Marketing’s Weekly Dose of the Truth

Ken Magill

About Us

Email Response Rates Pathetically Low


By Ken Magill

Email response rates are pathetic in comparison with other channels, according to a recent report from the Direct Marketing Association.

Email campaigns sent to house files drive an average 0.12 percent response rate—response being defined as the user having taken some sort of action called for in the message, according to the DMA.

In comparison, letter-sized direct mail drives an average 3.4 percent response rate, almost 30 times that of email, according to the DMA. And this is despite direct mail’s average response rate having dropped 25 percent in the last nine years, according to the DMA.

“Even though direct mail is less effective in driving response than it was a decade ago, it still is among the best media for generating overall response,” said Yory Wurmser, the DMA’s director of marketing and media insights.

The only channel with a lower response rate than email is display ads at 0.4 percent, according to the DMA.

Somewhat surprisingly, the highest response rate is delivered by the much-maligned—by consumers, anyway—channel of telemarketing.

Phone calls to customer files achieve an average 12.95 percent response rate, according to the DMA.

However, telephone marketing also had the highest costs: nearly $78 per order or lead for a house list, and $190 for a prospect list, according to the DMA.

And even with such low response rates, email still overwhelmingly outperforms all other direct marketing channels in terms of return on investment.

According to the DMA’s survey, email generates $28.50 for every dollar spent. And while this figure is dramatically lower than the ROI figure the DMA published for email last fall, it’s still comparatively high.

“We found that ROI for email is 4x that of direct mail, so there’s not a systematic bias in the data for direct mail,” wrote Wurmser in an email exchange with The Magill Report. 

And as has always been the case, email marketing’s greatest asset—its low costs compared to other channels—is also its biggest curse.

“The double-edged sword with email has always been that the more you send the more revenue you generate,” said Loren McDonald, vice president of industry relations for email service provider Silverpop.

However, as marketers continue cranking up the volume of email they send, they run the risk of driving up their unsubscribe and complaint rates to unacceptable levels, he said.

“Every company has to find that perfect balance,” he said. “At some point you cross a line where the cost to reacquire customers, the number of unsubscribes you get, and the number of spam complaints you get outweigh the added revenue.”

McDonald recommends, among other things, setting up triggered-email programs, such as abandoned-order messages.

“The smarter marketers are continually adding one-to-one behavioral-based email marketing,” he said. “We have a client who has 40 different email messages going out every single day and she doesn’t press the ‘send’ button on a single one.

“The beauty of these emails is that any one of them might be going out to just 100, 1,000 or 10,000 people, but the conversion rates—10, 20, 30, 40 sometimes 50 percent—makes the math work,” McDonald said.

Another issue affecting email response rates is marketers with large percentages of inactive addresses on their files, said McDonald.

“I’ve been hearing lately that a lot of clients have files that are 50, 60 and even 70 percent inactive,” he said. “When you start to get numbers like that, you start to get deliverability problems.”

The dilemma over whether or not to remove inactive addresses has been covered extensively in this newsletter.

That debate aside, McDonald recommends focusing more on the most active addresses in the database.

“If you start to focus more on those people who are actually doing things, such as opening, clicking and buying, then you’re dealing with that core group and getting more out of the people who are actually engaged,” he said.


Show: Newest | Oldest

Post a Comment
Your Name:
Please type the letters in the image above

Terms: Feel free to be as big a jerk as you want, but don't attack anyone other than me personally. And don't criticize people or companies other than me anonymously. Got something crappy to say? Say it under your real name. Anonymous potshots and personal attacks aimed at me, however, are fine.

Posted by: Dela Quist
Date: 2012-06-25 17:01:26
Subject: I should have known better :-)

Very good analysis Yory - I made the mistake of attributing Kens rather emotionally charged (but thats your job right Ken :-b) use of the word pathetic to describe email response rates to the DMA report. I then compounded my error by NOT reading the actual report!!! Rather than worry about rates marketers should worry about numbers. How many people are you converting and how much money are you generating. Percentages are for the show, numbers are for the dough
Posted by: Yory Wurmser
Date: 2012-06-25 11:46:55
Subject: Re: The problem with unclarified stats

Steve, I'm gratified that you ran your own numbers, and that the overall figure for email came close to my figures (albeit still higher). I also agree that the direct mail numbers are higher than I would expect -- but there are a couple of factors that make me feel that this result is plausible. Because of the per piece cost of direct mail, marketers have a stronger financial incentive to run a priori analytics that reduce volumes and increase response rate. Where the life-time value of a new customer is high enough, for instance credit cards, the response rate is typically much lower (around 0.5% or lower). In these cases, the cost constraints are counterbalanced by the revenue possibilities. The short-term economics for email do not have these constraints, which is why a lot of emailers run large and relatively undifferentiated campaigns. I have no doubt that highly targeted, triggered email campaigns have response rates many times what an undifferentiated campaign would have, and perhaps would rival similar mail pieces. With mail's current economics, however, I would bet that a far greater proportion of mail campaigns are highly targeted. Second, I was pretty rigorous in excluding answers for mail that fell above two standard deviations away from the mean, which cut some answers that didn't pass the sniff test. Lastly, the data for mail goes back ten years, and each year we've had a decline in response rates. The data on mail is also completely consistent with other reports. Even if I took the mean email response rate (0.33) and cut the direct mail response rate in half (1.70%), the conclusions would be the same. Response to mail is declining, but from a pure response perspective (as marketers currently use the two media, which is a big caveat), mail currently beats email handily. From an ROI perspective, the opposite is true.
Posted by: Steve Henderson
Date: 2012-06-22 12:32:08
Subject: The problem with unclarified stats

