Application process
Civil Service 250-Word Behaviour Statement Examples: A Fail and a Pass, Side by Side
Published
The 250-word behaviour statement is where most Civil Service applications die. Not at interview — at the sift, where a panel member reading at speed scores your statement on the same 1–7 scale your interview will use, against the same grade-level descriptors, usually with a pass mark of 4.
The good news: the difference between a failing statement and a strong one is almost never the underlying example. It's the telling. This article gives you the structure, the word budgets, and — because seeing it beats being told about it — the same statement written twice: once at a 3–4 that fails, once at a 6 that doesn't. Annotated, so you can apply the difference to your own draft tonight.
How the sift actually reads your statement
Three facts change how you should write:
Sifters read fast. In a large campaign, a panel member may spend two or three minutes per statement, scoring dozens in a sitting. Evidence that has to be inferred doesn't get scored — the ownership, the decision, the reasoning all have to be explicit on the page.
They score against descriptors, not impressions. The question in the sifter's head is "can I point to the line that meets this descriptor?" Your job is to write pointable lines.
Most of your competition sounds identical. The median statement says, in effect, "I am reliable and hardworking," with 80 words of scene-setting and a vague result. A statement showing one clearly-owned decision stands out within three sentences.
The structure and word budgets
For roughly 250 words:
- Situation + your role (~40 words). What was happening, what was at stake, and what was yours — "I was responsible for", "I owned", not "I was part of a team that". Two or three sentences maximum.
- Actions (~140 words). The scoring zone. Two or three actions, each carrying: what you did, how, and why you chose it over the alternative. The word "because" is where marks live. At least one action should be a decision that was genuinely yours; at least one should involve another person — someone persuaded, coached, or handled.
- Result (~50 words). Quantified wherever possible — numbers, percentages, time saved, feedback received. Small numbers still count. A result that outlived the task (a process others now use) is premium evidence.
- Reflection (~20 words, optional). One sentence, pitched at the grade — about judgement or ownership, not "the importance of communication."
The most common structural failure, by a distance: 80 words of context, 50 of action. Flip it. The sifter already knows workloads are high and stakeholders can be difficult.
The same statement at a fail and at a pass
Behaviour: Working Together. Both versions below describe the same underlying events, both are ~250 words.
FailVersion 1 — scores around 3–4/7
"In my current role, our team relies on receiving monthly data from a partner organisation to complete our reporting. Over several months the data was arriving later and later, and often in an inconsistent format, which caused significant problems for our reporting deadlines and created tension between the two teams. Something needed to be done to get the relationship working better.
I have always believed good working relationships are essential, so I made an effort to improve things. I communicated regularly with our contact at the partner organisation, keeping things professional and friendly even when the data was late. I made sure to thank them when things went well and remained patient when they didn't. When problems arose I raised them politely through email, copying in my manager so everyone was aware. I also attended a joint meeting between the two teams where the issues were discussed openly, and I contributed to the discussion about how things could improve.
As a result of these efforts, the relationship between the teams improved and the data began arriving more consistently. Our reporting deadlines became easier to meet and there was noticeably less tension in our interactions. My manager commented that the relationship was in a much better place. This experience showed me the importance of maintaining good relationships with partners even when things are difficult."
Read it as a sifter. It's pleasant, professional — and nothing in it is attributable. The candidate "communicated regularly", "remained patient", "contributed to the discussion": all disposition, no action a scorer can point to. The one concrete event (the joint meeting) was someone else's — the candidate attended it. The problem apparently improved on its own, "as a result of these efforts", with no mechanism visible and no numbers. And the friction — the entire reason this example is a Working Together story — is handled by copying in my manager.
PassVersion 2 — scores around 6/7
"My team's statutory reporting depends on monthly data from a partner organisation. Over a quarter, their submissions became later and less consistent — we missed one internal deadline and tensions were rising. I owned the relationship day to day, so I decided to fix the cause rather than keep managing the symptoms.
