A critical part of mobile application development is ensuring that your audience is able to and enjoys using the application. To make sure that you get it just right, it's common practice to conduct a few rounds of usability research to help you learn how users expect to use your app and how you can improve the user experience. In fact, Thou Shalt Not Skimp on Testing" is UX Testing Commandment #1 in Proto.io's 10 Commandments of Usability Testing for Mobile Apps.
Research shows that when mobile app developers allocate significant amounts of time to usability and research, their apps have higher download rates, higher engagement rates, and higher usage times. Research also shows the Usability, or how easy or difficult an app is to learn how to use and how user-friendly the app is, is the key to higher app engagement. Testing a mobile app for usability is critical to an app's success.
While some testing is better than no testing at all, collecting accurate enough data to make development decisions has proven to be difficult. The trouble with common usability tests is in the tests themselves. Generally speaking, a test is short (usually 15-30 minutes), extremely focused on one specific task or goal, and the user often has an incentive to make it to the end of the app's test. But is this how users in the real world would use your app?
Countless studies have proven that the answers your users give during usability testing can differ significantly from the way they would use your app in the real world. A well-known example is the Pepsi vs Coke taste tests. In the late 70's, Coca-Cola owned the majority of the market share for the
Wrong. The majority of people revolted against New Coke, and the company lost significant market share. Considered one of the biggest blunders in marketing history, New Coke was pulled from shelves within a few months of release, and "Coke Classic" came back to replace it. Sales showed that the population preferred Coke Classic. But Coca-Cola’s extensive research predicted the opposite. Why?
Turns out, when sampling a sip or two of cola in isolation with no other stimuli, people preferred the sweeter taste of Pepsi. But in real life, you don't drink .5 of an ounce of soda* and consider yourself refreshed. You generally have a can of Coke with a meal, around a campfire, or in a bar. You generally intend on drinking the full 12-ounce can or 20-ounce bottle. The method in which user research is conducted is just as important as the test subject.
This translates into app design and development by showing that the way a user engages with your app in a test environment might differ greatly from the way they would use it in real life. The manner that we present test subjects to be researched is just as important as the subject of the research, and developers must be conscious of this during each and every test.
Related Content: Making Hybrid App Development Feel Native
Additionally, most usability testing takes place while the app is still in development, meaning the app in the test often differs from the end product. Even the slightest changes in parts of your app can have an effect on the user's experience and/or their understanding of how to use your app. For example, if there are aspects of your app that are dependent on other elements, such as a drop-down that limits options or a geolocation-dependent feature, the user's experience is likely affected by these aspects and thus needs to be part of usability testing.
It's important for app developers to test each element and feature on their user's journey through the product. It's also important to continue usability testing throughout the development process. Allocate time to at each stage of development and version release, so you can ensure that the UX remains consistent throughout the entire process. Lastly, be sure to test each version of your app on as many device types and models as possible, starting with the oldest devices and operating systems available and working your way up to the latest versions.
Have you ever conducted market research that predicted one outcome, but real-world data proved it to be inaccurate? Share with us on our social channels below!
*there's actually legit research on the usage of soda vs pop vs soda pop vs whatever-else-anyone-calls-it. Check it out on Huff Post.