Why It Works: How Marmite took a weakness and turned it into a strength

When Richard Flintham and Andy McLeod, a young creative team, set to work for Marmite, they began by considering what Marmite meant to them. They quickly realised that Flintham loved it — and McLeod hated it.

A tagline was born, which they describe as a simple statement of fact. In an article for Creative Review, Flintham described ‘Love it or hate it’ as just a “truth”. All the campaign did, he said, was tap into the strong feelings that people already had about the product.

But it wasn’t an easy conversation to have with the Marmite team. Telling your client that a considerable number of potential customers despise their product is never going to be easy. Luckily, the brand was brave enough to run with it.

Why It Works: How two meerkats made insurance memorableBut actually, I would argue that it was less of a gamble than it might have felt. Because there are a few pointers from behavioural science that suggest the campaign was destined for glory.

And it was, of course, and even 30 years after its introduction, Marmite has stuck by its most distinctive brand asset. In fact, it became such a powerful campaign that divisive issues that crop up in conversation are often described as ‘Marmite’.

A flaw too obvious to ignore

Marmite is made from brewer’s yeast and — though I need hardly mention it — has a powerful, salty and highly distinctive savoury flavour. It repels a lot of people. And everyone knows that. Marmite decided to own that fact and place it at the forefront of their campaign.

The approach works due to a bias called the stolen thunder effect, which suggests that by admitting a flaw, you remove the power it has to cause damage.

Evidence comes from a 1993 study by Kipling Williams at the University of Toledo and his colleagues. They recruited 257 participants and asked them to read one of three versions of a criminal assault trial in which there was a piece of damaging evidence about the defendant. The accounts differed in just one aspect: sometimes the harmful evidence was brought up by the prosecution (thunder), and sometimes it had been mentioned by the defence lawyer first (stolen thunder), but importantly, not rebutted. Sometimes, this piece of evidence was left out altogether.

The researchers asked participants whether they thought the defendant was guilty. Have a look at the data below. The higher the score, the more guilty.

Trial Version	Mean perceptions of defendant’s guilt (scale 1-9)
No thunder (damaging evidence omitted)	5.04
Thunder (prosecution team uses damaging evidence)	6.61
Stolen thunder (defence team admits damaging evidence)	5.83

Adapted from Williams (1993)

Not surprisingly, participants considered the defendant less guilty when the damaging evidence was left out (no thunder).

Damaging evidence inevitably impacted the case when mentioned — but much less so when disclosed by the defence (guilt score of 5.83), rather than used by the prosecution (guilt score of 6.61).

Why is this? Maybe admitting a harmful fact acts as tangible proof of the honesty of the lawyer. So, any subsequent claims become more believable.

Why It Works: How Oasis definitely (maybe) applied behavioural science to ticket salesThis suggests that if Marmite were to simply leave the potentially repellent taste as a feature to be discovered and criticised among consumers, it would be held against them. Better to be up front about it and limit the damage.

Confessing a flaw also means that the other half of the claim — that some people love it — is more likely to be trusted.

But admitting a weakness works in another way.

Silly old me

There’s evidence that the simple act of acknowledging a shortcoming will help your brand seem not only more trustworthy, but also more appealing. The classic study showing this was led by Elliot Aronson at Harvard in 1966.

He recruited participants and played them one of several recordings of an actor answering quiz questions. The actor had been supplied with enough answers to achieve an impressive 92% score.

In one set up, after answering the quiz, the actor pretended to spill a cup of coffee (a small blunder, or pratfall). Participants were asked how likeable the contestant was.

Another group of respondents heard the recording but with no coffee blunder. Results showed that the clumsy contestant was 45% more likeable:

	Mean attraction score
Pratfall	30.2
No pratfall	20.8

Marmite becomes a more attractive brand by conceding that lots of people vehemently dislike it means, ironically, that more people buy it. Which translates into lasting success.

So, whichever side of the Marmite fence you land on — whether you’re a fan like Flintham or you’re more of a hater like McLeod — I’d argue that you have to love their behavioural science-based creative approach.

Richard Shotton is the founder of Astroten and author of The Illusion of Choice, a book about applying behavioural science to marketing.