Data is the lifeblood of organisations, the oil that keeps the machine turning. Once collected and scrutinised, it enables analysts to understand user behaviour, improve the quality of products and services offered, spot trends and predict demand. But it’s not as easy as it sounds.
At the Department for Work and Pensions (DWP), its store of 200 terabytes of data is growing at a rate of 10% a year, with an increasingly complex mix of sources – from documents to video and voice. It’s a phenomenal resource but one that the department has found challenging to make the best use of in the past. “It was mostly used around understanding our customer base, citizens as a whole, and big socio-economic movements,” says Charlie Boundy, deputy director for advanced analytics. “Now it’s increasingly about understanding our services, our performance and how we measure success.”
Making better use of data is at the heart of the Digital Future strategy at DWP Digital, which also aims to deliver great services, build new user-centred solutions, and build capacity and capability. A move that helped to drive momentum was the establishment of DWP’s chief data office in October 2021. “We now have a tangible group within the department that takes responsibility for data governance, security and data enablement,” Emma Presley Abbott, head of enterprise insight, says. “That’s been a key part of making sure the exploitation of data becomes much easier and a much more common part of our ‘business as usual’.
“We’ve realised we need to learn from the millions of transactions and interactions that we have with users, rather than just storing data to calculate a payment. There’s a real appetite all across the department, at all levels, to make the best use of the data that we have.”
Presley Abbott’s team is responsible for the digital platform that acts as a one-stop-shop for the whole department, and facilitates the enterprise reporting service (delivering reports from data sources). She aims to tackle some of the main challenges that have prevented data from being used in a meaningful way – accessibility, data quality and traceability. “The ambition for my team is to make that data more readily available to colleagues so that it can be better used for decision making,” she says. “We want to make the whole data landscape feel like a part of everybody’s role, rather than this mysterious, abstract concept.”
That’s already having results. Employment programmes such as Kickstart and Way to Work have been launched, evaluated and optimised within short timeframes to ensure they have the maximum impact. “You have to constantly and quite proactively monitor their rollout, providing the insight required to pull different levers, and make changes on the fly,” says Presley Abbott.
Those reporting capabilities also came in handy during the Covid-19 pandemic, when the need for immediate access to accurate data was paramount, Boundy adds. “You can imagine with a stable economy and quite stable unemployment rates, people are usually happy to deal with data on a weekly or monthly basis. During the pandemic, everybody needed it, every day, every minute, in areas where we’d not shone a spotlight before. We had to mobilise to do a lot of very rapid insights and be a lot more reactive and responsive to user need.”
The past two years have also opened up the possibility for cross-departmental data sharing, which Boundy is exploring through a pilot project with several government departments that aims to provide insights to aid economic recovery. “Historically departments have been, quite rightly, very nervous about any form of data sharing, because there’s a huge data protection risk,” says Boundy. “But we’re looking at how we can coordinate departments who’ve got a shared problem to target our efforts better.”
New technologies are also being considered, such as machine learning, blockchain and cognitive computing. Boundy is clear that DWP Digital has already drawn hard lines around how and where artificial intelligence (AI) will be used. “There’s no appetite for automated decision making using artificial intelligence. Anything involved in entitlement or calculation of a payment does not have AI involved at all. Where it makes a big difference is in triaging work. Who needs a bit more support? Who needs a different type of service, rather than a one-size-fits-all approach? We want to get to a point where we are differentiating a service based on the specific needs of the customer.”
DWP’s evolution to a fully data-centric organisation won’t happen overnight. But Presley Abbott says she’s excited to be part of the journey. “We provide services to millions of people, at real times of need in their lives. Being able to use data to make sure those services are working effectively and the customer experience is as good as it can be is quite hard to beat.”
Ready to make a difference? Find out more about the career opportunities offered by DWP Digital
