As a big, diverse, service-focused organisation, we have a lot of data. We work with very vulnerable people and deal with illegal substances, so a lot of this data is highly sensitive and confidential. We regularly work with people over a long time, sometimes years, so we can see patterns in what works, how our services change and what people need from us. This helps us — and the people who use our services — keep learning. But so often this data is caught up in clunky systems and is hard for people to make sense of. We’ve started asking what we can do to make data more useful to everyone.
At the end of last year we were privileged to work with DataKind UK, a network of data scientists who support charities to explore and get more from their data. Over a weekend we worked collaboratively to ‘dive’ into a set of anonymised data with a group of around 30 volunteers. The weekend was exhilarating, exciting, interesting and tiring in equal parts. I am still blown away that people would give up so much of their free time to help.
We gained some really key insights into our data, and took away some really clear themes. Some things confirmed what we thought we knew based on our experience — such as the importance of first impressions and strong relationships. Other times we busted myths. We’re now spreading the messages across the organisation and translating this into action through changing how we design and deliver services, and how we measure our outcomes.
We produced over a hundred pages of charts and information across the weekend. We then grouped these into broader areas, such as factors affecting drop out, length of time people are with us and the effect of multiple visits to our services.
For example, the chart below shows that across many years the length of time it takes people to leave our services successfully is reducing.
But for me the insights we gained into the data isn’t the thing I keep coming back to.
It was the first time I experienced a group of people all using a dataset with the sole purpose of understanding it. There was no expectation that we would find out anything, no assumptions, no need to turn it into a report that evidence how we use funding or compare us to another service. We weren’t constrained by any of the things that ordinarily we have to consider when we pull data together.
Of course, we’ve used our data in the past to inform our decision making but not on this scale and not with this freedom. It’s really easy to get into a cycle where we are ‘feeding a beast’ and not take the time to step back and really use the data we have at our disposal to make real changes for people who use our services.
With the right set of data, the right skills and the right tools you don’t need to have one right question — you need lots of questions and you need to keep asking them. This means in some cases leaving your assumptions at the door and changing your mind, it means using the evidence to make decisions but also understanding that all the answers don’t lie in the data (and being brave enough to say this).
Now, we’re recruiting three new roles to our Data team to help us build on this work: a Data Operations and Product Manager, a Lead Data Scientist and a Data Engineer. I’m really looking forward to welcoming new people to our existing team to help take forward our data strategy.
We’ve thought a lot about how we can do this and are open to change. We know:
- We need to build tools and develop products so that everyone can ask questions quickly at the time they think of them and without needing a lot of background detail. Our data is valuable and it should be shared across our organisation (and beyond), not limited to a specialist group of staff.
- We need to find a way to change the way we work so that staff have the space to ask multiple questions and not worry that they’ve wasted time or don’t have a solid ‘thing’ at the end. For me, this questioning is at the core of changing how we think about our data.
It’s early days but it we’re on an exciting journey. If you’re interested and would like to talk more, please get in touch with me at firstname.lastname@example.org.