Is it worth hiring David Beckham to promote your brand? A psychological test

He’s probably the most famous man on the planet, but the problem is there’s no chance of exclusivity. Beckham already endorses a string of products from the Emporio Armani fashion label to Burger King. So is it worth hiring him to endorse your product?

One way to answer this question is to look at the psychology behind the way people associate brands and celebrities in memory. If you pay Beckham to endorse your product in an ad, then you’d probably like to think that people who view the ad will, in the future, see Beckham in the paper or on TV and then think of your product. But how do you know that seeing him won’t lead customers to think of the other brands he endorses (Armani, Burger King, Gillette, Sharpie, H&M, Samsung, Adidas or Pepsi), rather than yours? This is the question that Katie Kelting and Dan Rice have addressed in their new study.

The researchers predicted that a key factor would be the degree of fit between Beckham and the products he endorses. They presented 235 undergrad students with 8 magazine style adverts. Six of these were fillers and included non-celebrity ads for an SUV and a brand of dog food. Two of the ads were critical to the study. One was the “target ad” showing Beckham alongside a digital camera (you could imagine this is the product we’ve hired him to promote). The other ad showed Beckham endorsing one of his other clients. Crucially, this “interfering ad” came in one of three versions. Some students saw Beckham endorsing an energy drink (a strong fit seeing as he’s a retired footballer); others saw him endorsing an MP3 player (a moderate fit); and others saw him endorsing a baseball bat (a weak fit).

Next, the students were distracted by a six-minute maths task. Finally came the crucial test. The students were asked to think back to the adverts they’d viewed a little earlier, shown a picture of Beckham, then asked to name the product he had endorsed earlier. Would they name the digital camera in the target ad?

As Kelting and Rice predicted, it all depended on the degree of fit between Beckham and the other brand he’d promoted. Among students who saw Beckham promoting the camera and an MP3 player (both deemed to be of moderate fit with Beckham’s image), 88 per cent recalled that he’d endorsed the camera. By contrast, for those who saw Beckham promoting the camera and either the energy drink or the baseball bat, far fewer recalled that he had endorsed the camera (59 per cent and 57 per cent, respectively). In other words, brands that had either a strong or weak connection with Beckham seemed to have a more powerful interfering effect on people’s memory for the target brand*.

This interpretation was supported by further analysis. Focusing instead on participants’ recall of the interfering brands, their recollection was stronger for the energy drinks and baseball bat than for the MP3 players. Comparing levels of recall for the interfering brands and the target brand (the camera), recall was again higher for the energy drinks and baseball bat. “A brand sharing either a high or low match with the celebrity endorser will win the battle of activation and retrieval over a brand that only has a moderate match with the celebrity endorser,” the researchers said.

With so many celebrities today endorsing multiple products, Kelting and Rice said their results have real-world implications for the marketing industry. On this point, one further finding is worth mentioning. The researchers also looked at attitudes towards the different ads, which is one of the most common ways that adverts are assessed in marketing. Attitudes were least positive for the ad showing Beckham endorsing a baseball bat. In a real-life situation, a marketing expert might assume that this ad is therefore not much of a threat to Beckham’s other endorsements. But in fact, as we’ve seen, this ad was one of the most memorable and interfering precisely because of the poor fit between Beckham’s profile and the product. “Making strategic decisions without considering the impact that other brands in a portfolio may have on recall could lead to suboptimal decision making,” the researchers said.


Katie Kelting, and Dan Hamilton Rice (2013). Should We Hire David Beckham to Endorse our Brand? Contextual Interference and Consumer Memory for Brands in a Celebrity’s Endorsement Portfolio. Psychology and Marketing DOI: 10.1037/e621072012-150

*note that these memory and interference effects were found only for cued recall (after prompting with a picture of Beckham), not in another part of the study in which participants engaged in free recall without any prompts.

Post written by Christian Jarrett (@psych_writer) for the BPS Research Digest.

4 thoughts on “Is it worth hiring David Beckham to promote your brand? A psychological test”

  1. It's interesting that the researchers considered an item of sports equipment to be a poor fit for a sportsman to endorse. Maybe my low familiarity with Beckham's image makes me see this differently from the typical advert viewer.

  2. hi Rachel – good point. They believed this was a weak fit because Beckham is so strongly associated with football. But you're absolutely right that people will vary on what they consider to be a good fit with a celebrity. A related point the researchers made is that the range of topics and products that people associate with a celebrity is likely very wide and constantly developing as people these days read so much about celebrities' lives and their personal interests.

  3. I'd agree with Rachel, I can see a stronger link between Beckham and the baseball bat than with the Mp3 player. Unless of course the advert draws out the exact qualities which is links him to the MP3 e.g. performing at the highest level.

    Also, I would think products with technical qualities (e.g. MP3) may be more suited to celebrities with a high level of perceived expertise in the area (e.g. Dr Dre's Beats) and unfortunately Beckham has a reputation for being slightly dim.

    Although a very interesting study! (I'm doing my dissertation on this subject).

  4. I see the intuitive linkage displayed in the hypothesis, but is this experiment not statistically invalid? To the law of small numbers suggests 235 undergrads is a pretty small sample…

Comments are closed.