How to Write a Policy Brief

A policy brief has perhaps two minutes of a busy reader’s attention, and everything must survive that constraint.

A policy brief translates research into something a decision-maker can act on. Its audience is rarely expert in your field, usually short of time, and answerable to constituencies you do not control. Those three facts shape every choice in the document — and explain why an excellent literature review makes a poor brief.

Identify your reader precisely

“Policymakers” is not an audience. A minister, a civil servant drafting regulation, a local authority officer and an NGO campaigner need different framings, different evidence and different asks. Decide which you are writing for and what specific decision sits in front of them.

That decision determines relevance. Evidence that would be central to an academic paper may be irrelevant here if it does not bear on the choice being made.

Lead with the problem, stated as a cost

Open with what is going wrong and what it costs — in money, in outcomes, in people affected. A brief that opens with background loses its reader before the argument arrives.

Quantify wherever you can. “Delayed discharge affects thousands of patients” is weaker than “delayed discharge accounted for 1.2 million occupied bed-days last year, at an estimated cost of £550 million”.

Keep it genuinely brief

Most policy briefs run to two to four pages, and the constraint is real rather than stylistic. Length signals that you have not identified what matters. If your material will not fit, the problem is usually that you are including evidence for its own sake rather than because it changes the decision.

Structure for navigation

Readers move through a brief non-linearly, so the structure must support scanning.

• Title — states the issue and hints at the recommendation.
• Summary box — three or four bullets, readable alone.
• The problem — scale, cause, who is affected.
• The evidence — what research shows, briefly.
• Options — realistic alternatives with trade-offs.
• Recommendation — what to do, by whom, when.

Present options, not a single answer

Decision-makers distrust briefs that offer only one course of action, because real choices involve trade-offs between competing goods. Present two or three genuine options, assess each against consistent criteria — cost, feasibility, speed, political acceptability, distributional effect — and then recommend.

Include the status quo as an option where it is credible. Showing what happens if nothing changes is often the most persuasive part of a brief.

Handle evidence honestly

The temptation in advocacy writing is to present the evidence as more settled than it is. This is both dishonest and tactically poor: a reader who discovers a significant contrary finding you omitted will distrust everything else.

State the strength of the evidence plainly. “Three randomised trials support this; all were conducted in urban settings, and the effect in rural provision is untested” is a sentence that builds credibility rather than undermining it.

Write for a non-specialist without condescending

Define technical terms on first use, avoid disciplinary shorthand, and explain methods only where the method affects how much weight a finding deserves. Your reader is intelligent and busy — not ignorant.

Avoid academic hedging conventions that read as evasion outside the university. “The data suggest a possible association” may be appropriate in a journal; in a brief it reads as having nothing to say.

Use visuals that carry an argument

One well-chosen chart often does more than a page of prose. Choose it to show the thing that matters — the trend, the gap, the comparison — and caption it so it can be understood alone. Resist decorative graphics; a brief with three charts and no argument is worse than one with none.

Be specific about the ask

End by stating exactly what you want the reader to do: which body should act, under what power, on what timeline, at what cost. A brief that concludes with “further consideration is needed” has wasted its own argument.

Where implementation depends on something outside the reader’s control, say so. Naming the dependency is more useful than pretending it does not exist, and it demonstrates that you understand how the decision actually gets made.

Anticipate the political constraints

Policy recommendations exist inside a system of budgets, statutory powers, electoral cycles and institutional interests. A brief that ignores these reads as naive to the people who have to act on it. You do not have to endorse the constraints, but acknowledging them shows you understand how a decision actually gets made.

Where your preferred option is politically difficult, say what would need to change to make it possible. That is more useful than pretending the difficulty does not exist.

Name the beneficiaries and the losers

Almost every policy redistributes something. Briefs that describe only benefits invite the obvious question of who pays. Stating the distributional consequences — which groups gain, which bear costs, and whether compensation is feasible — strengthens the analysis and pre-empts the first objection a reader will raise.

Check the format requirements

Many organisations and modules specify a template: word count, section headings, whether references are permitted, whether an executive summary is separate from the summary box. These are usually enforced strictly because briefs are read in volume and consistency aids comparison. Read the specification before drafting rather than reformatting afterwards.