The UK Government Digital Service

For the exclusive use of i. mabrouk, 2019.
Julian Birkinshaw
LBS Ref: CS-18-28
Scott Duncan
HBP: LBS210
August 2018
The UK Government digital service
Mike Bracken sat in his office in Holborn, London, in August 2014, just over three years since accepting
the job as Executive Director of the UK Government Digital Service (GDS), thinking about where the
team had come from, what it had achieved and where it was going.
A team within Cabinet Office, GDS had been formed as a catalyst for change in 2011, tasked with
transforming the UK Government’s digital offering from the inside. To the tech community, GDS was
an unmitigated success story, having made millions of people’s lives more convenient and with an
estimated £1.7 billion in potential savings every year. GDS had received acclaim across the world:
The Washington Post claimed: “Since its launch in 2010, GDS has emerged as the gold-standard in
the global world of digital government.”
There was also a broader, more ambitious, agenda: to be a catalyst for transforming the UK
Government. Bracken and many of his colleagues saw GDS as enabling the shift from a traditional
process-driven approach to one focused on delivery.
But while the results of the 25 exemplar projects (the largest government transactions that were being
transformed to meet user needs) had been impressive, media reports citing (often anonymous)
government officials suggested that beneath the surface there was a backlash against GDS’s centrally
driven transformation. He observed:
We have achieved a great deal over the last few years, and our approach is
demonstrably better than the old way of working. And yet there are still people who
think that working in an agile way should be optional. Inertia is a powerful force.
With the next general election in May 2015, there would be no significant policy changes in the coming
year. Bracken’s challenge was how to maximise the chances of success in the next government.
Julian Birkinshaw is Professor of Strategy and Entrepreneurship, Academic Director of the Institute of Innovation and
Entrepreneurship, and Deputy Dean at London Business School. Scott Duncan, London Business School case writer.
London Business School cases are developed solely as the basis for class discussion and are not intended to serve as endorsements,
sources of primary data, or illustrations of effective or ineffective management
© 2018 London Business School. All rights reserved. No part of this case study may be reproduced, stored in a retrieval system, or
transmitted in any form or by any means electronic, photocopying, recording or otherwise without written permission of London
Business School.
This document is authorized for use only by ilyas mabrouk in 2019_Fall_MSc International strategy and influence_organizational dynamics taught by Amine Ezzerouali, Skema Business
School from Sep 2019 to Dec 2019.
CS-18-28
Background
The UK’s first digital initiative, open.gov.uk, was piloted in 1994 and was an attempt to put various
government services online. Directgov, launched in 2004, was another attempt at setting up a web-
based portal, with a parallel initiative, Business Link, providing digital services to businesses. But by
2007 most digital transactions with government still went through hundreds of individual departmental,
agency and local authority websites. A review at that time estimated the Government could save £400
million over three years by channelling all its electronic activities through Directgov and Business Link.
But these entities lacked central control over website domains. A related problem was the UK
Government’s poor track record in delivering major IT projects. For example, the National Health
Service IT project had been written off in the mid-2000s at an estimated cost of £9.8 billion.
In 2010 then UK Prime Minister David Cameron appointed internet entrepreneur Martha Lane Fox as
the new ‘UK Digital Champion’, asking her to advise the Government on how delivering public services
online could make them more efficient.
Her report, Directgov 2010 and Beyond: Revolution not Evolution, recommended the formation of a
new, centralised digital team with absolute control of the overall user experience across all digital
channels and with a head reporting directly to the Cabinet Secretary.
Francis Maude, the Cabinet Secretary, pledged support for the proposal and gained approval in
principle through an internal funding body called the Public Expenditure (Efficiency and Reform) Sub-
Committee, or PEXER for short. An interim CEO, Chris Chant, was appointed in January 2011. The
Deputy Director was Tom Loosemore, a highly experienced developer of digital services.
Experimenting with a new approach to software development

