Measuring non-transactional websites – Part 1

This article, written by Neil Mason, was originally published on Clickz.com and is republished here with permission.

Last week I was delivering one a workshop on measuring web effectiveness to a mixed group of people from different business sectors. One of the things that I have observed having done these workshops over a period of time is that (at least here in the UK) peoples expectations are changing and that the breadth of existing knowledge of people coming on the courses is getting wider. Not so long ago, most people attending the workshop would all be pretty much starting from the same place and the most stated reason why people were coming on the workshop was because they wanted to know what “web analytics” was all about and how to get started in it.

These days the audiences tend to be a bit more diverse in their objectives and I’m beginning to find, in general, that people coming onto the workshop from organisations that have some sort of online business model tend to have more pre-existing knowledge than those that come from organisations where their online channel don’t have any transactional elements to them. This is of course, a gross generalisation but it’s not without some foundation!

So, the people with more e-commerce backgrounds are generally coming to the workshops saying “We’ve been doing web analytics for a while now and I want to understand how to take it to the next level”, whereas delegates from the public sector or non-for-profit organisations tend to be looking for the fundamentals. This set me thinking about why this might be and indeed whether we needed to differentiate the workshops in the future to meet the needs of these different groups. How would I change the content of the workshops between “e-commerce” delegates and “non-transactional” delegates?

The conclusion I am coming to is that the courses would look quite different. I don’t think it would just be a case of tweaking some of the sections here and there, there would probably need to be a change of structure and emphasis because the tools you need and the way that you would use them are very different.

The reason why the e-commerce type of delegates tend to be further ahead on the curve is because a lot of the reporting functionality in typical web analytics systems tend to be orientated around defined outcomes. Functionality such as campaign tracking, conversion funnels etc all work best when there is a defined conversion point, preferably with a value attached. When your core goal is to provide quality information rather than sell something, then these features have less utility.

There has been some discussion recently on web analytic forums and in the blogs about measuring non-transactional websites using web analytics tools. Increasingly people talk about the notion of “visitor engagement” and how to measure it. Various methodologies and algorithms have been put forward for debate but some of the challenges are around the complexity of the measurement and the ability to collect some of the data in the first place. I think one of the major problems is that just recording what was clicked and when it was clicked just doesn’t give informational or content sites what they really want to know, which is usually whether the visitor was able to find what they were looking for and was it any good. We can measure how long people hang around for example it’s difficult to interpret whether an average time on site of “x” minutes is good or bad and what, if anything, to do about it. We need more insight than that.

That insight is going to come from other data sources and from surveys and visitor feedback mechanisms in particular. As well as knowing what was clicked and when it was clicked, we need to know who was visiting, why they were visiting, whether they found what they were looking for and whether what they found was any good and gave them the information they were after. Trying getting that lot out of a web analytics tool!

So, I guess I need to get on with some course design work. At the moment I probably spend about 60% of my time on a workshop talking about site analytics and the rest of the time talking about other tools such as audience panels, performance data and surveys. I think for a workshop on measuring the effectiveness of non-transactional websites probably at least half of it needs to be on visitor profiling and feedback and the other half on the rest.

Next time I’ll share some thoughts on measuring non-transactional websites. Till then…