Thanks in part to canny marketing by MHCLG, we had hundreds of people register for our session at Digital Leaders Week… but only a fraction of those people actually turned up on the day. Nevertheless, the discussion overran the allotted time, and I found it really useful – both in clarifying my own thoughts and picking up some new ones. Thank you to everyone who came along. Below are my initial reflections on the session, and how I ought to be thinking about Data to Insight’s role in supporting (not building!) the community which has grown up around our data tools.
I'm attending regional PIMG meetings over the next couple of months to set out the Data to Insight stall and give a sense of the project's immediate goals, etc. and as part of that I'm really interested to gather as much input as possible from LAs. If you have any thoughts on the below, please do chip in either here or by email if you prefer.
Communities aren’t “built”
Communities grow of their own accord, wherever something worthwhile is happening, and the best aspiration the Data to Insight project can have is to assist the community which is growing naturally around this work. Rather than trying to shape it, we need to pay attention to the shapes it wants to make, and help those flourish. There are definitely things we can do to help in that respect, but they're less about building something new or "better" than they are about revealing what's there, and helping it work.
One size doesn’t fit all
Much as different LAs will have different levels of engagement in communities like this one, different LAs will, despite their commonalities, have different analytical needs at different times. With a few exceptions (like the ChAT), we should probably expect each of the data tools or ideas that we support to be adopted by a subset of the wider community. As such, we might not necessarily need to develop every tool as if it’s going to be used by every LA, and we might think instead about how best to help smaller communities within the wider network do what they need to do. We could definitely benefit from building our tools to be “tweakable” by end users – and, ideally, for those tweaks to be shareable.
It can be hard to find good work
We share well within regions, but even then it can be hard to know who’s got a good solution already in place for any given challenge. One of the key things that a dedicated Data to Insight resource could do is to find those examples of good work and help share the approaches/tools – ensure visibility across the community.
We’re good at sharing tools, but less so at sharing practice / approaches
Some LAs reported that they feel LAs and regional groups are good at discussing and sharing social work practice, and getting better at discussing and sharing data tools, but there is a gap around discussing and sharing performance frameworks, methods and attitudes. One of the things Data to Insight could try and do is to broaden the discussion of data tools out into a discussion of performance practices more generally – less “what do your numbers say” and more “which numbers matter, and what do you do with them?”
There are ways past the technical barriers
…like open source software licenses, which protect creators against unforeseen consequences of sharing their work; or like collaborative funding approaches such as we used to refine the Demand Modelling tool; or like the apprenticeships and shared learning we're moving on. But there needs also to be motivation to overcome the barriers. Just because there’s a way around a difficulty doesn’t mean people will go out of their way to take it.
Some people just want the tools, and that’s fine
We had confirmation at the session that there are definitely some LAs which make heavy use of the ChAT, statutory return tools, etc., and do so partly because these help reduce the burden of data work in small teams. They want this stuff to exist and are happy that other LAs are driving the design and development.
Goodbye lecture theatre
The “lecture theatre” approach to dissemination, whereby we gather an audience and recite information to that audience, doesn’t always help those audience members who could most benefit from the help. When we have a great data tool or method, we need to do more than say that it exists; our "super users" can work with that, but elsewhere we’d do more good if we provided brief examples of how to use it, and followed those examples up with one-to-one interactions.
Diffusion
Innovation isn’t mandated; it diffuses, and sometimes very slowly. We can help by finding innovations and building relationships to help those innovations spread more quickly. We can also help by finding easy first steps for community members to take – to draw people into a “lightly engaged” state, and then from there help them find – or instigate – the specific subgroups or projects which inspire them. As Data to Insight, sometimes it will then be enough for us to know about those projects, and help other people find them.