Loosemore was tasked with hiring and building the team that would develop a prototype or ‘alpha’
version of a single website for the UK Government. He asked for only three months to do it. He recalled:
Rapid prototyping is a normal way for large commercial organisations to test their
web services, but it had never before been done in government.
Rapid prototyping was part of incorporating an ‘agile’ methodology of software development. The agile
software movement emerged in the 1990s as a reaction against the process-heavy ‘waterfall’ model
of software development, which had been the standard approach for large providers and system
integrators. This approach had, until now, dominated government digital projects. The agile movement
promoted adaptive planning, a time-boxed iterative approach, and encouraged rapid and flexible
response to change, all done through collaboration between self-organising, cross-functional teams.1
Loosemore assembled a skilled team from inside the civil service and his outside contacts. The
prototype, alpha.gov.uk, was launched in May 2011, a 12-week process costing £261,000. Initially it
© 2018 London Business School

Page 2
CS-18-28
contained answers to the top 100 most frequently asked questions in government, like “What to do if
you lose your passport”. In the language of high-tech start-ups, it was a minimum viable product:
The alpha was just sketching in code…The unofficial objective, which I never wrote
down but people knew, was for ministers to look at it and go: “I want one of those”.
That’s why the design was quite shiny, more so than you'd normally do on an alpha.
But it was quite shallow. It was a prototype. So you look at it and go, “Wow, that could
be amazing”. And you have a real clickable, usable sense of what the future could
be.
In the two months following its launch, there were over 100,000 visits to alpha.gov.uk, with nearly
1,000 people leaving structured feedback and a further 3,000 passing comment via Twitter.
Finding an Executive Director

The search for a permanent successor to Chris Chant began in April 2011. Mike Bracken emerged as
the preferred candidate. Bracken was the Director of Digital Development at Guardian News & Media,
and he had also worked on open data initiatives with the UK Treasury. He said:
I could see there was a genuine opportunity to make something happen, the forces
were in alignment. But the challenges were also huge – legacy IT systems, lack of
recognition and reward for specialist skills, and culture change and senior-level buy-
in needed to make that change happen. The task was a hugely exciting opportunity.
Before accepting the role, Bracken articulated his requirements to Francis Maude:
We really need three things: the right environment and building, people with specialist
skills and a single central (internet) domain. They expected me to say: “I want lots of
money”. I didn’t say that. They expected me to say: “I want the powers of
departments”. I didn’t say that. And they expected me to say: “I want some form of
hierarchical and status-based power in the centre”, and I didn’t say that. What I did
say was that I want a building, people and domain.
Maude accepted Bracken’s requests and he was appointed Executive Director in July 2011. Control
of the gov.uk domain was formally transferred to GDS and backed up by the power to design and edit
its contents at another PEX(ER) meeting. Bracken saw his role as implementing the proposals in Lane
Fox’s review:
First: create a digital centre of government. That had never been done before, and
most governments still to this day do not have one.
Second: fix publishing – that is, how the government comes across online.

Page 3
CS-18-28
Third: fix transactions – redesign and improve the way individuals and businesses
interact with government (such as applying for passports, driving licences, stamp duty
on houses, tax and benefits, with more than 1 billion actual transactions every year).
Fourth: go wholesale – that means helping third party organisations create new
services and better information access for their own users by opening up government
data and transactions.
Creating the GDS culture

