Optimisation isn't just about web, test your app too!

Garret Cunningham

How to allocate e-commerce resources efficiently

As usage of mobile devices has become ubiquitous, so has usage of mobile apps. This has resulted in a debate among e-commerce businesses about whether they should concentrate the bulk of their resources on their apps or their mobile websites?

Many choose both, creating a companion app to accompany their website. This can be wise, but there’s a common pitfall you should avoid with this strategy.

Different user behaviours

The biggest benefit of mobile apps for e-commerce retailers is that they tend to convert really well, especially compared to mobile web. If users take the time to download an app and give it dedicated space on their phone screen, they tend to be among the most loyal customers and the most likely to make a purchase.

In contrast, people tend to pick up their phones and surf the web randomly all day long. They aren’t necessarily shopping when they land on your site so they might not be ready to make a purchase decision. This is why e-commerce businesses usually see inflated session and user figures but lower conversion rates and few transactions on their mobile websites compared to their apps.

The pitfall occurs when businesses assume that since apps convert better than websites, they should convert some of their app features to their mobile web. However, this isn’t necessarily the case. You must perform testing and research to better understand how users are interacting with your app and mobile web to make educated decisions about the best features for each.

Viewing products, not an Instagram story

A UK online retailer created a very attractive mobile app that included some really cool features that made using it feel more like a social media experience than a shopping experience. The percentage of overall sales generated from the app increased from 10% to 30% soon after they launched it, so they naturally wanted to incorporate some of these features on their mobile web.

During testing, we discovered that while users were interacting with the features on mobile web, they didn’t really understand what they were or how they were supposed to use them. They wanted to view products, not an Instagram story. So while there were high interaction rates on mobile web, the conversion rate actually fell by 1%, which would have resulted in a £200,000 loss over 12 months.

Making app changes is more difficult

Keep in mind that making changes to mobile apps is usually much more complicated and time-consuming than making changes to mobile websites. Testing platforms allow changes to be made to mobile web in real time quickly and easily. But with apps, the developer must submit changes to the App Store or Google Play where they are reviewed and approved. Then users must download the new app. So it could be days, weeks or longer before users are actually using a new mobile app.

We performed research for a UK online retailer that wanted to understand some of the challenges its customers were facing in using their mobile app. Based on the research, we came up with several different tests to perform. But instead of us making changes on a testing platform like we would with a mobile website, the app developer had to make the changes. This was a longer, more complicated and more time-consuming process, but it ended up being worthwhile.

In one of the tests, we made product descriptions more prominent and increased their visibility on the app because users were having difficulty finding them, and we also eliminated a click that was required to get to them. This boosted the conversion rate by 1.7%, which increased revenue by £500,000 over 12 months.

Don’t make faulty assumptions

Users exhibit different types of behaviour on mobile apps than they do on mobile web, so e-commerce businesses can’t assume that the success of app features will translate to mobile web. Testing is the best way to determine which features work best on apps, on mobile web and on both.

Download this PDF here