Instead of another chasing email, I asked our contact for thirty minutes and started by asking what the process looked like from her side — because I suspected the problem was upstream of her, not carelessness. It was: our data specification had grown over two years into something her team partly guessed at, and their busiest week collided with our deadline. That conversation changed my approach entirely. I redrafted our specification with her, cutting three fields we no longer used and adding examples for the two that caused most errors, and I proposed shifting our deadline four days — I checked first that our own reporting could absorb it, and it could, comfortably. I took the revised arrangement to both managers as a joint recommendation from the two of us, which meant it landed as a partnership fix rather than one team blaming the other.
The next three submissions arrived on time and error-free, our reporting has hit every deadline since, and the joint-review approach is now how both teams handle specification changes. What I learned: when a partner keeps failing you, the fastest fix is usually to ask what you're doing that makes their job harder."
What a sifter can now point to: a relationship owned; a deliberate choice ("instead of another chasing email… because"); genuine perspective-seeking that changes the approach — descriptor language, made concrete; a concession checked and offered; friction converted into a joint recommendation rather than escalated; a quantified, durable result; and a reflection above "relationships matter."
Notice what didn't change: the events. The candidate did all of this in both versions — Version 1 just never said so.
Five rules that move a statement from 4 to 6
- Every key action is "I". "We agreed a new process" is unscoreable. Who agreed it? If it was you, say so.
- One decision with a rejected alternative. "I chose X over Y because…" — the single highest-value sentence pattern in the format.
- Name the friction, and resolve it yourself. A sceptic, an unhappy stakeholder, a pushback — handled by you, not routed upward.
- Quantify the result. "Improved significantly" is a claim; "45% fewer repeat contacts in two months" is evidence.
- Your manager appears at sign-off, not at the decision. From EO upward, "I suggested it to my manager, who took it forward" hands your marks to your manager.
Grade calibration: the same statement scores differently by grade
The 1–7 scale doesn't move, but the descriptors do — and a statement that scores 6 at EO can score 3 at SEO. At EO, sifters want ownership and initiative within your own work. At HEO, an end-to-end piece of work with the decisions clearly yours. At SEO, trade-offs, leading others, and risk owned. At Grade 7, direction set, senior people influenced, decisions made on incomplete evidence. If you're writing statements for a specific grade, our complete guides for EO, HEO, SEO and Grade 7 each include a full statement rewritten fail-to-pass at that grade's calibration.
The fastest way to improve a statement is to see where it actually lands before a sifter does. CS Interview Coach's AI scores your written answers on the same 1–7, grade-calibrated rubric — and tells you which of the five rules above you're breaking. Try it free, or get every grade guide, the 250+ question bank and unlimited practice with Interview Sprint at £24.99/month.
Common questions
Roughly: situation and your ownership in about 40 words, two or three actions with the reasoning behind each in about 140 words, a quantified result in about 50 words, and optionally one reflection sentence. The most common structural failure is spending half the words on context.
Yes, but weighted heavily toward Actions. Situation and Task together should be two or three sentences at most — sifters score the decisions and actions that were yours, not the scene-setting.
'I', consistently. Sifters can only score evidence attributable to you. 'We delivered the project' is unscoreable; 'I re-planned the schedule and delegated the analysis' is evidence.
On the same 1–7 scale as your interview, against the grade-level behaviour descriptors, usually with a pass mark of 4. Sifters read at speed — evidence that has to be inferred effectively doesn't exist.
You can, but you shouldn't go far under. The word limit is your evidence budget — a 160-word statement has voluntarily surrendered a third of its scoring opportunity.
Yes — application systems typically cut you off. Draft long, then cut context before you ever cut actions.
Across different behaviours, yes — a strong piece of work usually flexes across two or three by shifting emphasis. Don't reuse the same statement verbatim at sift and interview for the same behaviour; panels notice.
Different formats scored differently — a personal statement assesses your experience against the advert's essential criteria, and forcing STAR onto it wastes your word count.
Usually 4 to pass, but appointments run in merit order, so write for a 6 — the same example told with explicit ownership, a named trade-off and reasoning attached to every action.
Practise before the real thing
Reading how panels score is step one. The candidates who pass are the ones who practised saying their answers out loud — and got scored feedback against the official Civil Service Behaviours framework before interview day.
Start Practising Free3 free practice questions · No card needed · Cancel anytime