hi, i’m amr. every time offline marketing comes up in a digital team, someone quotes the direct mail response rate and someone else laughs. both reactions skip the useful part: what that number measures, and which list it was measured on.
the two numbers behind the headline
the figures most often quoted come from the ANA and DMA response rate reporting: about 9% for a house list (people who already know you), and about 4.9% for a prospect list (cold, in-market names). the reported ranges are 5-9% and 2-5%. an automotive-focused compilation puts the overall average for that sector at 3.84%.
read the two headline numbers side by side and the real message appears. the same mail piece pulls roughly 1.8x the response from people who already know you (9 divided by 4.9). the list is doing more work than the creative.
what a “response” actually counts
a response in these reports is any measured reply: a call, a visit, a code scan, a web visit. it is not a sale, and it is not a qualified lead. that makes it a different kind of number from a click-through rate or a conversion rate, which is why comparing it straight to email or paid social is shaky. the sources that line them up in a table don’t name where their email and paid social figures come from, so i wouldn’t build a budget case on those comparison rows.
a second warning: the publication year isn’t consistent across the pages repeating these numbers. one compilation attributes the ranges to a 2025 ANA/DMA report, others give the same averages with no year at all. treat them as directional benchmarks, not as a forecast for your mailing.
what the arithmetic looks like on a 5,000-piece drop
- prospect list at the 4.9% benchmark: about 245 responses
- house list at the 9% benchmark: about 450 responses
that is straight multiplication of the benchmarks, not a prediction for any account. the number that decides whether the campaign was worth it comes after: cost per response (print, postage and list cost, divided by responses) and then cost per customer once responses are qualified.
how to make direct mail measurable like a digital channel
- give every piece its own door. a unique short URL or QR code per segment, and a dedicated phone number if calls matter. without it, mail shows up as “direct” traffic.
- hold out a control group. leave a slice of the list unmailed, then compare against it. otherwise you can’t separate people who responded to the mail from people who were going to buy anyway.
- split house and prospect results. blending them hides the 1.8x gap and makes both look average.
- count the downstream step. track responders into qualified leads and closed deals, not just the response count.
how this connects to the rest of the stack
GA4 captures the QR and short-URL visits as their own campaign source, and HubSpot stores “direct mail” as the lead source plus the segment, so responders can be followed through to deals the same way paid traffic is. once mail has a source field and a holdout, it stops being a faith-based line item and starts competing for budget on the same terms as Meta or Google.
where this stops applying
if you don’t have clean, current address data, the house-list figures don’t apply to you at all, and a cold-prospect rate can be the optimistic case. mail also isn’t a fit for low-value, high-volume offers where postage and print eat the margin before the first response arrives.
start scaling — if offline spend is sitting outside your reporting, a unique URL and a holdout group fix that this week. let’s connect.
Sources: ANA/DMA response rate figures as reported by 4over4 and DemandLocal compilations (house list about 9%, prospect list about 4.9%; automotive average 3.84% as reported via MailPro). Report year varies by source. Figures are published industry research, not verified results from Amr’s own client accounts.