Exploring the role of data in post-Covid recovery

by Eddie Copeland
325 views

Last week, I wrote about how local government could build on some of the trends seen during the Covid crisis response to create a lasting and positive legacy. I summarised my thoughts in this table:

Others are thinking along similar lines. I particularly liked the diagram below, which Matthew Taylor, Chief Executive of the RSA, shared in his recent article.

The part of this diagram where we need to spend the most time and energy is surely the top right-hand “Amplify” quadrant: positive developments we could build into new practices.

That quadrant is worth fighting for. It’s also risky.

It’s worth fighting for because the natural tendency of most systems under stress is to seek to revert back to how they were as soon as possible. Yet the challenges of recovery will not be met by simply returning to the pre-Covid status quo. The needs are different and greater. We’ve also witnessed the early seeds of opportunities to do things differently that should not be wasted.

It’s risky because confirmation bias may lead us to assume that the changes we now need are those we’ve always said should happen anyway. (Feel free to critique me for doing that in this article.)

For those who do advocate change, the best means I can suggest to avoid confirmation bias is to ensure we set out our arguments for why specific measures are necessary, which can be judged on their own merits. That’s what I’ll attempt to do here.

While at LOTI we’re interested in the broad role that digital innovation can play, in this article I’ll focus my remarks specifically on how we might think about the future of data.

How do you respond to a crisis?

So how might we think about exploring the Amplify box in the diagram above? I’d suggest three approaches are likely to emerge:

Let’s discuss these in the context of data.

Specific Fixes — A number of urgent data requests have arisen during Covid where it’s been apparent that councils simply don’t have the data they need. One example is how local authorities have needed to distribute business support grants. Many have discovered that while they have good records of local companies on their business rates database, they lack email or bank details for the majority. That makes it incredibly difficult to get payments out promptly. We can and should fix specific issues like this and ensure councils have those details in future.

New Opportunities — A crisis also prompts us to think about how things could be done differently and better. Perhaps the single greatest new opportunity we could aim to realise on a data front would be shifting from static to dynamic (if not real-time) data on a greater range of issues. As public sector staff, from CEOs to front line workers, have sought to respond to the crisis, the limitations of relying on static weekly, monthly or annual figures have been laid bare. As factors such as transport usage, high street activity and use of public spaces become deeply important in understanding the nature of recovery, more dynamic data could make a real difference.

Generic Capabilities — While the first two categories of activity are worth pursuing, I’d argue the single most positive legacy that could come out of a crisis is that we put in place generic capabilities — core foundation stones — that make us better able to respond to whatever comes next. Some of those capabilities will be about what individual councils need to have in place to use data well. However, given that few crises respect local authority boundaries, arguably the most important set of capabilities concern how different organisations can collaborate with data.

Putting in place the foundation stones for data collaboration

For years there has been discussion about the factors that make data collaboration between different public sector bodies hard.

Five stand out.

  1. Technology — some technologies make it hard to get the data out (e.g. lack of APIs); worse, some suppliers charge councils to access their own data.
  2. Data standards — the use of different standards, formats and conventions for recording data, and the lack of common identifiers like Unique Property Reference Numbers (UPRNs) makes it hard to compare, link or match records.
  3. Information Governance (IG) — Ensuring that London’s public sector organisations can use data in a way that’s legal, ethical and secure — in short, worthy of citizens’ trust and confidence — is key. Yet councils’ different approaches to IG can make the process take a long time — sometimes months.
  4. Ways of working — councils’ different processes require and produce different data.
  5. Lack of skills — when data skills are at a premium, councils understandably need staff with data competencies to work predominantly on internal projects, with little time available for collaboration.

There’s a host of reasons why progress to resolve these barriers has been slow. But perhaps the greatest is the perception that the effort required to address them is greater than the reward of doing so.

The Covid crisis is changing that perception.

Elected members, chief executives, resilience leads, national government and even the media are realising that it can’t be acceptable to take weeks or months to get access to critical data. Yet it would be a mistake for the Covid legacy to be confined to making sure specific datasets are available next time. Rather we should ensure cities have the capabilities to access and use data responsibly and rapidly based on whatever needs arise.

Looking at a London context, I’d propose reframing the barriers listed above into five key foundation stones for data that need to be put in place to make the capital a stronger, more resilient place. I’ve listed some specific recommendations for what should be done under each heading.

