Once you’ve gleaned early learnings from your standard metrics, don’t hesitate to look further. When studying a set of data, certain cross-calculations and correlations can generate surprising insights that are completely unexpected but incredibly valuable.
Before planning your A/B tests, examine your digital analytics data. In doing so, you’ll be able to determine which elements to test first, or even clearly define hypotheses and objectives for your A/B tests.
Be mindful in how you structure and lay out your reports and dashboards. Above all else, put yourself in the shoes of the person who will be viewing this data.
You know you must communicate your data in the clearest, most comprehensible way. The problem? Not all your colleagues are analytics specialists.
Start out simply. Examine standard indicators around technical compatibility, broken links, sales opportunities, searches using your internal search tool that return no results, etc… These basic indicators alone will give you interesting information to work with.
Above all else, it’s the quality of your data that matters most. If your data is lacking in quality, your analyses will be skewed. When starting a new analytics project, focus first and foremost on accurate implementation.
Tag everything – all pages on your site, and all screens of your mobile apps – to avoid problems down the line with your data analysis due to missing or insufficient information.
One criterion that deserves extra attention when selecting your analytics tool is cost. Even if your analytics tool is supposedly “free”, it can still end up being costly.
Certain decisions are much more pivotal than others – and your choice of digital analytics solution is one of them. Beyond just studying the functional scope, be sure to also examine aspects that are often overlooked when choosing a tool:
To celebrate AT Internet’s 20th anniversary, we’ll be posting our 20 Golden Rules of Digital Analytics: a digital analytics tip each day for the next 20 days.