Should we implement a map feature or not? Building an analytics platform
We are building an analytics platform for our system, now focusing on introducing new metrics or reports. We wanted to introduce a map that shows visits to the websites of our Users. However, what we can do in terms of time and development, we can only afford to create a map grouped by countries. However, this will prove the most useful to those Users who have their customers coming from several countries, which mostly is not the case. Most of our Users have their activities in 1.8 countries. What should be the decision here?
- Develop the basic map showing only traffic grouped by country - the map looks cool, however, it will not tell us anything useful. On the other hand, it is placed at the very bottom of the interface, so even if it's not useful, it's not cluttering the interface very much.
- Wait with the development until version 2.0, with the map that allows grouping by country, city, region, etc. - we introduce it in another iteration, use our efforts towards building a proper map mechanism. The map is cool
- Don't develop it at all. Not worth the time and effort.
jeanid last edited by
It's a very common misconception that geographic data must be displayed on a map. The real question is whether the insights that this data is supposed to provide are indeed of a geographical nature, where the actual location of the sites on the map, and their spatial relationship with each other, are meaningful.
For example, if I've launched a marketing campaign that targets a specific metro area, and I see that my sales get a dramatic improvement in the center of that area and they gradually decline as we get further away from it - that's something that can most effectively be seen on a map. If I just get a table containing many dozens of these locations, with the sales data, it will be super difficult for me to notice this trend. I've made a geographical decision and I need to measure and validate its effect geographically.
On the other hand, if I run a campaign targeted at programmers and I get a spike from the big tech hubs - places that are known to me - it will be at least as easy for me to spot this trend in a table as it is on a map, and probably much easier (because it doesn't rely on my knowledge of geography and doesn't provide lots of irrelevant info). So that information, although it technically has to do with locations, isn't really of a geographical nature.
You need to understand whether your use cases require high geographical awareness, and prioritize the importance of the map based on that.