How embedding learning is helping to deliver better data science in Essex

Liz Ridler
Delivery and Evaluation Lead Public Service Reform

Essex County Council

Data sharing and data science are pretty hot topics at the moment. I have been to numerous conferences, workshops and discussions that focused on how doing more stuff with data will be a panacea. We may yet still have some way to go to reach this promised ‘cure-all’ solution in the public sector but we’re all keen to share our journeys and learning to reach enlightenment! 

This ‘collaborative learning’ in itself is brilliant and something we should celebrate – it shows we are shifting our understanding of how projects and organisations work from linear and planned upfront, to cyclical and iterative, each cycle based on learning. Making reflection and learning second nature – failing quickly, or asking for forgiveness rather than asking for permission – enables us to understand what works and what doesn’t and supports a more honest and collective way of working. 

So, in this spirit, I’d like to share some insight around what we have  learned here in Essex thus far through our programme to use system-wide data to predict outcomes for potentially vulnerable people, which the Centre of Excellence for Information Sharing has supported us in.  And importantly how the journey so far has influenced our delivery. 

We can’t know it all up front 
It seems easy to say we want to share these data sets, to enable us to make that change.  But the reality is far more complex and importantly not static – so you can set out to do one thing but then find circumstances have changed.  To share individual level data successfully we need to be clear about the purpose of sharing the data. However, this is a catch twenty-two because it can be hard to be clear, and possibly limiting, to set this out in advance of sharing and then modelling the data. A process of iterative development, delivering small and building up, is helping us to address this. 

We need to think through the ethical implications
We are dealing with sensitive issues and data and so we need to have the right conversations up front. We have used the new Cabinet Office data science ethical framework to help us. Having the right conversations reassures stakeholders that we are thinking more widely than project deliverables and helps us make sure we have the right governance and processes in place to protect people and modify our project accordingly. 

Combining skills and disciplines makes this work 
We are dealing with system-wide issues that don’t keep to organisation boundaries.  To deliver this work we need input from subject matter experts who know the policy area, analysts, data scientists, information governance, technologists, communications specialists and of course commissioners and service leads who will use the insight. This mix of skill sets and competing demands needs to be harnessed towards a common deliverable. 

We are developing a more flexible and agile working approach to enable us to respond to what we have learnt to date about delivering Essex Data. This enables us to more effectively deal with an emerging solution, and accept that change is inevitable, not a failure. We hope that the changes we have made and will continue to make, in response to our learning will move us more quickly towards the panacea offered by data science.   

I have picked a few key learning areas to explore here but there are more, as well as other information about the Essex Data Programme at www.essexfuture.como


Leave a Reply

Your email address will not be published.