Bracken’s first move was to consolidate the GDS team in Aviation House, a 20-minute walk from
Whitehall. The environment was, by design, as different as possible from a traditional government
department. It looked and operated like a technology start-up and quickly developed a bustling, high-
energy atmosphere. An early move was to send someone out to buy iPhones for everyone. They were
handed out with the simple instruction: “This is your GDS phone. We trust you to use it responsibly.”
Openness was an important feature of GDS’s approach. Bracken was a regular blogger and he
encouraged his team to blog as well (see www.blogs.gov.uk) and to code in the open. The idea was
also to turn content into data that others could use.
Another priority was to build the right team of people:
We were hiring like mad, we were frantically opening the doors for a generation who
had been excluded before from government – the web generation. So all I did was
basically put the door open sign on and said, “Look, this seems to be serious.” I just
went round and talked to them and said, “I think all the forces are aligned, it’s worth
having a go”, and a lot of people went, “Well, I’ll give it a year!”
The operating budget for GDS in 2011-12 was £22.3 million, with about 200 people working in the
organisation. Many of the people hired could easily have made more money working in the private
sector.
Bracken’s approach was to act quickly, to get things launched and then fine-tune content based on
feedback from users. An early project focused on online ‘e-petitions’, a system for collecting signatures
on petitions to government. As Bracken recalled:
We did it in four weeks with a few guys. We bought the hosting for about £100,000
and were vastly successful. Unfortunately, we had no idea how fast demand for e-
petitions would go up – we had 1,700 visitors per minute on the first day and the site
crashed. But we reacted quickly, and we learnt a lot about scaling this way.

Page 4
CS-18-28
Today, I think one in every eight people in this country has used the e-petitions site
or participated in it and it was a model for how you can do this stuff differently.
GDS developed a way of working that became captured in a set of operating principles like “Show,
don’t tell” and “Publish, don’t send.” In Bracken’s words:
Rather than go in and trying to win hearts and minds, we just thought the best result
is when you do stuff people like. Lots of things we couldn’t touch because they
involved legacy technology – large contracts which would have taken a year to unpick
– so we were very tactical in the first six months because we were still building.
The beta version of GOV.UK was planned to launch in January 2012. It would include three things: a
public test of the site delivering the citizen-facing aspects of GOV.UK; a private test of a shared
GOV.UK platform that explained what each government department did; and a first draft of a ‘Global
Experience Language’, which would provide clear standards in the future.
It was delivered on time. On the day of release, Tom Loosemore wrote on his blog:
We have re-written, re-designed and re-thought 667 of the needs people have of
Government (broadly, those currently catered for by Directgov) – making them as
findable, understandable and actionable as we can. We’re using open software and
tools as much as possible.
Design principles
Building GOV.UK around user needs meant paying attention to design. Ben Terrett, Design Director
at ad agency Wieden and Kennedy, joined GDS as Head of Design in January 2012 and established
a set of design principles:
1. Start with needs: user needs, not government needs.

2. Do less.
3. Design with data.
4. Do the hard work to make it simple.
5. Iterate. Then iterate again.
6. Build for inclusion.
7. Understand context.
8. Build digital services, not websites.
9. Be consistent, not uniform.
10. Make things open: it makes things better.

Page 5
CS-18-28
Terrett and his team started to look for an overall theme, on the basis that “public services should not
only be functional, they should also be beautiful”. The obvious place to start was some kind of British
aesthetic.
The visual art direction brief was literally, make it look like it doesn’t come from the
west coast of America, because all websites look like they’ve come from there and
there’s no need for this one to look like that! We tested a bunch of typefaces. The
one that kept testing really well was called ‘Transport’ – it was used on the road signs
in the 1960s! So we used one called ‘New Transport’ and it’s exclusive to us, so that
other people can’t use it.
Working across government

Bracken spent a lot of time working with the various government departments to persuade them to
buy into his strategy. To help him with this, he created the Digital Leaders network – a digital expert
from each main government department, plus one each from Scotland, Northern Ireland and Wales.
As Kathy Settle, one of Mike Bracken’s team, explained:
The Government added impetus to the digital agenda in March 2012, when the
Chancellor announced that, “from 2014, new online services will only go live if the
responsible minister can demonstrate that they themselves can use the service
successfully. The Government will also ensure that all information is published on a
single ‘GOV.UK’ domain name by the end of 2012.
However, progress was not entirely smooth. During 2011 Mike Bracken and the GDS team had
experienced mixed reactions from officials in other government departments:
We were told, you can’t touch our legacy systems, you don’t understand them and
they are too big; your new methodology isn’t secure; open-source development is
dangerous; you don’t have the skills. For each challenge, we developed evidence to
counter it. Gradually, we won people over.
Bracken had several sources of leverage in tackling these concerns. One was the “constant and
unwavering” support of Francis Maude, who was focused on achieving efficiencies in government.
In addition, control of IT spending sat with the government’s Chief Technology Officer, Liam Maxwell,
through the authority of PEX(ER). During this period, Maxwell sat in the Cabinet Office building but
with a close working relationship with Bracken, who recalled:
Liam and his team, through the judicious use of IT controls, were able to counter
much of the resistance. Nonetheless, the greatest source of pushback was the CIO
community, who were used to working with big IT contracts and procurement teams

