Last year during this time I attended a meeting where we went through our 2012 stats. When they came to the library numbers, people began debating what certain reference stats mean. The problem was the library staff was ignored and the stat was down played (although we had done well). This is not a big deal in the larger picture, but it was a learning experience.
I thought to myself that we must figure a way to better control the numbers surrounding our work. We don’t need to cover over bad stats, but we need a more efficient way to show the work we do. And, we need to provide our customers with the information they need.
It took most of the year, but I started to come up with a way our library staff will use stats. We create a lot of data, and a lot of it is counting stats: Number of requests; number of ILL requests; Number of downloads on certain journals.
What we don’t do create is context for these statistics. This is the first step. Taken by themselves, data and statistics don’t mean anything. People don’t have the time – or the patience – to extrapolate meaning from counting statistics. If they do, they will do it wrong. Or, they do it in a way we don’t like. The bottom line: We have to do this for them. We must create value for our customers by showing them the meaning/context out of our statistics.
Spreadsheets: They are a solution and a problem. Spreadsheets are tools created to organize data. They are not communication tools. What we need to do is communicate our statistics, but we have to move the data off the spreadsheets.
The goal is to package information in ways your customers can understand. This helps people make decisions. Like any organization there's lots of narratives on why things are done and why things are not done. However, we need to give our customers the ability to drive decision making with data. Not stories. And, if we create this information well, our customers can pass it on to other people.
The idea is that we can do this using the tools we already have. That means writing a textual summary – a narrative -- to our statistics. We also need visuals, like graphs, which are easy to do in MS Excel (for simple graphs) and even nicer in Illustrator (for more complicated visuals). You can be very versatile and creative in this area.
The narrative. It needs to highlight an aspect of your statistics in a way that people understand. It can be just a few sentences. If someone is looking at your stats to help them cut a service or a product: Give them an analysis. If it’s monthly stats, give them something that stands out to you. What runs against the grain? Our customers expect us to know these stats. That fact that you’re doing the stats makes you the expert. They want us to analyze the statistics and highlight the important stuff for them. The idea is to let them make better decisions.
The narrative must have appeal to create impact. You also have to know the audience and that should be your focus. Stakeholders have different needs. Think of scope. Some people need more granular information. Some don’t. Generally, the higher up you are in the org chart, the less detailed the information you want. Also, watch the language. Traditional language in the discipline is not understood by those outside it.
We don’t need to cook the books. We have a variety of output measures.
We also have to remember that different people may take your work differently.
The bottom line: Counting stats only go so far. We have to add value to our customers by providing context to our statistics
These ideas are mostly stolen from a great talk at the 2011 Electronic Resources and Libraries conference in Austin, Texas. The speakers were:
- Jamene Brooks-Kiefer at Kansas State University
- John McDonald at Claremont University in Austin, TX.
- Michael Levine-Clark at the University of Denver
You can watch their presentations here: http://vimeo.com/27015920
Photo by Carbon Visuals