One of the overriding theme of this analytics conference is using data to back up assumptions.
eBay did a very good presentation this morning about how their marketing teams research their customer-base through focus groups, usability studies, lab testing and even home visits. They actually send teams into people’s houses to watch eBayers at work in their own environment and in their own time. This kind of approach might yield a small sample, they’re never going to be able to canvas all 276m buyers and sellers, but it gives them something to go on when assessing a new feature, bell or whistle.
They then back up these what they find through detailed analytics and testing, processing many petabytes (!) of data in doing so.
The secret sauce seems to be a combined approach to improving the user experience and in turn increasing ROI and profitability.
You have to ask questions of the data you collect, but sometimes it’s difficult to know where to start. By asking your customers via traditional methods like surveys and suchlike, they can provide you with the ideas of how to interrogate and navigate your analytics reports to discover patterns which you can then work on to optimise or eradicate.