Page 6
CS-18-28
and with a handful of large-system integrators. So, to tackle this, we decided to

abolish the CIO title and push departments to adopt our lean principles. This period
of resistance was personally and professionally the most challenging period of my
career to date.
The official release of GOV.UK

Francis Maude officially announced the release of GOV.UK in October 2012. From that date, every
visitor to the www.direct.gov.uk and www.businesslink.gov.uk was automatically redirected to
www.gov.uk. One month later, the Government published its Digital Strategy, setting out how the
Government would become digital by default:
We estimate that moving services from offline to digital channels will save between
£1.7 and £1.8 billion a year.
In late 2012 the next phase of work began, including the transformation of 25 of the largest government
transactions (later known as ‘exemplars’), the building of a performance platform to allow live tracking
of government digital services, a digital marketplace offering, and an expansion of the Government’s
identity assurance programme. In April 2013 Liam Maxwell was moved over to GDS to report directly
to Mike Bracken. Bracken explained:
This was a massive shift. It meant that we had broken the old procurement mentality
whereby a government department would either hire hundreds of contractors or bring
in a big IT services company whenever they needed help.
Also in April 2013 it was announced that the website GOV.UK had won the Design Museum Design
of the Year Award. It was chosen from 98 entries to claim the overall prize, beating off competition
from the Shard, the Olympic cauldron and a revolutionary folding wheel. This was the first time a
website had won the prize. The jury noted its “well thought-out yet understated design, making the
user experience simpler, clearer and faster.”
The second half of 2013 and the first half of 2014 saw GDS continuing to focus on transformation of
digital services and across government. Cross-government events, like ‘Sprint Alpha’ and ‘Sprint
Share’, brought people together to share their experiences of working on transformation. By August
2013 the discovery phase had been completed for all the exemplars. Many government agencies
began migrating to GOV.UK.
During 2013 GDS grew to around 400 people. In terms of usage, GOV.UK had half a billion views by
January 2014.

Page 7
CS-18-28
Media attention and resistance

Much of the media attention around GDS was positive, but there were also some signs of resistance
to its progress. For example, an article in the Financial Times in June 2014 2 hinted at sensitivities in
IT procurement decisions. Two departments, BIS and DECC, were the first to shift to the new ways of
procuring IT developed by GDS and they had encountered teething problems during the transition.
The article said:
Another Whitehall official admitted the two departments had been suffering from
“transition issues” but defended the changes: “This looks, quite unfairly, like an
attempt by DECC and BIS to shift blame for their own mistakes. All departments
remain responsible for their own technology and service procurement.
When asked about potential tensions between GDS and departments, Bracken said,
GDS has hard powers: IT spending controls, domain power – that’s GOV.UK and
standardisation power, basically saying what good looks like. However, these formal
powers are rarely used. We are effectively a vehicle that works on soft power, driven
by user need … Our strategy will continue to be delivery – delivery of such a high-
quality platform and services that people in government want to work with us and
users are delighted by the services that are in use. These two things alone remove
all the other obstacles to working together.”
Pondering the future of GDS

