GOV.UK defines this behaviour as: "Show pride and passion for public service. Create and engage others in delivering a shared vision. Value difference, diversity and inclusion, ensuring fairness and opportunity for all." (Success Profiles: Civil Service behaviours, GOV.UK, Open Government Licence v3.0.) Notice what the definition does not say: managing people. Panels score Leadership from candidates with no reports at all — the behaviour is direction, engagement and fairness, wherever you sit.
This page covers the forms the question takes, with four marked model answers deliberately spread across the spectrum: leading without authority, leading a team under strain, resetting direction after change, and the inclusion clause that candidates most often leave unevidenced.
That last point deserves emphasis. The definition’s third sentence — valuing difference, ensuring fairness — is scored, not decorative. Panels increasingly ask for it directly, and an answer with a concrete fairness mechanism beats any amount of sincere phrasing.
Practise these questions out loud with a live AI interviewer — free. Start with Try Me
How the marking guidance works
Each model answer below is marked against the four criteria a live aurate session scores:
See how a full session is scored
aurate is a practice tool. Marking guidance describes what strong practice answers show — it isn't an employability assessment.
Why it's asked: The purest form of the behaviour, and the standard opener for candidates without line management. Panels want to see what you substituted for authority: clarity, credibility, and making the work's purpose vivid enough that people opted in.
A cross-team data-quality problem had been bouncing between three teams for months — each owned a third of the pipeline, nobody owned the join. I had no authority over any of them; I was the analyst who kept finding the errors downstream.
I stopped circulating error reports and instead asked one person from each team for a single 45-minute session, framed around a question rather than a complaint: 'what would make this stop being all of our problem?' Before the session I traced twelve failed records end to end, so we argued from specimens instead of impressions. The join errors turned out to cluster at two handoffs with no agreed format.
I proposed we three become the standing owners of those handoffs — a shared two-page format spec and a monthly half-hour check, which I offered to run. All three said yes, mostly, I think, because the ask was small and the specimens made the cost of drift undeniable. Join errors fell from a weekly occurrence to two in the following quarter, and when a fourth team later joined the pipeline, the other two owners inducted them without me — which is when I decided it had actually worked.
Marking guide
Why it's asked: The strain form. Panels are listening for what you protected (standards, people, priorities), what you consciously let slip, and whether your team ended the period more or less willing to follow you.
For six weeks our five-person processing team ran at three — one maternity leave, one long-term absence — while intake stayed flat. The instinct in the team was to quietly work longer and say nothing; the previous cover period had run on goodwill until it snapped.
I led it differently in three ways. First, I made the trade-off explicit instead of private: I ranked our four work types with our manager in the first week, got her sign-off that the bottom category would build a visible backlog, and put that in writing to the team — so nobody was carrying a secret failure. Second, I moved the daily check-in from progress-chasing to load-balancing: ten minutes, one question — 'what's about to fall?'. Third, I protected one full training afternoon that the team assumed would be cancelled, because our newest member was three weeks in and abandoning her induction would have traded six weeks of strain for a year of struggle.
We held the top three categories inside standard the whole period. The backlog in the fourth peaked at nine days and was cleared within a fortnight of returning to strength — and the team's end-of-quarter feedback specifically named the explicit trade-off as the thing that made the period feel run rather than endured.
Marking guide
Leadership answers sound better than they hold up
This behaviour draws the gap between rehearsed and real: panels push on the moment it cost you something. A live aurate session applies that pressure to your actual examples and marks the result on the same four criteria used above. Two free sessions. No credit card.
Try it freeWhy it's asked: The reset form — common after restructures, mergers and policy pivots. The panel knows you cannot make people like a change; they are scoring whether you gave it honest meaning, absorbed the anger without deflecting it, and found the team a real stake in the new direction.
After a restructure, my new team of eight was an unhappy merger of two units that had each lost their old remit — one group grieving a policy area they'd built careers in, the other convinced the merge was a takeover. The new remit was genuinely important and nobody cared.
