"Reaction [beta]"
Five days bonus! M&S revisited 17 May 2006
Yesterday, Mark posted the following comment about our Marks & Spencer study:
"I love these sorts of studies, however, the fact that you aren't focusing on each website's target audience as your test audience, to me, means that the study can only talk in general design terms, and even then it's not that useful. For example, as a male, if I go to Marks and Spencer's' homepage and see a feature promo for lingerie then I'm probably going to look immediately at the navigation to see if anything at all in the site is likely to be relevant to me. Yet your conclusion is that since people didn't spend much time looking at the feature promo then Marks & Spencer are wasting (or using inefficiently) screenspace. Now that has to be a spurious conclusion. However, if you studied only women in the demographic being targeted by the feature promo then you would have more meaningful results..."
...A really interesting comment. However, we should point out that M&S targets both men and women, so I'm not sure that we were incorrect to test their homepage with a mixed user group. (Although it would be true to say that M&S's clothing arm caters more towards women, and that their homepage was women-focused at the time we tested it).
We should also say that we didn't conclude that "since people didn't spend much time looking at the feature promo then Marks & Spencer are wasting (or using inefficiently) screenspace". What we actually said was:
"What's striking from the heatmap is that almost all visual attention centres on the page header (containing the main navigation) and the left-hand menu. Great if M&S wants to connect users with products as quickly as possible, but not so great if they want them to engage with the features in the page's main area. It's worth noting that only one of our forty users actually clicked an item in the main body of the page..."
...Anyway, enough excuses. It just so happened that we were carrying out some further analysis on the five days eye tracking data when Mark posted his comment, so we thought we'd take a closer look at the issue of gender. Below are three heatmaps - the first showing the visual activity of only the women that participated in our study, the second showing the visual activity of only the men, and the third showing the combined view for both women and men.
Women only
See a full size version of the "Women only" heatmap
Men only
See a full size version of the "Men only" heatmap
The original heatmap (Men and Women combined view)
See a full size version of the "Combined" heatmap
As you can see, these findings are somewhat counterintuitive. The women were extremely focused on the navigation menus and rarely looked at the main body of the page. However, while the men were also predominantly navigation-focused, they were a lot more willing to venture into the main body of the page - even though the majority of the feature located therein targeted women!
It is often said that men are more visually-orientated than women - perhaps this adage goes some way towards explaining these results.
Next article: Shades of accessibility
Previous article: Five days: HMV.co.uk
Bookmark this page
Trackbacks
To create a TrackBack to this entry simply append ping/ to the permalink URL for this page.
3 comments so far
flipsockgrrl 18 May 2006 12:11 AM
The men were more interested in looking at pictures of women than in reading words! I'm shocked, shocked I tell you -- shocked that you are surprised by this behavior. I would expect usability consultants to take a slightly more sophisticated approach to analysing the results of behavioral studies.
First, in the context of these results, assuming that the gender difference is due to 'men being more visually oriented' is simply nonsensical -- this is an *eye*tracking study. It's all about vision. Unless one is using a screenreader (JAWS etc), it's awfully difficult to read a navigation menu without looking at it.
Second, in Western society we have ingrained from childhood the idea that women exist to be looked at and evaluated by men (see, oh, about two centuries of feminist writing for evidence of this). Of *course* there's a gender difference in reactions to the M&S home page. For men, the photos are something attractive and noticeable in an otherwise crowded and rather dull web page.
For women, the photos demonstrate how we should look in order to be acceptable to society. After a lifetime of being bombarded with such expectations, it's simply a matter of self-preservation that makes many women habitually avoid looking at photos of the impossible ideal.
Other than this quibble over interpretation of results, I really appreciate you making available the six heat maps from this study -- you've gained quite a bit of whuffie :-)
Etre 18 May 2006 10:37 AM
Flipsockgrrl: Thanks for your comments, you make some really interesting points.
We weren't surprised that men were interested in looking at the ads featuring women. What we were surprised by, is that M&S is targeting these ads at women, and that women are far less likely to look at them than men. This makes them almost completely impotent (as confirmed by the fact that only one person in forty clicked on them).
It's probably worth reiterating that the results we've been publishing are the preliminary findings of an ongoing study, and in no way constitute sophisticated analysis. There are many more statistics, heatmaps, etc that we could have chosen to publish, but didn't. Our aim was simply to get some eye tracking information out there - in a manner that is accessible to the layperson - because there's so little of it available outside of academia. Sparking some healthy debate along the way has been an added bonus. ;-)
Mark 25 May 2006 09:40 AM
Thanks for the response. I'd love to see this survey broken down into 'relevance' to the audience, intent to purchase and perceived value of content.
If you pre-selected a group of known M&S users how would their site usage differ to people who have no interest in M&S? And then if you cut the results down against people who use the site AND purchase online... or people who use the site AND purchase from the site.
If the site is relevant to the user I'd still like to find out what other behaviour they exhibit in deeper pages. Again - if they are familiar with the site maybe they know exactly where to go in the nav to get to the products that are relevant to them - in which case the nav is doing EXACTLY what it is supposed to do.
Any chance you can break some of this stuff down further? If you don't have details about M&S usage perhaps you could carry out a belated exit poll measuring relevance of the site, purchase intent (assumed) etc. I think when you combine this behavioural data with the eye tracking you will really be onto something. Otherwise it's too vague for a survey. And I say this with an interest in what you're doing from a personal and professional curiousity... I understand your point about this being a preliminary analysis - I'm just keen to push you to see what other data you collected because I love this stuff ;) Thanks again for making this available! Have you got a partner in Australia?