Sitting in his office in August 2014, Mike Bracken reflected on the progress achieved. Publishing had
been ‘fixed’, with all 24 ministerial departments and nearly 300 agencies and arms-length bodies now
using the GOV.UK platform.
Fixing transactions was well underway – 25 exemplar services had been identified and by September
2014, 13 were in public beta or in full service. They had also improved transparency, with 105 service
dashboards and performance details on all transactions now available to the public, with many updated
in real-time.
The final objective, going ‘wholesale’, was underway. There was a plan to make all information held
on GOV.UK easily reusable and to make publishing common-data formats simpler and easier.
Depending on the outcomes, GDS would then look to select a few services to work with to build
prototypes and pilot API use.
Turning his thoughts to the broader challenges and opportunities facing GDS, Bracken observed:
We’re here to reform government, but we haven’t been able to say that until recently.
It has taken three years and an immense amount of work to be able to say that. It

Page 8
CS-18-28
was only in October 2013 for the first time I publicly said that we’re here to
fundamentally reform the institutions of the State.
David Thorpe, Director of Policy at GDS, offered his perspective:
Domain control has allowed us to get a lot of things done very quickly and to show
how things can be done. Whitehall alone has around 17,000 professionals with policy
in their title: traditionally they have understood an important part of their job as writing
the requirements which someone else has to implement. This creates the
implementation gap we see so often in the public sector.
But delivering on these ambitious objectives was far from straightforward. The UK Civil Service was a
conservative institution and it had been clear to Bracken from the outset that fundamental change
would be difficult to achieve. With the next General Election less than a year away, Bracken had to
think about both short-term steps and the long-term future of GDS.
In terms of next steps, one approach was simply to continue to fix transactions and improving services,
according to the ‘strategy is delivery’ mantra. Alongside this, a second approach would be for Bracken
and his team to spend more time getting buy-in from influential civil servants and elected ministers.
A third approach would be to build up GDS’s reputation outside government, in the UK and overseas.
This would increase the visibility of how GDS was transforming digital government and help reinforce
its potential value as a catalyst for change.
Bracken also needed to develop a point of view about the future form of GDS.
Should it maintain its position as a focused, central unit working in its current way but gradually
influencing more parts of the public sector? Or should it see its role as building up a digital capability
across other government departments – and gradually working itself out of a job?

Page 9
CS-18-28
Exhibit 1: Homepage of GOV.UK website
Exhibit 2: www.gov.uk/transformation

Page 10
CS-18-28
Endnotes
1 See following recent practice-based articles for an update on agile working in organisations:
Birkinshaw, Julian. “What to expect from agile.” MIT Sloan Management Review 59.2 (2018): 39-42.
Rigby, Darrell K., Jeff Sutherland and Hirotaka Takeuchi. “Embracing agile.” Harvard Business Review 94.5 (2016):
40-50.
Wouter Aghina, Aaron De Smet, Gerald Lackey, Michael Lurie and Monica Murarka. “The five trademarks of agile
organizations,” January 2018. McKinsey Quarterly. By.
2 www.ft.com/content/1621d0aa-f7d9-11e3-baf5-00144feabdc0#axzz3DCt18Au2

Page 11

The UK Government Digital Service

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The UK Government Digital Service

Uploaded by

Copyright:

Available Formats

For the exclusive use of i. mabrouk, 2019.

The UK Government digital service

Experimenting with a new approach to software development

© 2018 London Business School

Finding an Executive Director

© 2018 London Business School

Creating the GDS culture

Another priority was to build the right team of people:

© 2018 London Business School

1. Start with needs: user needs, not government needs.

© 2018 London Business School

Working across government

© 2018 London Business School

and with a handful of large-system integrators. So, to tackle this, we decided to

The official release of GOV.UK

© 2018 London Business School

Media attention and resistance

Pondering the future of GDS

© 2018 London Business School

David Thorpe, Director of Policy at GDS, offered his perspective:

© 2018 London Business School

Exhibit 1: Homepage of GOV.UK website

© 2018 London Business School

© 2018 London Business School

You might also like