Meet me halfway

In my first job after graduating, I worked on our company’s website. I knew I was on to a winner when I found out that my manager was searching the web using AOL because she couldn’t change her homepage. Of course, understanding the internet was a tiny fraction of her job. She was too busy working out how to grow an organization on three continents.

The irony was that, even though she had a much better idea than me of the direction in which the organisation should be headed, she would often defer to my “expertise” in web metrics.  Little did she know that I was one of the people Katie Delahaye Paine describes in “How idiots track success“. In my defence, I was young and naive.

As a former manager, I know that leaders often make decisions based on incomplete information; asking the people with expertise for their opinion and combining it with an understanding of where the library should be headed. Here’s the problem with this approach when it comes to library websites:

The internet is magic

Not only is the internet magic, but we suspect our “digital native” users have magical powers and don’t need our old-fashioned library way of doing things any more. Unless you are a web professional, the chances are you have a less  than firm grasp of how your users find information and what you should do about it. And you might just about have a handle on what you “reckon” about the web when a web manager comes along and shakes what little confidence you have by asking you what you think of a whole bunch of numbers describing the situation!

With this in mind, we’d be just as well off asking our decision makers to rummage through fish guts to interpret user behavior as ask them to use web metrics. They might poke at them a bit but they won’t like it.

This is a shame; as we try and change how we use assessment to improve our services, our website is a hugely useful source of information but the people in charge of those services don’t have the time to study web metrics for the sake of it. Meanwhile, our web managers get the metrics but no one can tell them what we want. In my last update, we looked at using comprehensible numbers to enable decision makers to take an informed point of view. How would we apply this idea to our use of the library website as an assessment tool?

Avinash Kaushik, author and  digital marketing evangelist, gives us a really good place to start. He describes four attributes of great metrics. They should be “uncomplex, relevant, timely and instantly useful”. I would add that we should start small and keep within each decision maker’s area of influence and expertise. For example, if I was the person responsible for our computer labs I’m not sure what I would do with a report telling me the number of site visits or time on site for each user, there’s nothing that I can grab on to and use. However, I could look at:

  • The bounce rate; why are people leaving this page instantly? Are there any keywords or referring pages which give a really high bounce rate?
  • The number of people trying to schedule a room; how many abandoned the site before completion? Is the booking software too complex? Are the rooms fully booked at key times?

There’s a whole bunch of stuff I could look at and different things I could try but the most important thing is that if I am a manager starting from a zero base of knowledge, I need to find the numbers instantly comprehensible and useful or I’ll go on to one of the dozen other things on my to-do list. But as I build up my experience of using these metrics, they become less like magic and more like something I can have a halfway decent a conversation about.

Photo by Kevin Trotman


Leave a Reply

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

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

Google photo

You are commenting using your Google 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 )

Connecting to %s