Get your Civil Service application through the sift

When the department announced a move to digital-first claims, I led a review of how my team's paper process would be affected. I read the strategy and spoke to the policy team to understand what it was meant to achieve: faster decisions for claimants and lower handling costs. I mapped our 14 manual steps against it and found three that would still need paper because of a legal signature requirement. Rather than wait for the national design, I proposed a short interim process to my Grade 7, with the risks set out, and briefed my team on why the change mattered for claimants. The interim process went live a month before the national system and cut our average decision time from 12 to 8 days. I learnt to read a policy for its purpose, not just its instructions.

In my last role, customer refund requests were tracked in three separate spreadsheets, and requests regularly went missing. I suggested a single shared tracker and asked my manager for two weeks to trial it. I built the tracker with a status column and a weekly check, wrote a one-page guide and showed each colleague how to use it, listening to their concerns about extra admin. Two colleagues worried about duplicate entries, so I added a simple rule that the first person to log a request owns it. After the trial, missed requests fell from around ten a month to none, and the team adopted the tracker permanently. I now ask the people doing the work what slows them down before I suggest a change.

As project lead for a supplier contract renewal, I had six weeks to recommend whether to extend the existing contract or go to tender. I gathered two years of performance data, spoke to the three teams who used the service, and asked commercial colleagues about the market. The data showed the supplier met 92% of its targets but its costs had risen 18%. I set out both options with their costs, risks and impact on users, and tested my analysis with finance before it went to the board. I recommended a one-year extension with tighter targets while we prepared a tender, and explained the trade-off openly. The board agreed, saving an estimated £40,000 against an immediate tender, and the tender launched on schedule the following year.

When I took over a team of eight after a restructure, morale was low and two people had asked to move. I met each person individually to understand their concerns, then agreed a clear set of priorities with my Deputy Director so the team knew what mattered and why. I set up a fortnightly session where anyone could raise problems, and made sure quieter colleagues had the chance to lead parts of it. When a colleague with caring responsibilities struggled with meeting times, I moved our core meetings to the middle of the day. Within six months the team's staff survey engagement score rose from 48% to 71%, nobody left, and we delivered all four of our business plan commitments. I learnt that people follow a direction they helped to shape.

I needed to persuade regional managers to adopt a new reporting template that they saw as extra work. Before the meeting, I spoke to two of them to understand their objections: the time it took, and reports they felt nobody read. I cut the template from four pages to one, and prepared a short example showing how their figures would be used in the monthly briefing to ministers. In the meeting I started with the problem the template solved for them, answered concerns directly and agreed a one-month trial. I followed up in writing with a summary and a named contact. Nine of the eleven regions used the template in the first month and all eleven by the third, and I shared the ministerial briefing back so they could see the result.

During a peak in casework, our team and the neighbouring team were both falling behind, and we were competing for the same specialist adviser's time. I suggested to both managers that we share a single queue for questions to the adviser, and offered to set it up. I agreed simple rules with colleagues from both teams so the most urgent cases went first, whoever raised them, and I put my own notes on common questions on our shared drive so people could answer simpler queries themselves. Over the next two months, the wait for an answer from the adviser fell from five days to two, and both teams cleared their backlog before the deadline. Several colleagues added their own notes to the guide, and it is still in use.

When our team began using data dashboards, I realised that neither I nor most of my colleagues could build or read them confidently. I completed the department's online data course and practised on our own team's figures. I then ran short lunchtime sessions, starting with what each person needed for their own work. I paired with two less confident colleagues and built their first dashboards alongside them, then stepped back so they could do it on their own. Within three months, five of the seven team members were producing their own monthly dashboards, saving our analyst about a day a month. One of the colleagues I supported went on to a data role. I keep a list of the skills I want to build next and review it with my manager each quarter.

I answered written enquiries from the public, and complaints showed that people found our replies hard to understand. I read a sample of 50 replies and the complaints about them, and found that most problems came from technical terms and missing next steps. I drafted plain English templates for the ten most common enquiries, each ending with what happens next and when, and tested them with two colleagues and a member of our public feedback panel. I agreed the wording with my manager and our policy lead so it stayed accurate. After three months, complaints about clarity had fallen by 60%, and the average time to reply dropped from ten days to six because fewer people wrote back asking what we meant.

Two weeks before a statutory deadline, a colleague left unexpectedly and I took on her 120 cases on top of my own. I listed every case by deadline and complexity, agreed with my manager which lower-priority work could pause, and set myself daily targets. I cleared the simpler cases first in focused blocks, flagged the 15 complex ones early so a senior colleague could take two of them, and updated my manager every morning on progress and risks. When a system outage cost me a day, I adjusted the plan and paired with a colleague to check each other's cases quickly. All 120 cases were completed before the deadline, with no errors found in the quality check. I learnt that early, honest updates make it much easier to get help in time.