First Impressions are Hard to Change

Most people have heard the old saying, “First impressions last”. Most of us have also probably had an occasion when we’ve made a bad first impression and wished we could change it.

The psychological phenomenon of priming is reasonably well documented and also easy to demonstrate in certain contexts. Ask people to quickly estimate one of the following calculations (don’t show them the other) and you’ll get very different answers:

1 x 2 x 3 x 4 x 5 x 6 x 7 x 8 = ?

8 x 7 x 6 x 5 x 4 x 3 x 2 x 1 = ?

The order of words have been shown to have a similar affect. People give more weight to what they encounter first. In the case of the example above, people guess a much higher number for the second calculation than for the first.

But whilst we might be quite careful on a personal level when making a first impression, few people consider that brands or stores or websites are susceptible to exactly the same phenomenon.

From my own consumer behaviour work I’m very aware of how what people encounter first can influence how they perceive things subsequently. Store window displays and entrances set the tone for the way subsequent parts of the store are viewed. People who take a dislike to a brand rarely see any need to revise their opinion of them. If your first customer experience is a bad one, customers will be more sensitive to the next failing.

At an evolutionary level this makes perfect sense. We’re able to progress so quickly and achieve so much as a species because we develop general rules that work for us and then apply them; were we to constantly re-evaluate the accuracy of everything we wouldn’t get very much done.

Imagine doing your supermarket shop without automatic processes: you would have to stop and evaluate the alternative products every time you bought something. That might involve studying the price per weight cost, reading the nutritional information, examining the list of ingredients, perhaps even asking the people around you what they thought of the alternatives, or watching several other people make a purchase and copying what most of them choose. It would take forever!

Recently published research looked at the way first impressions are revised. Researchers primed participants with either positive or negative information about people shown on a screen. They then provided new information about the person, but subtly changed the background colour that was displayed.

When the researchers gauged the spontaneous reaction of people to the person they found that the new information they’d provided was only referenced, in other words people only revised the initially primed view, when the background colour matched the colour that had been shown with this new information.

Put another way, if you’ve initially decided someone is an idiot and they subsequently give an impressive talk on marketing, you will still think he’s an idiot except when he’s giving talks about marketing. In other circumstances, even if they don’t relate to the initial circumstances where you formed that impression, you will default to your negative view of him.

Similarly, if your first impression of a salesperson is poor, you are likely to form the view that the store has poor service. If you subsequently encounter someone there who provides you with excellent service it probably won’t matter; the damage has been done. You will form the unconsciously held impression that if you’re served by that person again you will be OK, but anyone else is likely to be similar to that first person.

I’m convinced from work I’ve conducted and from other studies that I’ve seen that the same applies to brands, stores and websites. Frequently retailers try and cram far too many products and far too much information into their store’s entrances on the basis that this is the highest traffic zone in the store. However, what they don’t usually consider is the impact this has on how their store is perceived.

If you think about your favourite websites, brands and store experiences, pay particular attention to what you see or remember seeing first. The chances are it says much about the impression you have more widely.


Sources: Kevin Hogan: The Science of Influence (priming example)
Bertram Gawronski, Robert J. Rydell, Bram Vervliet, Jan De Houwer. Generalization versus contextualization in automatic evaluation. Journal of Experimental Psychology: General, 2010; 139 (4): 683 DOI: 10.1037/a0020315

Image courtesy: Xavier Donat

Leave a Reply

Your email address will not be published. Required fields are marked *