which they prefer.
Write the first reply now
Agree the wording for business hours and after hours in advance, so nobody composes it in a hurry. Keep it to the enquiry and name the business.
Ask what you need to quote
A first reply that asks for the suburb and the job means the next call is about when, not what.
Cover the hours nobody watches
Decide who replies at night and on weekends. Or decide plainly that the reply comes next morning, and say so in the acknowledgement.
On wording, keep the first text or email factual. Under the Spam Act, a message of no more than factual information plus details such as who sent it is treated differently from one with a purpose of offering or promoting goods or services[3]. Every commercial message must still identify who authorised it[3].
Phone calls leak the same way when nobody answers. Missed-call text-back covers one way to keep that conversation open, and what the Spam Act looks at in the text.
Where does AI automation fit?
AI automation is aimed at the first gap: knowing an enquiry exists, at any hour, and acknowledging it. On our automation page, the enquiry agent that answers website questions at any hour is live. The agents built to work by text message are not, because our SMS service is not live yet, and the page says so plainly. Anything an agent sends in a client's name is drafted for their approval by default.
One caution, in the research's own terms. The 2007 study timed calls made by people[1]. It says nothing about whether an instant automated reply earns the same head start, so we will not claim it does. An agent can hold the conversation open and collect the details. A person still makes the call that turns an enquiry into a job.
If you have timed your first reply and do not like the number, tell us where your enquiries come in. One of us reads it and gives you a straight answer about what fits.
Questions and answers.
What is a good speed to lead?
There is no official standard. The best-known study found the odds of contacting a web lead by phone fell by over 10 times within the first hour, and dropped 100 times between a first call at five minutes and one at thirty[1]. A practical target is the fastest reply your business can keep every time, for each door and each part of the day. Time your own first reply before you set one.
Is the 21 times figure real?
It comes from the 2007 Lead Response Management study, which found the odds of qualifying a web lead dropped 21 times when the first call came at thirty minutes instead of five[1]. It is often quoted as if it were about sales. The study measured qualification, not sales, and says it did not address close ratios[1].
Where does the 100 times figure come from?
The same 2007 study. It found the odds of making contact with a web lead by phone dropped 100 times when the first call came at thirty minutes instead of five[1]. Contact meant reaching a live person on a call of a set length, not making a sale[1]. The study drew on six unnamed companies and is now nearly two decades old.
Does an instant auto reply count?
Not in the original research. The study timed phone calls from sales or lead qualification staff to web-form leads[1], so it cannot say whether an automatic acknowledgement has the same effect. An auto reply that tells the person who will be in touch, and when, is still worth having. Just do not count it as your first real reply.
Does the Spam Act apply to my first reply?
It can. The Act's consent and unsubscribe rules apply to commercial electronic messages, judged by purposes such as offering to supply or promoting goods or services[3]. A message of no more than factual information plus sender details is treated differently, though it must still identify who authorised it[3]. Keep the first reply to the enquiry. This is general information, not legal advice.
Sources.
- The InsideSales.com/MIT Lead Response Management Study (Dr James Oldroyd and InsideSales.com, presented 16 October 2007; copy of the study page as printed from leadresponsemanagement.com on 17 July 2008), InsideSales.com and Dr James Oldroyd (copy hosted by Mortech). Read 14 October 2026.
- The Short Life of Online Sales Leads, by James B. Oldroyd, Kristina McElheran and David Elkington, March 2011 (opening paragraph only; the full article is paywalled), Harvard Business Review. Read 14 October 2026.
- Spam Act 2003, Compilation No. 10 (10 March 2016): ss 6, 16, 17, 18 and Schedule 1 clause 2, Federal Register of Legislation. Read 14 October 2026.
Find out how long your reply takes.
Our automation page shows which AI agents are live today and which wait on our SMS service. The contact form reaches one of us directly.