Enter a search term such as “mobile analytics” or browse our content using the filters above.
That’s not only a poor Scrabble score but we also couldn’t find any results matching
Check your spelling or try broadening your search.
Sorry about this, there is a problem with our search at the moment.
Please try again later.
When the news about Marks and Spencer’s sales results broke a couple of weeks ago it immediately got my attention.
The Chief Executive and other senior figureheads clearly laid the blame of the 8.1% drop in sales and resulting share price dip on the launch of its new website.
The new site comprised a smart redesign coupled with a platform shift from Amazon’s services to its own.
There have been many documented cases of website usability causing a huge impact to revenue (the $300m button being the most famous) so I wondered whether this too was one such example or whether more cynically perhaps, the City had been handed an unfortunate scapegoat.
The answer became all too clear after running tests using whatusersdo. We asked six users to perform some typical tasks on the site, finding two items of clothing, adding them to their basket and going part way through the purchase process. Half the users completed the test on a tablet and half used a desktop.
The tests soon revealed several usability and performance issues that ranged from those causing frustration and requiring a workaround to those of a critical nature that would either have been unrecoverable outside of a test scenario or cause the user to have left the site.
Critical performance issues
In terms of providing a good user experience in online retail and helping customers towards a purchase, preventing a user from getting to the checkout page has to be the most fundamental error.
Allowing them to get to the checkout but not to complete the selection of a delivery option is just as bad. Allowing items disappear from a customers shopping basket is just plain cruel.
Amazingly these tests uncovered all these scenarios, seemingly not a result of poor design but rather because of technical or performance problems.
The following clip shows a user who is unable to progress through the purchase process after having added two items to her bag.
She clicks on the checkout button several times but it simply reloads the shopping bag page. This user does eventually succeed after trying again via the home page but in the real world many others may not.
In this test, the user has opted to have her items delivered to her local store. She enters the town name in the search box (Bristol, so not a minor village) but no results display.
She even changes the name of the town to no avail.
This clip shows the user rightly become very confused that the dress she added to the basket earlier is no longer displaying in the basket. I could not work out any plausible reason for this.
If the item had become out of stock while she was completing the purchase then surely an error message could have been displayed.
Platform migration constraints
We know users do not take kindly to having to register and create an account and many retailers cater for this now by allowing people to checkout as a guest user.
In changing retail platforms, evidently M&S could not port across users’ account details and opted to require them to change their passwords.
As someone who forgets passwords on a shamefully regular basis, I did not even notice this request when it was made to me. However, the more attentive among the M&S customer base evidently did and were not impressed, as shown in the clip below in which the user has trouble receiving the password reset email which takes an overly long time.
She states she would have left the site had this not been a test situation and no doubt many others along with her.
Usability issues with filters
There was one particular filter category that M&S provide to narrow product selection that seems distinctly more trouble than it’s worth – the ‘Size’ category.
On selecting this either with or without another filter, users reasonably expect the items before them to be narrowed down to those available in the selected size. Instead of this though, the site still shows all the items but simply shows a small text indication, which most users miss, on the product page.
So, because the user is focusing on the product image and working out whether they like it or not, they click into the product details page only to find that it is out of stock.
Invariably this causes much frustration and confusion. This is not the only site that behaves this way and I’m at a loss to understand why. The behaviour of this filter is completely inconsistent with others.
For example if I choose colour blue, I’m only shown items in blue and all the pink ones are removed from my view. Why can't the size filter behave this way?
If the intention is to try and display as much stock as possible in the hope that the user will select perhaps a different colour of the out of stock item in their size then the indication must be made much clearer than it currently is.
There are other usability issues with the filters too, particular on a tablet. The filters seem to behave unpredictably, closing themselves just as you’re browsing through one of them. Also the text used for the category headings is very small (as is some of the other text on the site) on a tablet's default view.
Users found the interaction between the product ‘Quick View’ feature and the selected filters to be unintuitive. When they clicked the browser back button they found that their selected filters had all cleared. This could be very frustrating having spent the time narrowing down a large selection.
Users noticed a lack of detailed product information which they were looking for to support their buying decision. This may have prevented them from committing to the purchase.
Some users looked for more detailed information to explain what certain size categories meant, such as how long a regular dress would be. There was no explanation of this in the size guide.
Other sites also provide the model’s height to help gauge the fit of an item which would have been helpful here. Also some filter categories just seem plain wrong and a bug or two have clearly been missed. Is there really a difference between ‘Regular’ and ‘Standard’? Also ‘Petit’ and ‘Petite’, anyone?
Again tablet users experienced a poorer experience when it came to navigation.
Having chosen the ‘Women’ section from the main menu, they wondered what the difference was between the 'explore womenswear' and 'clothing' options. ‘Explore Womenswear’ actually displays more options for browsing by brand or style but I wonder if this heading is too subtle.
Users did comment on some positive aspects of the site, such as the slick look and feel and certain new features. The filters feature some interesting categories such as ‘Dresses under £40’ which is a handy shortcut as well as the usual occasion or style headings.
Also users can make use of a ‘Style Advisor’ feature. This wizard-like feature asks users to select various size, colouring and style criteria and in return it displays recommended items of clothing, saving these details for future visits and email updates.
Although it is a bit too longwinded, this could be an attractive feature to anyone who is blinded by the range of choice in most retailers.
However, it is a shame that this feature is powered by a third party (Dressipi) and not by Marks and Spencer themselves since account details are not linked up and it is necessary to enter email and name details separately.
I would like to know more about the issues above like whether rigorous usability testing was done well in advance of the launch, though it’s really not my place to pry.
What I can clearly see however is that this handful of users exposed many serious concerns which would have hindered purchasing. Multiply that by the customer base and there, perhaps, is your 8.1%.