Image showing Lang, Pulse and robot Iadu conducting a user testing.

Qualitative and quantitative data to optimise the User Experience

20.06.16

By combining analytics and user testings, we can best optimise an existing app. We can identify weaknesses to then validate and optimise them.

UX reviews in the lifecycle of a product

From a business perspective, a good User Experience is often a matter of life or death for a digital product. For instance, if a product is too bloated, people will automatically look for a lighter alternative. Complex digital products come with a longterm lifecycle. Within this lifecycle, more and more features might be added. The application becomes bloated while the core experience suffers. The application needs a UX review.

It is part of our job to evaluate and improve the User Experience of existing applications. Considering both time and financial resources, we aim to revive the core experience of the application. To do so, we need both qualitative and quantitative analyses.

Qualitative and quantitative data as complements

Usually, we start with a qualitative approach, for instance with user interviews and workshops. With these, we can isolate a first set of weaknesses of the application. We then need quantitative data to validate those assumptions. Analytics data* can support or refute our initial insights. For instance, if we see that the exit rate on an order form is very high, then it makes sense to overhaul and optimise the form. Or we can identify if carefully compiled content is being read. If users for instance only spend five seconds on a page, the editorial efforts were all in vain.

After having learned more through the quantitative data, we come full circle and do qualitative user testings. We often let target users interact with a clickable prototype to obtain A/B feedback. Thus, we can ensure that the optimisations which we propose lead to decisive improvement and create added value. After that, we can start with the implementation.

* Which analytics tool to use depends on the project. For instance, we use Piwik in sensitive environments, since it allows you to keep ownership of your data.