This week, your inbox is probably filled with Black Friday emails. And they’re all pretty much the same. “20% off now!” or “Don’t miss out!” or “Buy this TV or else I’ll club this baby seal!”

And since every retailer seems to follow the same approach, everyone else follows suit.

You don’t want to be that one brand that misses out on one of the biggest shopping days of the year.

But have consumers become desensitised to hard-discounting? This is the question I had, and here are the statistics to answer.

It’s no secret that more emails during peak sales periods result in more sales. Because email is a form of advertising. And advertising works. This is no secret.

But when we look around the advertising landscape, the most successful seasonal ads aren’t about hard discounting. Take, for example, John Lewis’ latest ad.

Sure, there’s some astute product placement (and who doesn’t want to buy their parents a telescope for Christmas) but overall, it’s not a direct-sales message.

And yet – when Black Friday comes around, email marketers around the world rejoice and send out messages that sound, to quote a Thai vendor at a street market, “same same but different”.

So I wondered – do the results warrant this method of marketing? That is, to hard-discount via email, with very little message variability?

If you’re from the “last-click-attribution” camp, that’s one thing, and you’ll probably disagree with this post. “Moar salezzz” and whatnot, all with a myopic view of Black Friday, without considering the opportunity cost of such messaging.

But if you believe in the value of a digital brand, read on.

According to a sample of 3,892 Black Friday-themed email subject lines

We algorithmically rank subject lines using Phrasee Pheelings, which deconstructs subject lines into its core emotional components. (See this post for background info.)

And, we also collect huge amounts of emails sent by worldwide brands.

Using a Markov chain Monte Carlo method, we are able to simulate and predict the effectiveness of one subject line versus another, in terms of opens, clicks and conversions.

I won’t bore you with the statistics. Here it is in a nutshell: based upon a sender’s profile and the content of their subject line, we are able to simulate consumer response to it.

Using a Bayesian model, we are then able to quantify its effectiveness in terms of opens, clicks and conversions.

We aggregate these metrics into what we call a “Phrasee Score” – a weighted statistic that is a proxy for overall subject line effectiveness.

TL;DR: A low Phrasee Score is bad. A high Phrasee Score is good. For background info, read this post.

So what do the stats say?

We analysed all 3,892 subject lines and got the emotion scores for the five key emotional drivers in a direct marketing environment – curiosity, directness, friendliness, offbeat, and urgency.

Note: the data was from the US and the UK. We didn’t separate them, and indeed there could be cultural differences. But this is outside the scope of a single blog post.

The data was all from the retail/ecommerce industry, so once again there could be differences in product-specificity.

But this is a blog post, not a novel. So, while there are limitations to the methodology used, the results remain thought provoking – and likely the most in-depth look at Black Friday subject lines out there.

Here are the descriptive statistics from the sample set:

The trend in Black Friday subject lines is towards very direct and urgent. Well of course – sell, sell, sell! That’s the industry mantra on the day after Thanksgiving, right?

But when you look at the correlation between the Phrasee Score and the results, you can see that there is a positive relationship between a higher Curiosity score and a higher Phrasee Score... and that Urgency has a negative effect. Interesting!

But: remember, correlation does not equal causation. What these statistics don’t mean is, “Never use urgent drivers in your Black Friday emails.”

To interrogate the stats further, I used a K-Means Cluster method to identify “emotional clusters” that existed within the data. This is where the real interesting trends identified themselves.

Note: a ‘cluster’ is a statistically-related group within a data set. So, above we find there are three noticeable clusters – each of which centre around different emotion scores.

The most used cluster, #2, used in 53% of cases, is very urgent and very direct. And not much else. And the effect on the Phrasee Score? -21% versus the baseline. That’s not good.

Some examples from this sample set:

  • Black Friday Sale: Save 20% + Free Delivery
  • Black Friday Special - Save $100
  • Half Price Black Friday Deals - Ends Sunday

The last one is particularly weird. How can you have a Black Friday deal ending on Sunday? What the what?

The best cluster? #1. But it was only used in 12% of the sample set. It was less direct, less urgent, and more curious and friendly. And the effect? A 14% lift versus the average Phrasee Score.

Some examples:

  • BLACK FRIDAY is here - want a little help with your Christmas shopping?
  • How you can avoid stress on Black Friday
  • Tick tock, tick tock, the BLACK FRIDAY countdown begins...

Note that they are all obviously about Black Friday, that’s clear, as denoted by the directness centroid. And yet – none of them are the pure sales messages. Interesting.

So what should you do on Black Friday?

The above are all broad trends, of course. Your brand’s voice is different from everyone else’s out there. The key is to test.

But not just any testing. The key is to test out a wide variety of language constructs and to have a learning methodology in place.

Try to inject some curiosity in your lines. And some urgency. And some friendliness. A pure sales message may work the best for you, but if you don’t test out different emotional clusters, how will you know?

Stand out from the crowd & don’t just be another spammer

What has your experience been with Black Friday emails, both as a marketer and as a consumer? Chuck your ideas in the comments below.

Here's the thing: don’t get lost in the clutter and just come across as desperate.

Use a variety of language to understand what makes your audience tick. Be creative. Be courageous. And mostly, be awesome.

Because, thankfully, Black Friday comes but once a year, and if you don't get noticed, you don't make the sales.

Parry Malm

Published 24 November, 2015 by Parry Malm

Parry Malm is the CEO of Phrasee and a contributor to Econsultancy. Connect with him on LinkedInTwitter or Google+.

25 more posts from this author

Comments (3)

Avatar-blank-50x50

Camilla Zajac, Copywriter at Green Light Copywriting

Interesting post, thanks!

over 1 year ago

Pete Austin

Pete Austin, CINO at Fresh Relevance

Here's a reality check on that example data. Just basic maths:
* cluster 2 is 53% of cases and gets -21% lift vs baseline. That's about half the data.
* So the other half of the data must get about +21% lift to balance it and keep the baseline at 100%.
* So there must be other cluster(s) getting better than +21%
* Yet cluster 1 is described as the "best" and only gets +14%.

over 1 year ago

Parry Malm

Parry Malm, CEO at Phrasee Ltd.

Pete, there were more clusters. I removed the boring ones. But thanks for the math tips, obviously an area I'm pretty weak at.

One of these days you should stop trolling and write your own content. I'm sure it would be interesting.

over 1 year ago

Comment
No-profile-pic
Save or Cancel
Daily_pulse_signup_wide

Enjoying this article?

Get more just like this, delivered to your inbox.

Keep up to date with the latest analysis, inspiration and learning from the Econsultancy blog with our free Digital Pulse newsletter. You will receive a hand-picked digest of the latest and greatest articles, as well as snippets of new market data, best practice guides and trends research.