usage data is like a bikini?!?


Tony keeps talking about how the library homepage should be amended by looking at the usage data, and whilst usage data is important I’m not convinced that it’s the be all and end all of things.All this focus on usage data reminds me of the joke:

What’s the similarity between statistics and a bikini?

What they reveal is interesting, but they conceal the vital bits!

For one I’m not convinced by the Google Analytics click-through data as it seems to me that it only looks at the sub-domain of the links and so any clicks to a base url get amalgamated together. And just looking on the homepage there are the following base urls:  http://library.open.ac.uk; http://www.open.ac.uk; http://www.open2.net; http://intranet.open.ac.uk; http://voyager.open.ac.uk. So not being able to distinguish between people who leave the site to go to the catalogue and those who go to another institutional page or an external site makes it very difficult to then use this data as our primary decision-making tool.

Secondly, we get other feedback direct from our users that we need to take into account – either as a result of surveys and user testing, or unsolicited comments. These are often insightful and show what people are thinking of the site when they use it – what they are trying to do and what problems they have doing it.

Lastly, we within the Library have a view on how we think the website should be used and what we want to push to our users to help them use our resources. We may not be right, and that’s where site analytics are invaluable in seeing how the site is used, but they can’t be used in isolation.

Tony is right though: we do need to get better at using the data we gather to inform our decision-making. We gather alot of information (not just web analytics) and need to be more proactive in working out what we want from this and then using it. Tony’s been talking about using the website usage data as the basis for a re-design of the library homepage for over a year, and given how many staff I know read his blog it’s quite embarassing that it’s taken so long for us to do anything about it. But our new front page will be launched on the 1st Sept and the usage data (among other things) has been used to come up with the design.

Advertisements

2 thoughts on “usage data is like a bikini?!?

Add yours

  1. “whilst usage data is important I’m not convinced that it’s the be all and end all of things”
    I don’t think it is; you say “we get other feedback direct from our users that we need to take into account – either as a result of surveys and user testing, or unsolicited comments. These are often insightful and show what people are thinking of the site when they use it – what they are trying to do and what problems they have doing it.” Yep -they’re great ways of identifying user intention, invaluable even, at times. But I do think the mass action of tens of thousands of unique users is also something that should be considered, alongside online questionnaires, interviews and user testing with low numbers of people under contrived conditions.

    “it seems to me that it only looks at the sub-domain of the links and so any clicks to a base url get amalgamated together”
    Are you talking about tracking people who click a link to a page outside the domain that you’re tracking with Google Analytics? You can track these however you want (I know for a fact I’ve shared this type of howto with Library folk before: http://www.google.com/support/googleanalytics/bin/answer.py?hl=en&answer=55527 )

    (Also bear in mind the OU Site Intelligence analytics, which track usage across parts of open.ac.uk (I’m not sure which parts?))

    “we within the Library have a view on how we think the website should be used and what we want to push to our users to help them use our resources.”
    Last time i gave a presentation to the Library re: using analytics, I asked what the goals of the Library site. I also suggested that it would be sensible to identify questions about user behaviour that you could start to explore using appropriate analytics, stressing that the analytics are something you have a conversation with to get a better feeling for how the website is being used, but always bear in mind that you may be misreading the stats (NEVER trust a simple mean, for example…;-)

    I also suggested that named individuals take ownership of particular metrics/questions in order to understand how particular bits of the website are working for different audiences, based on audience segmentation, and seeing if that behaviour was as expected/as intended by the library. (eg I think I illustrated this by comparing usage of the website from people coming in from the College of Law, cf. somewhere else…) One possible goal here might actually be to better tune eg a journal collection for an audience, rather than e.g. specific website matters such as , I dunno, the possible need for a customised landing page for users from a particular referrer? (probably a bad idea, but it’s late…)

    “and the usage data (among other things) has been used to come up with the design”
    I’d be really rally interest to here how the data was used (and I think other folk would be too).

    eg here’s how the BBC have signaled some of their changes over time http://delicious.com/psychemedia/bbc+website+redesign

  2. You’re right Tony – we do need to be more holistic in how we look at all this data and know what we’re trying to get out. As well as having a better idea of what we want the data to tell us. We get so much data that is never really analysed, so we need to become much more smarter at using it. I wasn’t trying to downplay the importance of web statistics, just explain some of the other information that we get that has an impact on the design of the website.

    I would explain how the website usage data was used to come up with the re-design, except I wasn’t involved in producing it so I don’t know why the decisions were made, I’m just assured that it was used!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Up ↑

%d bloggers like this: