Not just a pretty face

Next time you walk through an airport duty free area, try and count the number of celebrities endorsing all sorts of things – watches, booze, perfume, more watches, sunglasses, fashion, handbags and vitamins.

There’s that athlete, that actor, with their piercing eyes and holding/wearing/being near the thing they’re selling. It’s as if they and the thing they’re endorsing are one.

You’re not just looking at another ad, you’re looking at a combination of behavioural science, educated guesswork and a lot of assumptions about what might influence your decision to buy or not to buy.

It’s said that for every dollar spent on advertising, fifty cents is wasted – they just don’t know which fifty cents. A recent study, The effectiveness of celebrity endorsements: a meta-analysis, looked at data from over ten thousand people to try and answer the question; do celebrities sell more stuff than anyone else? Yes and no, as it turns out.

Celebrity endorsements are on the rise. In the US, 20-25% of ads feature a well-known person. In Japan, the figure is closer to 70%. It must work, right? Until now, a meta-analysis hadn’t been conducted to see if they actually influence consumers’ responses. This study looked at the findings from over ten thousand people to finally get the numbers.

Advertising typically tries to influence us on three psychological levels; by giving us facts and knowledge about the product (cognitive), creating liking or preference for it (affective) and going that step further to stimulate desire for it (conative). It may not happen in that order (or at all), but celebrities are seen as a way of influencing each effect.

Celebrities are publicly recognised, so they draw attention. They’re also generally liked, which means consumers are more open to assessing what it is they’re endorsing and wanting to aspire to be “like them”. Ideally, the attributes of the object and the celebrity should be shared or at least consistent (classy, tasteful, expensive, etc). That’s why so few stars advertise low-end chainsaws, for instance.

Celebrity endorsements are designed to bring about a behaviour change – to purchase something, share information, volunteer, show support or even vote. But of course, it’s not as simple as “I, a famous person, use it. You should too!”

The study looked at a range of moderators of influence, including the sex of the celebrities, their type (eg. actor, musician, TV host), whether they matched the product, how explicit they were, how often the ads appeared and how familiar the product was.

‘nespresso georges clooney’ by Culture Culte via Flickr

Have a look at George’s image above for a quick guide to the results:

1. The celebrity should be genuinely recognisable to the public. Actors evoke stronger reactions than models, musicians or athletes.
2. The product should ‘fit’ the values the celebrity embodies. He’s not endorsing home brew beer, for instance.
3. The sell shouldn’t be explicit. He’s not licking his lips or closing his eyes in ecstasy, he’s holding a coffee.
4. The connection should be clear; we, as consumers, should make the desirable link between celebrity and product.

Male celebrities can evoke stronger effects than females, but then probably not when it comes to perfume or lingerie.

Actors seem to elicit stronger effects compared to models, musicians, athletes, or TV hosts. The product should match the celebrity (as in, don’t endorse guitars if you can’t play them), the celebrity should not be too explicit about their support for the product (so they’re not trying too hard), it should happen frequently (but not too frequently) and the product should be new or something the consumer doesn’t already have connections or strong associations with.

The study also pointed out that celebrity endorsements don’t perform as well as quality seals, awards or endorser brands – we still look for high standards, award-winning products and the backing of other trusted brands (like the Heart Foundation or Greenpeace).

So do celebrities sell stuff? Yes, they can, but it seems only when the right choices are made. The wrong celebrity matched with the wrong product trying too hard can backfire. A bad execution is never pleasant.

And other questions remain for further research – do we react as voters the same way we react as consumers? Can you sell a candidate like a car?

So as you fill in time before catching that next flight, or wander through a department store, take a moment to try and see if you’ve been influenced by the familiar famous face staring out from the glossy poster.