Foundation Stones of City Data for a Good Recovery

1 — Technology:

  • All systems to provide full and free access to data. Now is the time to ensure that all new local government contracts that procure technology (or a service that has a technology component) require suppliers to give the council full and free access to system data. Without that access, analysis and insight to inform decisions and actions are impossible.
  • Default use of the London Datastore as London’s data-sharing platform. If all 33 London boroughs need to share data with their 32 peers, it would require 528 data sharing relationships to be established. That simply cannot be the best option. Instead, boroughs should share data with each other via the London Datastore, involving just 33 connections (one for each borough). Where appropriate, this approach could also allow for pan-London datasets to analysed and/or visualised.

2 — Data Standards

  • Consistent use of UPRNs. A crisis often involves having to pull together data from different sources to build a more complete picture. That is made incredibly hard in the absence of identifiers that enable data to be linked. During Covid, where boroughs have urgently sought to understand which households are most vulnerable, those who consistently use Unique Property Reference Numbers (a unique ID for an address) have found it much easier to identify households with multiple indicators of vulnerability. The implementation of UPRNs across council systems should be prioritised.
  • Adoption of data standards for new technology deployments. Data standards make it far easier to link and compare datasets. Yet they are hard to implement retrospectively since they are tied to the different technologies boroughs use. However, we do have the chance to get things right with new technology. Building on LOTI and the GLA’s IoT Sprint Week — and recognising the role of smart street infrastructure in providing some of the more dynamic and real-time data that cities are likely to need to respond to future challenges — it’s right that we should lay some ground rules and standards. Failing to do so risks accidentally perpetuating the very data silos that cause so many issues with legacy technology, preventing the pan-London insights we need.
  • Data translation as a service. Where it’s impossible to implement a data standard, it may be that data hubs like the London Datastore (and similar platforms in use by Offices of Data Analytics in other parts of the UK) need to offer data translation as a service; mapping different standards to a common one that can then be used to build a pan-London picture.

3 — Information Governance

  • Adopting a common approach to IG. As separate legal and democratically elected entities, it’s right that boroughs should satisfy themselves that they’re using data in keeping with their values and commitments. Yet, when signing one information sharing agreement (ISA — a joint agreement on what data partners will share, for what reason and how) can take six months or more, boroughs should push ahead on developing a more streamlined and efficient process. Working with IG Leads, LOTI has produced a 7 step IG process that could help form the basis of a common approach across London. We now need to double down on putting that into practice.
  • Adoption of Digital DPIA tool and Information Sharing Gateway. A key part of the IG process is to conduct Data Privacy Impact Assessments (DPIAs) and Information Sharing Agreements (ISAs). Simple version control issues make these very time-consuming. Boroughs could benefit enormously from adopting dedicated collaboration tools to do this, including the digital DPIA tool LOTI is developing with CC2i and partners from Greater Manchester and Norfolk County Council; and the Information Sharing Gateway. Importantly, these tools will only deliver real value if they are used by the vast majority of boroughs.
  • Assurance of London Datastore team as trusted analysts. Connected to my earlier recommendation to use the London Datastore as London’s default data collaboration platform, one barrier to achieving this is that it may entail adding the GLA as a data processor to boroughs’ data sharing arrangements. In the coming days, we’ll be holding conversations with the GLA and the Information Governance Group for London to ensure boroughs have the assurances they need to be able to use the platform with confidence.

4 — Ways of working

  • Establishment of peer network of data professionals. Data professionals may work differently across boroughs, but they can still accelerate their learning by connecting with their peers to share details of their projects, insights and examples of good practice. LOTI and the GLA have launched a data analysts forum for the Covid crisis, which should endure beyond.

5 — Skills

  • Extension of digital apprenticeships to cover data roles. We know it’s hard for local government to recruit talented data professionals, yet those skills are certainly needed. Recent meet-ups of London borough data analysts has revealed a particular need during a crisis for staff who can create dynamic dashboards and produce useful data visualisations. In terms of foundation stones, we need a sustainable pipeline of such talent. One option would be to build on LOTI boroughs’ recent work to recruit digital apprentices and put concerted effort behind creating apprenticeships for data analysts, data scientists and other vital roles such as IG professionals.

From Covid to Recovery

Nothing I’ve outlined above is new. What is novel is the level of recognition Covid has brought to the importance of data in being able to rapidly respond to new challenges.

None of us knows what the few next months or years will hold.

But as we work towards a good recovery, we should take time to fix what’s broken. We should grasp new opportunities to do things better. Most of all, we should put in place the foundation stones to ensure we are stronger, faster and more adaptable for what whatever hits us next.

Find me on Twitter | Find LOTI on Twitter

You may also like