Meaningful metrics and visualisation for libraries – a workshop overview from Pi and Mash

Our recent workshop at the Pi and Mash unconference, held at Senate House Library in August, came about because we wanted to start a discussion about how we might make the most of user and collection data to help deliver library services. How can we ensure that we’re making data-driven decisions about services and not just reflecting a narrow view of what a library should do?

Library and information services have never been richer in terms of the amount of user and collection data available. However, separating the wheat from the chaff can be a difficult and arduous task. We’ve found lean analytics approaches to be useful for focusing on metrics that are relevant and, more importantly, actionable in a way that can potentially help shape the services a library provides.

 

A lot of analytics and related guidance is aimed at the business sector specifically, and hence does not translate particularly well to situations where goals are not profit-driven – the person who comes to a library every day to study exhibits a valued level of engagement to us that won’t necessarily be valued by the local coffee shop.

In addition, a lot of qualitative *and* quantitative data is collected in libraries, including web analytics, social media interaction, visitor and issue figures, collection usage data and publisher statistics. Qualitative data collection measures are also available to us, such as feedback forms, surveys and questionnaires, and observation.

Lean methodology promotes the idea of the ‘One Metric That Matters’. This, of course, may not be the only metric that matters for everyone, but we’ve found it’s a useful way of focusing on a particular goal and breaking down metrics into a more manageable chunk. On this basis, and knowing that our presentation time would be limited, yet the topic of metrics vast, Kate led our workshop with a focus on introducing tools relevant to data visualisation, as this (sometimes deceptively) simple method is not only a way to make it easier to see what the data might be telling you, but it can also be used to make data and processes more accessible to co-workers, funders and library users.

Web Analytics tools like Google Analytics and Piwik provide useful reports and dashboards that can be configured to capture the metrics that matter most to your service (although it can take a bit of configuring to circumvent the built-in emphasis on ecommerce and adwords). Goals and advanced segmentation are two key ways that it’s possible to wrangle large datasets to a more manageable and relevant level.

There are also a huge amount of charting and visualisation tools available to the library sector, and it’s quite easy to fall down a rabbit hole of JavaScript charting libraries! From smaller charting libraries such as the Google Charts API and Dimple JS, to behemoths like d3.js (an amazing visualisation library but one with a steeper learning curve), it’s mostly a matter of selecting the right tool both for the job and for your own workflow preferences (be they ability or time-based etc.).

Photo by jessamyn on Flickr

As part of our workshop, we created a sample of code snippets from different JavaScript libraries for people to experiment with on JSfiddle. There’s also a comprehensive list of tools and resources which we started and which workshop participants collaborated on during our session, created using PiratePad.

Our workshop only really managed to scratch the surface of what is a huge topic, and this write-up is just a brief select insight to what our session covered, but it was great to discuss the ideas and implications of meaningful metrics with library professionals from a range of different sectors and backgrounds. Asking participants what reasons lay behind their analytics interest, we found that it was a pretty even 50/50 split between wanting to use library data to drive services and provision, and needing better ways of presenting data and statistics to line managers, funders and other assorted parties.

No doubt, with the volume of library data being generated and collected (in its many potential forms), the need for libraries to get with the meaningful metrics programme is on the cards – not least because data-driven service/provision and decision-making are just some of areas which may ultimately benefit our sector’s long term aims and objectives.

The full set of slides from our workshop are available here: https://artefacto.org.uk/metrics-workshop/ and do let us know of any awesome visualisations you end up producing should you put these tools to your library’s test!