Everyone quotes the same follow-up statistics. Almost none of them are real.

You’ve been shown this one. Eighty percent of sales require five or more follow-ups.

It comes from a survey taken in 1942 by the Long Island chapter of a sales association, with a sample of fewer than forty people. The association’s successor, Sales & Marketing Executives International, says so on its own website, including the part about the sample size.

Nineteen forty-two. Before the interstate highway system. Before anybody had a telephone answering machine to avoid.

That number has been in sales training decks for more than eighty years, and it’s still in the pitch deck of whoever sold you follow-up software last month. Once you know where it came from, you start looking at the rest of them differently. So we went and checked the rest of them.

The short version

• “80% of sales require five or more follow-ups” comes from a 1942 survey with a sample under 40 people.
• “Seven messages gets a 60% response rate” misreads Hatch’s own report, where 60% was the average and seven messages described the single best campaign.
• “Only 37% of estimates close on the first visit” is a vendor figure with no published method.
• The SMS-versus-email response figures, and two other widely repeated numbers, trace to nothing we could find.
• The one number with a named research firm and a stated sample comes from a 2025 ServiceTitan survey.

The numbers you’ve been shown, and where they came from

“Seven messages gets you a 60 percent response rate.” This one comes from Hatch, which analyzed 163,212 estimate follow-up campaigns run by HVAC businesses on its platform over six months. Real data, decent sample. But read what the report actually says. The average response rate across all those campaigns was 60 percent. Seven messages described the single best-performing campaign, which hit 90 percent. Nobody claimed one caused the other. Somewhere between the report and the sales deck, an average and an outlier got welded together. Hatch sells follow-up software, which is worth knowing, but in this case the misquote isn’t theirs.

“Only 37 percent of estimates close on the first visit.” ServiceTitan’s figure, from a webinar recap on its blog saying its data shows only 37 percent convert immediately. It may well be accurate. There’s no way to tell, because the page states no sample size, no time period, no trades, and no definition of what counts as converting. ServiceTitan sells the follow-up tools the number supports.

“Sixty percent of sales happen after the fourth contact.” We couldn’t trace this to any original study at all. It circulates widely, and we found no source for it.

“SMS gets a 45 percent response rate versus 6 percent for email.” Usually credited to Gartner. We couldn’t find the Gartner research it’s supposed to come from, and the pages repeating it, mostly companies that sell texting, don’t link to one.

“Forty-four percent of salespeople give up after one follow-up.” Also untraceable. So is the variant that says 92 percent quit after four rejections.

The only figure in this whole area with a named research firm and a stated sample comes from ServiceTitan’s 2025 Residential Services Report, which found that 47 percent of contractors above $10 million in revenue say following up on estimates produces 11 to 15 percent of their income. Thrive Analytics ran the survey for ServiceTitan, more than a thousand residential contractors responded, and it was released in March 2025. Those are owners estimating their own numbers rather than measured revenue, and ServiceTitan paid for it. That’s still the best-documented figure in the category, which tells you something about the category.

We found the same thing when we went looking for a customer retention benchmark for the trades. The numbers everybody repeats mostly don’t lead anywhere.

So what do contractors actually say?

They ask each other, because no trustworthy benchmark exists.

In a thread on Mike Holt’s electrician forum from February 2006, a contractor asked whether their results were normal. They were winning one or two jobs out of every ten estimates, and they were following up hard to get there: five days, then ten, then ten, then twenty, then twenty, for sixty days or more.

Another contractor answered the same night that they turn about 30 percent of estimates into paying jobs, and called the other 70 percent “literally wasting my time.”

Those two numbers are as much as three times apart, from people in the same trade. And that thread is from 2006. Twenty years later, nobody has published a better answer.

Treat both as what they are. Two contractors on a forum, no definitions agreed on. We’re not presenting them as data. We’re presenting them as evidence that the data doesn’t exist, and that contractors are flying blind and know it.

The real reason follow-up stops early

Here’s the part the statistics miss entirely.

The contractor who started that thread was already following up more than most shops ever do. And their closing question wasn’t about the numbers. It was whether they should keep being a pain about it.

That’s the real reason follow-up stops. A third phone call about a panel upgrade does feel like pestering, and plenty of contractors would rather lose the job than be the one who won’t leave people alone. That’s not laziness and it isn’t a training problem. It’s a decent instinct producing a bad business outcome.

Email asks less of the customer. A message sitting in an inbox can be read later, or ignored, without anybody having to pick up the phone and say no. That’s the real argument for running follow-up over email instead of by phone, and it’s a better argument than any response-rate number anybody can show you.

