For companies that treat their web sites as another piece of marketing collateral, creating a good looking site is their end-game. For anyone that treats the web as an efficient and effective sales channel, managing their online presence is a continual cycle in customer experience management.

The cycle of create/monitor/refine relies on a three-pronged approach of creating a site, monitoring its use and refining the experience to meet new demands. To measure a sites effectiveness, one method is to use specialist analytics software, either open source, hosted (for example, Google Analytics) or proprietary (such as Webtrends).

Such tools capture information about web site behaviour, such as who has visited, when they clicked and how long they spent at each destination. The information can be extremely helpful for organisations hoping to develop targeted marketing, interactive advertising and search optimisation.

But creating honed behavioural analytics can be a complicated task. Each page needs to be “hooked in”, so that pages are tagged and individual hits and user behaviour recorded.

If behavioural analytics tell you “what the user did” on your site, quite often it is the transactional data (data they have entered into online forms) that tells you “why the user behaved” in a certain way.

However transactional data is normally held separately in an organisation MI Database and is only captured if the user completes a process – for example, receiving a home insurance quote. However, to find out why users don’t get to the end of a process firms have to undertake separate programming to capture the transactional data. Such programming helps to ensure that transactional information is recorded in databases and the intelligence passed to an analytics engine. But once again, such a process is complex and time-intensive.

So, here’s an idea – wouldn’t it be great if you could easily combine behavioural and transactional data? Such an approach would really create 1+1=3, as web designers and business experts can analyse and tweak customer experiences based on intelligence gathered from actual usage.

Evidence comes from our work at edge IPK. We worked alongside a large financial organisation, analysing behavioural data to identify the most common page that users left without completing an insurance quotation form. From the behavioural information, we knew that most users not completing a car insurance quote left on the page where they would enter their car details. The additional transactional data told us that users would select a car maker, but would leave before selecting a car model.

We quickly concluded that users were leaving at this point because they could not find the model of car they wanted to insure. The result was an increase in quotes and, more importantly, an increase in sales.

Happy customers and new business. Isn’t it time you tweaked your user experience analytics to ensure all potential metrics are covered?