Those of us that work in online marketing are very spoilt. With a well setup analytics solution we have access to 100’s of individual metrics about website/visitor interaction. We can use this information to identify opportunities for improvement and ultimately increase the revenue a website generates.

However, to really understand the huge array of measurements at our disposal we often need to take a step back from our screens and look beyond the numbers.

In fact, there are a number of metrics that simply cannot be put into context without real human brain power. Some examples:-

1) Average page views
2) Average time on site

Site owners typically consider these numbers to be an indication of positive website engagement. The higher these metrics are the better, right? Surely this indicates that your visitors are finding your site content of interest and are staying for an extended period as a result.

Now consider the following scenario:-

You have a site that sells rare coins, 1000’s of different varieties. Having looked at your analytics data you determine that each visitor views an average of 8 pages and spends 4 minutes on the site.

Great, you think. Your analytics agency has advised you that this indicates your visitors are engaging with your site.

BUT..

Your goal conversion rates are low.

Now add humans

You are concerned by the low conversion rates and decide to perform some live user testing. Some of the tasks involve finding a number of the sites rarer items. However, it is clear that the users being tested are struggling with the sites badly labelled navigation system.

All of a sudden those time on site and page view numbers don’t look as impressive. In the context of this example the objective would be to reduce these numbers.

You work with your design team, improve your navigation system and add a high quality search box. As a result average time on site and average page views halve but your conversion rate doubles.

My point

Web analytics solutions are a great source of information and can identify many of the problems a website may have. They provide the data but they will never do the analysis for you. You’re going to need humans for that!