I resisted the launch-workshop instinct and did three slower things. I met everyone one-to-one first and asked two questions — what should we keep from how your old team worked, and what should we never do again — then published the combined answers, attributed only to 'us', as our working principles. I was straight about what was gone: I never called the change an opportunity in the first month, because for half the room it was a loss and they knew it. And I found each person one piece of the new remit that touched their old expertise, so the first quarter's work built on identity rather than erasing it.
Six months in, the team's engagement survey had moved from bottom-quartile to median — not triumphant, but honest progress — and two of the strongest sceptics had voluntarily taken the new remit's hardest briefs. One told me later the turning point was the month I didn't call it an opportunity.
Marking guide
Why it's asked: The inclusion clause of the definition, asked directly — panels are scoring a mechanism, not a sentiment. Strong answers change how something works (meetings, sifts, allocation) so fairness stops depending on anyone's goodwill, including yours.
Our project's design decisions were made in a weekly meeting that rewarded exactly one style: quick, confident, spoken. I noticed — because the minutes showed it — that two of our strongest analysts had gone a full quarter without a recorded contribution, while their written work was consistently the best evidence in the room.
I changed the mechanism rather than exhorting people to speak up. Decisions now went round as a one-page note 48 hours ahead, with written comments welcomed before the meeting and every comment addressed in it by name; and for contested calls I added a silent first pass — everyone writes their position for two minutes before anyone speaks, and the chair reads the range aloud before discussion starts.
The effect was immediate and slightly humbling: the first silent pass on a contested call produced a majority against the option the loudest three of us had been about to wave through, for a reason none of us had raised. Within a quarter, recorded contributions covered the whole team, and the two analysts were presenting their own options papers. The format was adopted by a sister project after their lead sat in on one of ours.
Marking guide
Why it's asked: A calibration question panels use to check you've understood the level. Strong answers distinguish doing from directing from developing — and land on the target grade's mix, ideally with a sentence on how the role's context (frontline, policy, programme) shifts it.
Why it's asked: The loyalty-under-disagreement form. Panels want the sequence: challenge made honestly upwards, decision accepted, direction delivered without poisoning the team — and no daylight between what you said upstairs and what you said to your people.
Why it's asked: Leadership overlapping with developing self and others — asked here, it targets whether growing people is part of how you lead or an HR activity you tolerate. One named person, one specific investment, one visible step forward beats any framework.
Why it's asked: The weather-setting form: incidents, bad news, public mistakes. Panels are scoring the gap between your values and your behaviour at the worst moment, because that gap is what your team actually learns from.
Why it's asked: The strengths-format variant of the definition's first clause. It is scored on authenticity — under a minute, one honest moment where the public purpose of the work actually moved you, told plainly. Rehearsed civic language reads instantly and scores accordingly.
Yes — the Success Profiles definition never mentions line management. Panels score direction, engagement and fairness, all of which can be evidenced from project work, cross-team problems or peer situations. The influence-without-authority form of the question exists precisely for this, and a strong answer there is full-mark territory at most grades.
With a mechanism, not a statement of values: something you changed about how work, meetings or decisions operate so that fairness stopped depending on goodwill. Pre-reads that value written thinkers, structured input rounds, fairer allocation of stretch work — one concrete change with a visible effect answers the clause properly.
Working together is scored on the partnership itself — building relationships, sharing information, delivering jointly. Leadership is scored on direction and engagement: people moved somewhere because of you. The same project can evidence either, but the Action section must change focus — collaboration mechanics for one, direction-setting and followership for the other.
Yes for the 'tell us about a time' forms, with one adjustment: give the Result a followership beat. Numbers matter, but this behaviour's strongest closing evidence is people choosing to continue — adopting the approach, volunteering, asking for more — after the pressure that made it necessary had passed.
An honest, partially successful account often outscores a triumphant one, because the behaviour is tested hardest when things resist. Bottom-quartile to median is a credible arc; zero to hero is a suspicious one. Keep the real ending and spend your final sentence on what the team did next, not on redeeming the story.
Two free sessions. No credit card.