There’s a fair question underneath all this: do you present options in person, or send an email and hope the customer can make sense of the quote? That’s the real fork. The email isn’t there to replace the conversation. It’s there for the two weeks after the conversation, when the homeowner is deciding and you’re on another job.

What we can’t tell you yet

This is the part where an agency shows you its own numbers.

We don’t have any to show you. Not because the sequences aren’t running. They are, for clients right now. It’s because no client has agreed to let us publish their results with their name attached, and we haven’t asked hard enough yet.

We could publish an anonymous aggregate instead. Plenty of agencies do: a response rate with a percent sign on it, nothing behind it, and no way for you to check.

That’s the same unverifiable claim this article just spent a thousand words taking apart. Writing it here would make the whole page dishonest.

So the honest position is this. When a client says yes to being named, with real send counts and real booked work, we’ll publish it and you can check our arithmetic. Until then, don’t take our word for anything in the next two sections. Take the reasoning or leave it.

The sequence we run

Five emails. Here’s what each one is for.

One caveat before the list. We’re not going to tell you which touch produces the money, because we haven’t published the data that would justify saying it. Anybody who tells you the fourth email is where it happens is repeating the 1942 survey in a different outfit.

The recap, 48 hours out. Not a nudge. A restatement of what you quoted and why, because two days later the homeowner has already lost the paperwork.

The question, day five. One question, easy to answer. What’s the piece you’re unsure about?

The objection handler, day ten. Address the thing they didn’t say. Usually permits, usually whether the price includes everything, occasionally whether you’ll actually show up.

The useful one, day twenty. No ask at all. Something that helps them regardless of whether they hire you.

The close, day thirty. Plain. Where the quote stands, when it expires if it does, what happens next, and how to say yes.

That’s follow-up on open estimates. For customers you’ve already finished work for, the emails are different.

What to do with this on Monday

Tearing down other people’s numbers doesn’t book a job. Here’s what does, and none of it requires buying anything.

Pull your own close rate. Estimates sent last quarter, jobs won from them. Now you have your number instead of somebody’s conference slide. It takes ten minutes in ServiceTitan or Housecall Pro.

Count your touches. Go look at the last ten estimates that went cold. How many times did anybody actually reach out? If the honest answer is twice, that’s your gap, and it’s a bigger one than any subject line.

Pick the segment nobody works. See the next section.

Stop repeating statistics you can’t source. Including ours. If a number shows up in a pitch and the person can’t tell you where it came from, it’s probably no better sourced than a 1942 survey of fewer than forty people on Long Island.

The one nobody sends

There’s a category sitting in every contractor’s software that a lot of shops never work: estimates the customer approved that never got on the schedule.

They said yes. They agreed to the price. Somebody was supposed to call and put it on the board, and then a callback came in or a tech got pulled and it quietly fell through the cracks of a busy week. Nobody canceled anything. It’s money you already won, aging in a list.

We went looking for a published figure on how much of this sits in a typical residential contractor’s account, and we couldn’t find one.

So go measure yours. Pull the estimates marked sold that never became a scheduled job, and sort them by dollar amount. ServiceTitan already groups them for you; here’s where. That number will tell you more than anything in this article, and it costs you nothing but ten minutes.

Questions we get

Is it true that 80% of sales require five or more follow-ups?

There’s no modern evidence for it. The figure traces to a 1942 survey by the Long Island chapter of a sales association, with a sample of fewer than forty people. The association’s successor states this on its own website.

How many times should I follow up on an estimate?

Nobody can answer that with credible data, and anyone quoting a specific number is repeating something untraceable. What contractors report to each other varies widely. Pull your own numbers and follow up more than the two touches a lot of shops stop at.

What’s a normal close rate on estimates?

No trustworthy benchmark exists for the trades. In a 2006 electrician forum thread, one contractor reported one or two jobs per ten estimates and another reported about 30 percent. That’s as much as three times apart, and nobody has published anything better since.

Is it annoying to follow up on a quote?

By phone, it can feel that way, and that worry is why a lot of follow-up stops early. Email asks less, because a message sitting in an inbox can be read later or ignored without anyone having to say no.

What is a sold but unbooked estimate?

A job the customer approved that never got scheduled. They agreed to the price and nobody put it on the board. We couldn’t find a published figure for how much of this sits in a typical contractor’s system, which is why it’s worth pulling your own.

The figures above are cited so you can check them yourself. If a statistic on this page stops holding up, tell us and we’ll fix the page.