There is no reliable published response rate for roofing direct mail. The figures quoted most often, 9% for house lists and 4.9% for prospect lists, date from 2018, come from a survey of marketers across all industries, and were the highest that survey had recorded since it began in 2003, as MediaPost reported in 2019. So when you ask what roofing direct mail response rate to expect, the honest answer is a method, not a number. This post covers where the common figures come from, what they leave out, why one small campaign can't tell you your real rate, and how to work out the rate your mail actually needs.
Where do the usual response rate numbers come from?
Most figures you'll find trace back to one survey series: the Response Rate Report, published years ago by the Direct Marketing Association and more recently by the Association of National Advertisers (ANA). Marketers fill in a questionnaire about their own campaigns. The ANA keeps its own results pages for members only, so almost everyone quotes the report secondhand.
Here is what two public summaries of that series say. Note how far the numbers move.
| Survey | House list (past customers) | Prospect list (new homes) | What the summary tells you about the sample |
|---|---|---|---|
| 2012, DMA | 2.47% to 4.26%, by format | 0.94% to 1.44%, by format | Responses from 481 companies (Chief Marketer, June 2012) |
| 2018 data | 9% | 4.9% | None; both figures were the series' highest since 2003 (MediaPost, September 2019) |
The 2021 edition is the one with the most public detail. According to MediaPost's January 2022 write-up, the ANA surveyed 581 marketers from February through July 2021, and 56% of them worked at firms with annual revenue of $25 million or less. The same write-up says only about a quarter of the direct mailers in the survey mostly used prospect lists.
What those numbers don't tell a roofer
They mix every industry. None of the public summaries break out home services, let alone roofing. A catalog sent to people who bought from it last month and a letter to a homeowner who has never heard of you are averaged together.
House list and prospect list are very different jobs. For a roofer, the house list is past customers and people who already asked you for a quote. Mailing homes that don't know you yet is prospect mail, so the house list column doesn't apply to it at all.
Respondents report their own results. The public summaries don't say how each company counted a response or over what time window. And the prospect figure moved from roughly 1% in 2012 to 4.9% in 2018. A jump that big in six years may say as much about who answered the survey as about mail itself.
A useful way to read the benchmarks: they show that direct mail gets responses in many industries. They can't tell you what your letter to your ZIP codes will do.
What counts as a response for a roofer?
A response is a homeowner contacting you because of the piece: a call, a text, a QR code scan, a web form. It is not the same as a booked inspection, and it is a long way from a signed job. Keep the three numbers apart:
- Response rate = responses ÷ pieces mailed
- Booking rate = inspections booked ÷ responses
- Close rate = signed jobs ÷ inspections
Decide before the drop what you count. Wrong numbers, sales calls and past customers calling about an old warranty are not responses to this campaign. Two scans from the same phone are one person. If you change the definition halfway through, the rate stops meaning anything. Our guide to tracking direct mail leads covers the setup: a tracking number per campaign, a tagged QR code and a matchback of signed jobs.
Why one campaign can't tell you your rate
Small mailings produce small counts, and small counts bounce around by chance. The table below shows how many responses a mailing would bring in 9 out of 10 campaigns if its true response rate were 0.5% or 1%. The ranges come from a standard probability calculation. You don't need to redo it, just read the ranges.
| Homes mailed | True rate 0.5%: expected, typical range | True rate 1%: expected, typical range |
|---|---|---|
| 1,000 | 5 responses, 2 to 9 | 10 responses, 5 to 15 |
| 2,500 | 12 or 13 responses, 7 to 19 | 25 responses, 17 to 33 |
| 5,000 | 25 responses, 17 to 33 | 50 responses, 39 to 62 |
At 1,000 homes, a 0.5% mailing that gets lucky (9 calls) can beat a 1% mailing that gets unlucky (5 calls). You would drop the better letter and keep the worse one. Only at around 5,000 homes do the two ranges stop overlapping.
The practical rule: judge a letter on the total of several drops with the same offer and the same definition of a response, not on one drop. And change one thing at a time (the list, the letter or the offer), or you won't know what made the difference.
What response rate do you actually need?
A better question than "what's normal" is "what rate pays for this mail." That number you can work out in a minute:
Break-even response rate = cost per home ÷ (share of responses that become jobs × gross profit per job)
Example with made-up inputs; swap in your own. Say mail costs $1.00 per home all in, one in five responses turns into a signed job, and a job leaves $3,000 in gross profit. Then $1.00 ÷ (0.20 × $3,000) = $1.00 ÷ $600 = about 0.17%. That is one response for every 600 homes.
Now look at it per job instead of per piece. With the same inputs, 1,000 homes cost $1,000. At a 0.5% response rate that's 5 responses and 1 job, so about $1,000 per signed job. At 0.25%, it's about $2,000 per job. The price per piece didn't change; the cost per job doubled. That's why cost per piece alone is a poor way to compare mail options, and why our post on cost per job versus cost per piece walks through the full calculation.
The useful question after a few drops is whether your pooled rate clears your break-even with room to spare, not whether it hits 4.9%.
How to judge a response rate someone quotes you
Vendors, coaches and other roofers will quote rates. Before a number shapes your budget, ask:
- How many homes were mailed, and how many responses was that? Three calls from 100 letters is 3%, and it means very little.
- What counted as a response: calls only, or scans and form fills too?
- Were these past customers or homes that had never heard of the company?
- Over what window after delivery were responses counted?
- Was there a holdout group that got no mail, or does the number include people who would have called anyway?
- One market and one season, or many? A drop right after a hailstorm says little about a quiet month.
- Who ran the numbers, and is this a typical campaign or the best one?
- How many signed jobs came out of it, and at what cost per job?
A rate with clear answers to most of these is worth weighing. A rate with none of them is a marketing claim.
How to find your own number
Give every campaign its own tracking number and QR code, write down your response definition and counting window before the drop, and log each response through to booked and signed. After three or four drops, add up homes, responses and jobs across all of them and divide. That pooled rate, held against your break-even, is the number that matters. If you mail at volume, a holdout group shows how many of those jobs the mail actually added; the tracking guide explains how.
Where Scoutbound fits
Scoutbound is built so this count is easy to keep. It sends letters only to homes whose roofs show visible wear in aerial images, under your brand, each with an aerial image of the homeowner's own roof. Every letter carries its own QR code, which leads to your website, and a tracking phone number. Calls, texts and QR scans show up in your online portal, and calls and texts trigger an alert, so you can count responses per campaign instead of guessing. Signed jobs stay in your own records, so close rate and cost per job are yours to work out.