The idea of testing a web map isn’t promoted enough in the geo space. Testing is the heart of creating beautiful web map applications - where all stakeholders are able to act on data and have a better user experience.
But how do you actually run a good test on your web map application? Is it just a matter of trying a different zoom level, or color of the menu bar? It depends: Did you decide to test your hypothesis based on analytics or is it just a total guess?
To understand whether or not your tests are making a difference, you'll first want to decide what the goals are. Goals are unique to you, to your team and to the desired user experience of your web map. But a few industry standard goals are based around conversions like, activities/visitor, total engagement and bounce rates. These are key metrics that are a good starting point.
Then, you'll have to test ideas based on those goals. To come up with a test hypothesis, stay away from the random “I wonder ifs”. A good hypothesis should be trying to answer a question to a problem or an idea that seem unusual or have a large probability to increase conversions.
For example, if you notice that your map has a 25% bounce rate on desktops but 45% bounce rate on mobile, than there may be an issue with your mobile experience. Create an experiment that would shed light on why the bounce rate discrepancy from desktop to mobile is so high. Or if you're experiencing low activities/visitor numbers, perhaps users are hitting a blocker in terms of messaging, color schemes or the discoverability of your web map.
Go for simple and big at first
You can’t test every idea at once. So start with the hypotheses that will be easy to test and make the biggest potential impact. The struggle with testing is you’ll be faced to make a choice on which hypothesis to test that compete on big ideas and smaller ideas, that each could have a potential for big impact. Consider taking more time to test a big idea first. It may help to think in terms of a speed-versus-impact grid.
Trying to make little changes to your map typically isn’t going to have a massive impact on the user experience, but large changes can have a drastic change on how users interact with your map.
So instead of just changing a simple button your map, try using a different template, or theme for your map - always working backwards from your goal. The way we consume data varies based on the way its displayed. For example, if you have data points for 100 active real estate listings, you may want to test displaying thumbnail images of the listing on the left or right sidebar, as well as having markers on a map.
Try a lot of small ideas after
Once you’ve tested larger hypothesis, it’s now time to try small ideas. Think about all the ways you could test your hypothesis. Be small-c create: You don’t have to re-invent the whole wheel when testing small ideas, for instance, you should be willing to test some new ideas that are unusual. Switching your buttons from one color to another, or changing the default zoom level may be worth testing.
Ensure that you are referring back to your metrics to see how it’s affecting visitor interactions. Things like, activities/visitor and total time spent on map will tell you how engaged your visitors are.
When in doubt, keep it simple. You’ll be surprised how much difference a small tweak can make.