Discovery with 77 civil servants across the UK to settle one question: Office 365 or G Suite?
A major tooling decision risked missing the needs of distributed teams and users.
Researched users in different locations and created personas based on working context.
Highlighted accessibility, regional support, and change management needs.
Helped shift the focus from tool choice to rollout and service support.
The Department for Communities and Local Government (DCLG) was preparing for a major IT modernisation: a 16-month change programme that touched 77+ users distributed across England and 11 arm's length bodies, with a central tooling question: Microsoft Office 365 or Google G Suite.
Rather than treating that as a procurement question, the project ran a discovery to understand how DCLG users actually worked. The aim was to build user profiles that could ground the tooling decision in real evidence, and inform the rollout that would follow.
My role as a service designer and user researcher was to help shape and run the research, take part in contextual visits across the country, contribute to synthesis, and help build the personas the programme could use afterwards.
DCLG is not one office. Users are distributed across London, the South West, the Midlands, the North West, and beyond. Work happens at home, on trains, in regional offices, and at external sites. Teams cross departmental lines (DCLG, PINS, HCA, BEIS) and rely on each other through systems that do not link up.
The IT estate had not had a substantial refresh in a long time. Email had become a default for everything: communication, document storage, collaboration, audit trail. Lync adoption was patchy. Pockets of teams were quietly using SMS or WhatsApp on personal devices because the official tools did not feel reliable. Regional offices reported substandard kit, and accessibility issues sat in long support queues.
Moving straight to procurement without understanding all of this risked buying a tool that solved the wrong problem, or buying the right tool and failing on the rollout. The team needed evidence that could shape both the decision and the change programme around it.
Discovery combined depth and breadth: contextual visits in regional offices and London HQ, interviews across DCLG, the Planning Inspectorate, the Homes & Communities Agency, and BEIS Local Growth. We deliberately went to where the work happened (Truro, Warrington, Birkenhead, Bristol) so the picture would not be London-centric. Findings were synthesised into themes, then layered into personas and attribute frameworks so the programme could see both the patterns and the variation underneath them.
Email was the most-used tool in DCLG, carrying not just messages but document storage, version control, and audit trails. Pockets of teams were quietly using SMS or WhatsApp on personal devices for urgent messages because official channels did not feel reliable. Lync adoption was patchy. The bigger problem was not the absence of a chat tool. It was the cultural weight email had taken on.
Most users we spoke to worked from home at least one day a week and felt IT made that reasonably possible. But top-down support varied, hot-desking was rare, and users did not see the connection between working from home and giving up an office desk. The constraint was as much management style and office culture as it was technology.
Work and project teams crossed departments, organisations, and locations. DCLG worked closely with arm's length bodies and external partners, but IT systems did not link up and were not accessible to people outside the department. Amalgamated teams in shared offices used different IT systems, making file sharing, calendars, and distribution lists harder than they needed to be.
Most users had more modern devices and software at home than they did at work. Newcomers had often experienced more advanced systems elsewhere. Laptops were heavy, port replicators unreliable, and monitors (particularly in regional offices) old and small. The gap between expectation and provision shaped how users felt about their employer.
A hardware and software refresh would be welcomed. But many users said they did not have the time or opportunity to learn new systems at rollout, and that support tailed off when they actually needed it. Regional offices, accessibility users, and people who joined after rollout were the most affected. Throwing new tools at the department without addressing this would not change much.
The discovery moved DCLG from a tools-first procurement question to a service-shape decision. It produced a shared picture of how 77 users across the department and its arm's length bodies actually worked, and where the rollout would need to do the heavy lifting.
Seven personas with multi-layered attributes became the reference frame for the programme. Regional disparity, accessibility, and the IT Focal Point role were elevated from afterthoughts to design considerations. The recommendation against picking a 'better' suite, and in favour of investing in how change was managed, was a more difficult message to deliver than a tool recommendation, but a more useful one.
The work fed directly into the next phase of the programme, including a second wave of quantitative research, deeper investigation into line-of-business tools, and the change approach that would carry across DCLG and its arm's length bodies.