Email marketing ROI is usually attributed incorrectly

Let me ask you a question. The last time you walked into a 7-11 to buy a can of Coke, did you tell the shopkeeper, “I saw a Coca Cola billboard outside, and therefore I would like to purchase a can of this sweet, sweet nectar.”  Unlikely, right?

If you do email marketing, you know that some people do buy from your emails. For sure! It’s a powerful direct response channel. But that’s not all it is.

Take, for example, Travelodge. In October 2012, the company sent me a bunch of emails  Here’s a screenshot of my Gmail inbox, filtered for Travelodge during that period:

Travelodge Emails

Notice how I didn’t open a single one. According to the widespread email marketing axiom, I’m a disengaged email address. I’m a lost cause, one of those annoying people who simply doesn’t open their emails.

And yet, when my friend got married in Cardiff, where did I stay?

At the Travelodge. Why? Because Travelodge gets it: the inbox is a branding tool. 

Take a look through those subject lines. Nearly every single one is saying how I can make great savings, how it’s the market leading discount hotel brand.

Consider this: what are the odds of Travelodge sending me an email at the exact moment I’m ready to purchase a hotel for the night? Slim to none.

What Travelodge has done is consistently reinforced its brand ideals – low cost, adequate hotel rooms – through a series of emails.

I didn’t open them, sure. But I still saw the from name and the subject lines: discounts, cheap, cost-effective. And when I was in buying mode, I went to Travelodge’s website and booked direct, because this brand was top of mind.

I didn’t buy that can of Coke because of the billboard. I bought the can of Coke because the brand was top of mind when I was in purchase mode.

Don’t forget, email is a form of advertising

Lots of email experts will tell you that you need to be relevant, you need to segment, you need to be mobile, and so on. My point is not about this.

Relevance works. Segmentation works. Responsive design works (probably.)

But, if you aren’t emailing someone, you aren’t getting your brand in front of them in the first place, regardless of if they are in that traditional ‘engaged’ group or not.

So let’s fast forward to Christmas…

To set the tone, here’s an out of context image of a bunny wearing a Santa hat:

Bunny with Santa hat
(h/t Mike Smail)

At Christmas time, most retailers spend more money on advertising than any other time of the year. All of the major brands, from Macy’s to John Lewis to everyone else big and small, have Christmas campaigns: ’12 days of this’ and ’12 days of that.’ Whatever.

So advertisers are doing more:

  1. TV spots.
  2. Radio spots.
  3. Print ads.
  4. Offer/coupon inserts.
  5. Catalogues/direct mai.
  6. Outdoor advertising.
  7. Point-of-sale offers.
  8. Online display.
  9. Paid search.
  10. Social media stuff (ads, tweets, pages, whatever;) and so on, and so on, and so on.

And yet, people say that you shouldn’t send out more email.

“Because we don’t want to annoy customers”

I get it, you don’t want to be a spamming bastard. But let’s be frank, what really annoys people is seeing the same TV commercial over and over and over. And hearing the same annoying jingle on the radio at work. And so on.

How many times have you gone to sleep at night after watching TV with an annoying jingle stuck in your head? (The bane of my existence for about six months back in the 80s was the Meow Mix song. It still gives me shivers.)

Or, like the trend is these days, when you go to a website, and leave that website, then have banner ads stalk you around the internet. You go on YouTube to watch a squirrel on a surfboard and are bombarded with aggressive ads for a site you visited the day before.  That’s annoying.

So does email marketing annoy people? People can easily unsubscribe if they get annoyed, which is a lot easier than never hearing your jingle on TV or radio, or no longer seeing your banner ads across display advertising networks.

“We have to keep user experience in mind.”

A few years ago, there were a handful of user experience gurus in the world. Now they’re everywhere. That’s cool, we’re good dudes, I’m just saying it’s a category that barely existed when the dogs were still being let out.

Let me pose you this question: what’s a better user experience than a series of relevant emails, offering some great deals on Christmas gifts from your company?

So this is the thing. If you’re sending out a load of spammy shit on a regular basis, you’ve got a problem.

