Last week the IPA and Peter Field released fresh research on the effectiveness of brand purpose. They analysed 47 case studies from the IPA databank with a brand purpose and 333 without.
In general, the trade press interpreted the report as supportive of brand purpose as a strategy.
The headline from the IPA’s press release set the tone:
"Blanket criticisms of brand purpose advertising are unjustified, according to effectiveness expert Peter Field, who outlined his case for the potential power of brand purpose campaigns."
Beyond the headlines
At first glance, the headlines seem to reflect the findings. Most of the IPA’s analysis is on the performance of what it calls "strong" purpose campaigns. The research defines "strong" campaigns as "those that achieved at least one very large improvement across [a] basket of six business metrics". They represent a subset of all the purpose campaigns, roughly the top-performing half.
The research then outlines all the ways in which "strong" purpose campaigns outperform non-purpose ones: they’re more likely to drive very large business effects, brand effects and market share (see graph below). Quite the achievement.
But stop for a second and think about the statistical jiggery-pokery that’s going on.
The report is comparing a subset of purpose campaigns – only the best ones – with all the non-purpose campaigns. Unsurprisingly, these perform better. How could they not? After all, the weakest campaigns have been edited out. It's circular logic.
If that's all a bit abstract, imagine a trade body wants to prove that it is more effective than TV. So, it decides to analyse the impact of its campaigns and compare them with the rival medium. Unfortunately, the raw data shows it performs far worse. Undeterred, it creates a subset of its best-performing campaigns and compares that with all TV ads – the rubbish, the mediocre and the brilliant. Finally, it gets the desired results.
If it showed that research to you as proof of its higher effectiveness, would you give it any credibility?
You can't come to any meaningful conclusions if you fail to compare like with like. The analysis should focus on comparing all the purpose campaigns with all the non-purpose ones. Or if you’re interested in the best-in-class, contrast the "strong" purpose campaigns with the "strong" non-purpose ones.
To be fair, the report does compare one of those elements – it provides a contrast of all-purpose and non-purpose ones. And sure enough, here were get a different story from the headline. That comparison finds that on average, brand purpose campaigns generate 1.1 very large business effects, whereas those without a purpose generate 1.6. That’s a 45% variance in performance in favour of those brands not harnessing a purpose.
What’s perplexing is how little attention this key point gets. In a 798-word press release, the IPA allocates just three sentences to this finding.
Why mention it in passing? Shouldn’t the lower average performance of purpose campaigns be the headline?
Think back to previous IPA reports. The publicised findings have been the average effects.
Emotional campaigns trump rational ones.
Broadly targeted campaigns beat narrower ones.
They didn’t launch the fantastic Long and Short of It by saying that "strong" rational campaigns are superior to emotional campaigns.
A role for purpose?
All this is not to say that brand purpose, in the right context, shouldn’t be an option to consider. Brands come in all different shapes and sizes, with different histories and target audiences. So, it would be foolhardy to write off any tactic completely. As marketers, we should adapt our advice according to the task at hand and that may include brand purpose.
For example, the report shows that even "weak" brand purpose campaigns are more effective than average non-purpose ones in terms of boosting supplier and investor relations and employee satisfaction. If you are prioritising these objectives, then purpose might be a sensible approach.
And, finally, this article isn’t an argument for unethical business behaviour. Brands should pay a fair level of tax, avoid polluting unnecessarily, and support their communities. Of course. But just because you undertake good deeds should you broadcast them?
The IPA report takes the firm position that you should. Field states: "[Businesses] don’t just want to do something, they want to be seen to be doing something. As a marketer you would have to be rather dense not to take that on."
But, at the risk of appearing dense, just because businesses are keen to trumpet their ethics that doesn’t mean they should. Surely our job is to prioritise what’s effective over what gives us a personal glow of satisfaction? If so, best to keep on undertaking the good deeds you think are necessary but perhaps don’t broadcast them.
Richard Shotton is an author, consultant, conference speaker, and trainer
Will Hanmer-Lloyd is head of behavioural planning at Total Media