5 points of focus for the ‘next phase’ of GDS

August 27, 2015

As the dust settles following the announcement of several high-profile departures from the Government Digital Service (GDS) – not least of its Head, Mike Bracken – it’s worth taking a moment to reflect on what should come next.

Matt Hancock, Minister for the Cabinet Office, has assured that GDS will remain in place, saying: “e have the right people… to deliver the next phase of GDS. There is so much more to do, and I look forward to leading it at Ministerial level.” Yet it is widely expected that it will have a smaller budget and reduced headcount.

When any organisation has fewer resources, there is a need to focus on core strengths: the famous 20% of actions that produce 80% of the impact. With that in mind, what areas should GDS focus on as part of its “next phase”?

I’d suggest that there are five things it is uniquely well-placed to do:


GDS should define the standards of how front-end government IT should work, but be agnostic about who provides it. Just as Apple has strict standards for what developers can build for iOS, GDS should do the same for transactions (e.g. renewing a passport online).

It has taken four years to make 20 exemplar transactions. There will be a long wait if in-house teams have to convert all 770-odd government transactions themselves, let alone build the components of government as a platform (GaaP). (By contrast, HMRC digitised 500 transactions in 6 months using Adobe’s form software.) A smaller GDS cannot be expected to match the pace, innovation, experience, expertise, security and economies of scale of developers who spend all their time looking at just one product. But they can write the rules of the playing field.

That said, just as Apple makes some apps for iOS, GDS should build some pieces – but only where government is uniquely well placed to do so.


At any rate, the hard part of creating online transactions is not in digitising them. The real value comes from deciding whether they are needed in the first place; and if so selecting the information that is requested, presented and used. At these things GDS excels (see their work on how to design web forms for GOV.UK services).

This really matters. Channel shift will not save money unless people not only start but complete transactions online. As any online retailer will tell you, a positive customer experience is vital for making that happen. GDS has been a much needed users’ champion. It should continue to make sure that the citizen is front and centre of service design across government.


Mike Bracken was recently appointed as the government’s first Chief Data Officer. His job description specifically included the remit to ‘define open standards for the whole public sector’. Open standards (common formats / ways of recording information) reduce the technical barriers to sharing information between different systems. As I have previously argued, this is a key step to enabling better ways of working that can deliver real savings. As Mike leaves, GDS should continue this important role.


One (among many) of the reasons that some past public sector IT projects went badly was that government was not a good customer. Though some commentators complain about the old world of ‘big IT’, many of its shortcomings were of the government’s own making. For all the criticism levelled at large SIs and tech firms, their structure and the way contracts were delivered were often a response to what government said it wanted to buy, and the procurement processes they had to adhere to.

One of the most positive contributions GDS has made is to inject a critical mass of serious tech talent into the very heart of the civil service. They should use that expertise to help government become a smarter, more demanding customer, and to ensure that cases of defining specs one year and getting an inadequate system late and over budget several years hence are gone for good.

GDS should also make sure government retains an institutional memory of what does and does not work. As the book Digitizing Government (by Alan Brown, Mark Thompson and Jerry Fishenden) reminds us, many of the same IT issues come around time and again. It is vital that central government has a centre of digital excellence to learn from past mistakes.


Delivering results that save real money (an operational imperative for the current government) also requires scaling best practice across departments. I have previously argued that GDS cannot exist as an island of digital excellence in an otherwise unreformed Civil Service, but instead must be a catalyst for helping departments learn from and implement the very best ideas.

I completely endorse Mike Bracken’s assertion that there is a need for a ‘digital centre of government’. He has argued: “We can’t just keep making or buying technology solutions in one department and then just chucking it over the departmental wall and saying, ‘That will work for the rest of government’, because it never does. Ever.” Amen to that. That catalysing role will be vital if work on Government as a Platform is to be a success.


GDS in general – and Mike Bracken and his team in particular – have put in place some very positive measures to bring government into the digital age. GOV.UK has genuinely improved citizens’ ability to find information. Verify cleverly squares the circle of providing a single login whilst avoiding the politically-toxic idea of having a single citizen ID for government. (It should be rolled out to local government and the wider public sector as soon as possible.) The 20+ exemplar transactions are considerably more user-friendly than those they replace.

There should be no let up in those improvements.

But if GDS is set to be a smaller body, it is in everyone’s interest that it be one that can make the greatest impact.

GDS has often been compared to a tech startup.

Now is the time to behave like one by pivoting to where it has the strongest USP.

Join our mailing list