What I’m not saying is send out a load of crappy emails. What I am saying is, don’t be afraid to send out more emails so long as they’re good.

“We should let the user decide how often they want to receive emails.”

So here’s a fun task. Call up 10 of your customers who have purchased directly from emails, and ask them one question, with no preamble: “How often would you like us to email you: daily, weekly, monthly or quarterly?”

None will say daily. One may say weekly, one quarterly, and probably eight will say monthly.  The thing is, what people say they want is different from their actual purchase behaviour.

People think they don’t like to be advertised to. And, every consumer thinks that they’re less susceptible to advertising than anyone else on the planet. It’s like how everyone thinks they’re a better poker player than everone else at the table, but there’s only one cowboy counting his chips when the game is over.

If you only send emails to people once a month at their request, whether they’re regular customers or not, your brand won’t be top of mind.  Your competition’s will be.

So once again, I’m not saying spam the hell out of your customers. What I am saying is that if you restrict frequency of emails at the Christmas period, where everyone and their dog is after your customer’s hard earned dollars, guess what? Someone else will get there before you.

Re-visit your attribution model if you’re worried about email volume at Christmas

We, as marketers, do marketing because it works. It helps move product. It delivers revenue. It pays our paycheques.

And yet, the most popular strategy is to measure the ROI from the email channel based upon last click. Email marketers have an obsession with statistics: opens, clicks, unsubscribes – and sometimes miss the basics. A little thing you may remember from Marketing 101 called AIDA.

Most emails are sent out with that final A in mind – Action – and ROI is attributed based upon this. The thing is, email is an extremely powerful channel for the Attention and Interest phases, regardless of if an email gets opened.

So how can you determine email channel revenue uplift for your brand this Christmas?

If you’ve made it this far, you think one of two things: 1) I’m a tosser; or 2) I’ve got a point. If you’re in the #1 camp, fair enough, maybe you’re right.

If you’re in the #2 camp, consider this:

Say you’ve got a list of 500,000 email addresses that have not purchased from the email channel in the last 12 months (and stripping out unsubs and bounces.)

Note that the proven purchasers from the email channel should be held in a different segment; this test is about attributing revenue correctly to the email channel.

That is, anyone who is ‘disengaged’,  those who would get fewer or no emails following the prevailing beliefs of email marketing dogma.

Here’s what you do with that list of 500,000 who are perceived to not respond to the email channel:

  1. Set aside 100,000 and add them to a quarantine list. Send them no emails whatsoever.
  2. Take another 100,000 and let them tell you the frequency they wish to receive emails. Send them emails at that frequency. 
  3. With the next 100,000, send them your regularly scheduled programme, whatever it is.
  4. With the next 100,000, send the initial email with five subject line splits, and then do a re-send to non-openers with the best performing subject line.This shouldn’t be hard, any ESP should be able to do this off the shelf. This will be doubling the frequency to the non-openers.
  5. Lastly, the ultimate 100,000: send them loads of (relevant) stuff. Triple the regular frequency, to everyone. Openers and otherwise.

After the Christmas period, line up all your sales and dupe based upon email address, irrespective of whether or not they’ve clicked or opened correlate the number of emails sent per contact with the revenue generated.  Which one did the  best?

Of course, every business will be different, and your results may differ from your neighbour. But I’m confident enough to say that in the majority of cases one and two won’t buy the most stuff. Three may. And if it is the winning segment, kudos to you, you are already an email marketing ninja.

But my money, all things being equal, is on four and five.

If those last two out-perform the first three by a decent margin, and the uplift of five on four is negligible, re-sending to non-openers is your dominant strategy. If five is the clear winner, it’s time to invest more time and money into your email channel.

I’ve probably made no friends with this blog post…

I’m not gonna lie. I’m expecting a bunch of people to tweet about this saying that I’m an idiot. That’s cool, I can take it. Feel free to chuck in a comment below if you want to debate.

Simply put, email is more than a direct response channel, email is a key branding tool for digital brands,  so consider that this Christmas.

On December 26th it’ll be too late… and if your sales are short, there’s no one to blame but yourself.