From Data to Decisions
March 17, 2016
Marek Billington, Software Engineer
With the release of the latest Mohiomap update, there are many new features for our users to explore. Whether this is the new bookmarking tool for quickly getting back to your favorite maps, to the new dashboard view showing many insights into your documents and notes.
With the release of the latest Mohiomap update, there are many new features for our users to explore. Whether this is the new bookmarking tool for quickly getting back to your favorite maps, or the new dashboard view showing many insights into your documents and notes.
There are many new and different ways in which our users can utilize all these new features in Mohiomap. Current and new users will naturally end up experimenting with these new tools differently, finding their own purposes for each of them. One of my roles here in the Mohiomap team is the internal analysis of application usage, which informs our ongoing product improvement efforts. The purpose of this blog is to do a bit of Mohiomap “Inception” and explain a bit about how we use data visualizations to optimise data visualization.
Why, oh why?
We often ask ourselves “why?” - that is, for instance, “why do certain users find a specific functionality particularly useful?” or “why does a certain interaction often correlate with something seemingly unrelated?” As the first step with finding this “Why”, I usually start seeking out the high-level patterns that form within the different interactions that are made in the app. Then, by continually asking "Why" we can push further from those patterns into what they really mean, and how they are formed.
The objective always is to keep analyzing until we learn about the core value that is so important to our users. By piecing together these patterns and interactions we learn how our users manage Mohiomap, and it gives us the ability to go back and find or confirm the ways in which they use Mohiomap within their workflow.
We aggregate, anonymize, and collect our users’ actions on our app to gather user patterns. By analyzing these patterns from users across the different platforms that we support (Box, Dropbox, Evernote, and Google Drive), we can determine what features and use cases are most valuable to the respective user group.
“The patterns provide the information necessary to then make informed decisions.”
Following this, we process the anonymized data to gain insights. This is completed through the use of data visualization. Using the same frameworks that we use within Mohiomap to visualize data, we can quickly identify patterns and trends in the usage data. Whether the information comes out in the way you expected, or in some other surprising way, the patterns provide the information necessary to then make informed decisions.
That graph is pretty, now what?
As a next step we analyze the data visualizations, breaking down what can be seen from the different graphs and maps. Within massive clusters of points are the bits of data that all fit together nicely with nothing too special happening. However the outliers do what they do best which is stand out and make us ask, “Why are you over there?”. This then prompts us to delve deeper, refining the reason for why they stand out. Then restarting the process on more low-level format.
Getting to actionable insights
Through all this analysis, we are able to make well informed and actionable decisions. As a team, we are constantly gaining new insights into the many ways Mohiomap is used, and then proceed to make decisions about next steps and further improvements in the app. We have improved many aspects of the application this way - from small usability improvements to large and completely new features. The power of the visualizations that we use for our internal analysis helps in this - just the same way how Mohiomap helps our many users with the analysis of their documents, notes, files and deadlines.
That is why I look for the “Why” in the data by utilizing the existing Mohiomap visualization frameworks. By asking the right questions, we delve into the data, bring out the patterns, analyze them, and go from Data to Decision.