What do people actually do on your map?
It is likely you have a pretty solid idea of what you would like your users to be doing on your map. It is likely you want those users to investigate the subject matter layers you are presenting and marvel at the new information they have discovered from your work of cartographic web excellence.
Yes but, how do you know if they are?
What if your hard won users are getting distracted and are in fact looking for their grandmothers' house on the newly updated aerial image layer? It happens to the best of us. By only looking at your page loads on Google Analytics (or any other web analytics platform), it is very hard to determine the actual experience your user is having with you map product. To really understand what users are doing, you need to listen to their activities.
For our purposes here a 'map activity' is a pan, a zoom or a click. From looking at the patterns of these activities we can discern a great deal about a user's engagement in your map experience. Since we launched Maptiks we have been looking at user activities with great interest. We have recorded in the region of 9 million user activities over 750 thousand different map sessions. So we have developed some expertise at looking at how users interact with web maps.
We have noticed a number of key patterns:
1) The zero activity map load - we refer to this activity level as the Map Bounce Rate. This of course could be expected, or planned, but it might also indicate an area on your website where an interactive map could be replaced with a static map, saving significant money in some cases.
2) Small engagement map loads - This typically indicates the user is only partially compelled by the content, OR the user is being driven exactly to the correct piece of information immediately. This would be a desired outcome in terms of a store locator. The user wants to know something specific and the map's job is to provide that data, within a geographic context as quickly as possible.
3) Moderate engagement map loads - This would be the desired outcome if you are sharing a more information rich data experience, such as a realty map. You want the user to pan around and click. Interestingly, it is possible to start looking at user velocity when tracking this kind of engagement. "User velocity" being a measure of the speed and direction of a user as they travel through your map at any given time. The slower the velocity at a particular scale the greater attention that user is giving to that location.
Below is a map indicating a particular user's activity on a particular map. It shows a moderate level of engagement.
In this example it can be seen that the user moves quickly across North America, directly across Africa up to Asia, back south to Antarctic back to the North Atlantic then zooms in closely on Scotland.Their velocity remains high until they zoom in on Scotland. Although the zoom levels are not shown on this map it can be loosely indicated by the distance between each vertex on the line of user activity. Interestingly this user's behaviour pattern is somewhat circular, its almost a spiral.
4) High activity map loads - When activity levels typically get above 80 - 100 although this may sound like you have built THE best web map, there could be other more nefarious reasons for this, for instance:
In this case we can clearly see that the map's data is being programatically scraped.
These 4 patterns describe very general cases, the fact is the purpose of each hand written map is slightly different. They have different nuances, different delivery, different data, and different contexts all rolling up to a spectrum of different experiences.
Indeed, we can look at some particular user activity patterns that ring alarm bells too:
1) Just Scrolling North or Just Zooming In - both of these patterns indicate a user is having UI difficulty and the scroll wheel is not behaving in the expected manner.
2) All Users Zoom In an Arbitrary Amount - indicating that the initial zoom level of a map is not optimized.
3) Lots of activity in areas where there is no data - indicating you have lost the interest of your users.
Understanding how your users behave on your map provides an amazing looking-glass into human behaviour. Paying close attention to your user's activities will help increase map conversions, and thus your map's return on investment.