I have just ran an analysis from a few years of email campaigns from a couple of hundred companies running email marketing campaigns who also record conversions. If I add up all of their sent emails and add up the number of people who converted I get an "email response rate" of 0.16% But, what this does is totally skew the data with a massive weighting bias towards the non-targeted, unintelligent bulk marketing via email. Instead, if I take the same data and simply take the mean average conversion rate (or response rate) for these email campaigns I get 4.85%. This second figure is totally skewed towards the very large number of small, targeted, one-to-one, automated, triggered, intelligent email marketing campaigns which generate high conversion rates (which is a point which Loren makes). I suspect the direct mail figure you have quoted (3.4%) has been obtained in a way similar to my second approach - but hey, it is stats: they could mean anything.
Posted by: Yory Wurmser
Date: 2012-06-21 13:16:50
Subject: Some clarification

I appreciate everyone's concerns with the findings in this report -- and particularly the response figure for email of 0.12% -- so let me explain how I got there. All of the following methodology is clearly laid out in the report itself. The response data for all channels came from a survey deployed online in April to the DMA list, which contains a broad range of marketers, from the traditional to the latest social media gurus. It's a reasonable sample of the US marketing community, although it probably skews toward larger marketers. For mail, I directly asked respondents about response rates, since this a fairly standard metric for mailers. For email, I requested open rates, CTR and conversions per click. I then multiplied CTR by conversion per click to get the response rate. Loren's comment about definitions for response is astute, and Ken's reply accurately reflects my conversation with him. I went back to the actual questionnaire, however, and found that it didn't specify that email conversions (nor mail response rate) could cover any type of action. Loren may be correct that this may have lowered the relative response rate for email compared with mail. The comparison between email and direct mail also is inexact in the way that the averages were calculated. For direct mail, the answers to response rates more or less formed a normal curve. I took the mean and threw out several answers that fell more than two standard deviations above the mean. The median and mean in the end didn't vary by a lot. For email, the conversion numbers came in all over the place with a small cluster well above the other responses. This produced a large standard deviation, few outliers from the mean, and a big gap between median and mean. Within the report, I'm upfront about this and present both the mean response rate (0.33%) and the median (0.12%). I went back and forth on which to give precedence. In the end, I decided on the median, which I felt was both a more accurate reflection of average performance and a fairer basis of comparison with other channels. It was a judgment call, though. Finally, the 0.12% figure seemed low to me, so I checked it with a few marketers who said it was plausible. Let me also briefly address the question of bias. In no way is this report an indictment of email. Response rates are one component of measuring the success of a channel. The ultimate guide to the value of a channel is ROI, and this report, as Ken writes, shows email as having an ROI four times that of direct mail. Most of my research these days focuses on digital channels. I assure you I designed this report without any pressure to favor mail, nor is it part of my (or DMA's) agenda to do so.
Posted by: Ken Magill
Date: 2012-06-21 07:05:31
Subject: Response Rate Defined

Hey Loren: According to Yory Wurmser, the DMA defined "response" the same across all channels: the percentage of people who took some requested action. He told me the email response rates weren't all purchases either.
Posted by: Loren McDonald
Date: 2012-06-21 02:10:37
Subject: Apples and Oranges

I discussed with Ken what I believe to be an apples to oranges comparison for DM versus email and search, for example. The DMA is basically using the math of a 4% average CTR X a 3% average conversion rate - which = the .12% "response rate. But the DMA didn't define what the response rate = for DM. I've always understand response rate for DM to be a "response" - e.g., sending a BRC card back in; calling an 800 number; it is an action, but not necessarily a purchase. So in my mind they are comparing apples and oranges. But hard to know without more transparency around what is going into the definitions and calculations.
Posted by: Bob Frady
Date: 2012-06-20 17:52:14
Subject: 3.4% for Direct Mail?

Since when? The DMA has a LOT of clients who are direct mailers- looks like they needed some love. The DMA has been trumpeting an artificially high ROI for email for years. This new number is a lot closer to the truth.
Posted by: Steve Henderson
Date: 2012-06-20 09:49:03
Subject: Erm. No. I don't think so

Average of 0.12% response rate? Email generates an AVERAGE of 12 responses per 10,000 emails sent? Sorry, but I am calling Bullsh*t on that one. Where is this DMA report?
Posted by: Dela Quist
Date: 2012-06-20 04:21:04
Subject: Cigarettes less harmful than alcohol - Study funded by the tobacco industry

Puhlease!! I was going to leave it at that, but thought I would share this. The DMA in the UK spent quite a long time a few years back trying to prove that email had a higher environmental impact than DM!
Posted by: Gretchen Scheiman
Date: 2012-06-19 20:33:12
Subject: Where did the #s come from?

I would like to understand the basis for these numbers better. I have never seen anything near the very low 0.12% response rate to a house list through email, though I used to expect ranges nearing that for monarch-sized DM (slightly smaller than letter sized - people used to complain that letter sized could be a bit of a bait-and-switch in some industries as they appeared to be more important than they were). Then again, I have heard DM response rates are picking back up now that no one uses it anymore, and certainly if we're excluding postcards and such then we're taking the highest responding DM form-factor to compare against all email. But isn't that a bit of a skewed view of the two channels? These numbers don't smell right. Would anyone from the DMA be willing to share more about why these comparisons were chosen? And where the email metrics came from?