In our data science blog article we provide insights into our customer feedback analytics. This feedback does not disappear into nirvana – instead it is collected, automatically analyzed and distributed within the company to relevant stakeholders. Product managers and developers – to name a few – use the processed feedback to make decisions for future product releases.
At Avira we try to use the terms data-driven and customer centric not as buzzwords, but really execute on these two concepts. The automatic key driver analysis for customer feedback is one example where we developed an end-to-end pipeline to provide a basis for decisions on data collected from customers. In general, a key driver analysis is the study of the relationships among many factors to identify the most important ones. In our case, we are interested in the relationship between the general customer satisfaction and specific joy and pain points customers experience during the usage of our products. Such an analysis results, for example, in an estimate on how much the decrease of the customer satisfaction score is, given that a customer complains about a problem with the automatic update of our Avira Antivirus product. If this decrease is significant to other complains, we know that this issue needs to be prioritized.
The Data Science @ Avira blog post “Continuous and automatic key driver analysis” by Manuel Eugster explains the end-to-end pipeline for the continuous and automatic key driver analysis in detail. It illustrates the infrastructure and discuss the statistical model used behind the scenes. This is work developed by the Customer Insights Research team.