What do computer games and digital analytics have in common?
Digital analytics and computer games have a lot in common. You wouldn't immediately think so, but they do. Both disciplines are about exploring new worlds, finding and conquering treasures, saving the princess, uncovering secrets, mastering tasks with an existing skillset (of the hero), solving puzzles, solving riddles and reaching new levels. It is obvious that so many parallels make us think of web analytics.
The data we collect within a digital customer facing platform like website, shop or app provide us with insights of the platform. Without data, we are blind. The image below shows what we see when we don't measure. Take a closer look! Do you see it? The nothing?
Computer game level map without digital analytics.
Some people will say that we can pull data from the backend! Yes, that's true. Of course we can. However, the silo data from the backend has nothing, absolutely nothing to do with ANALYSIS, but only with a status quo reporting. Not to mention a long delivery time, inflexible or no combination options with other data and a visually unappealing presentation.
But we want to discover new worlds, find treasures, conquer treasures, rescue the princess, play the hero, solve puzzles and riddles.
Standard Tracking
That's why we build Standard Tracking into our digital platforms. As soon as our Standard Tracking is included, we will be able to fulfil around 80% of stakeholders’ requirements and answer real business questions.
Based on the computer game level map, this would look something like this (see below).
Computer game level map with a standard tracking.
We understand where the users click through (travelling by ship), what the best trading routes are, where the ships dock and where they are robbed or sink completely.
We know the obstacles on the route, but we don't yet understand why exactly there is an obstacle. We have hypotheses that explain the “why”. But we don't know.
Custom Tracking
To build knowledge and eliminate the hypotheses, we need to implement Custom Tracking. By implementing Custom Tracking, we bring “light into the darkness” and expand our “level map”. The black spots on the map become smaller. As implementation of Custom Tracking is always a trade-off between effort and benefit, tracking is implemented in the most important areas first. The level map becomes clearer but still has black spots.
Computer game level map with extended tracking - Prio 1 Custom Tracking was implemented.
For example, we suddenly realise that we can't safely drive past a river because we (i.e. our users) are being shot at from the beach. Or that we can now safely dock on land and our passengers can enter certain “buildings”, but not all of them get out again. With the current tracking, we still don't know why this is the case. But that was a conscious decision.
We can now work with this new data knowledge and develop strategies and solutions for recognised problems together with the stakeholders (business & tech). These are transformed into more Custom Tracking and implemented, afterwards we can measure whether our strategy has worked with the next deployment.
This continues iteratively until we have fully illuminated the level map. Now you might think that we have reached our goal. But that's not the case! Because digital platforms are constantly evolving and building new level maps that need to be illuminated.
Computer game level map with fully developed tracking - all features are visualied with custom tracking.
Okay, nice. But how does all this fit into real life?
A real-life use case
We integrated standard tracking into a customer's payment app at its launch. After the launch, there was a lot of attention on the data, and we were able to answer many standard questions:
Process Abortions & exits
Avg. Session duration on different pages and in the main onboarding processes
Number of account openings
Development of macro and micro conversions over time
Dependencies on mobile devices
and much more
But only the standard questions. We saw that we had a problem with the ID scan process during onboarding, but we had no way of looking inside the “building” of the ID scan process. We therefore developed a Custom Tracking with the business stakeholders, which enabled us to see which types of IDs are scanned and put this into correlation of abortions and conversion rates
which areas of an ID are captured by the scanning process by ID type
which errors occurred during the scanning process by ID type
how often users have to scan to get through the process by ID type
At the same time, we expanded the tracking to include error messages so that we could better understand on which pages and with which input fields within the onboarding process users were having input problems.
These are just a few examples that gave us a better understanding of our users.
Digital analytics is a process. Not a project.
During analyses, we always start with the basic questions. While we analyze and try to answer business questions, we discover blind spots we did not think about before. Sometimes, we already have data available to illuminate them. Sometimes we need to proceed and update the tracking accordingly.
As a rule of thumb, we are constantly working on expanding the tracking. Because digital analytics is a process, not a project. Digital analytics should evolve continuously with the digital platform, and the platform should evolve continuously based on data. As soon as you start with data, you never stop.
Our goal is to always provide the right data foundation to enable truly data-driven decisions in all matters.
Comments