Erudite Digital Team

Blog › Guide

How to write a policy brief that decision-makers actually read

By Eliakim Anzizi, PhD ·

A policy brief has one job: to help a busy decision-maker understand a problem and decide what to do about it. The reader might be a minister's adviser, a legislator, a programme director or a development partner. They may give your brief five minutes, often less.

Many briefs fail because they are shortened research reports. They describe the study, list the findings and leave the reader to work out what it means. A good brief does that work for them. This guide explains how.

What is a policy brief?

A policy brief is a short, stand-alone document that presents evidence on one issue and recommends action. The International Development Research Centre (IDRC) describes the best briefs as clear and concise stand-alone documents that focus on a single topic.

It differs from a journal article or research report. Those explain how knowledge was produced. A policy brief explains what the knowledge means for a decision and what should happen next.

Know your reader and your ask

Before you write a word, answer three questions:

  • Who decides? Name the office or person, not "stakeholders".
  • What do they already know and believe? A security agency, a finance ministry and a donor read the same evidence differently.
  • What can they actually do? Recommend action within their mandate, budget and timeline.

Then write your purpose in one sentence: "We want [this reader] to [take this action] because [this evidence shows this]." Keep that sentence in view. Anything that does not serve it can go.

How long should a policy brief be?

Shorter than you think. IDRC advises that a brief should not exceed 1,500 words or two pages. A guide published by ODI notes that policy briefs are usually two to four pages, or 1,000 to 2,000 words.

If you have more to say, put the detail in an annex or link to the full report. The brief itself should be readable in one sitting.

A structure that works

  1. A title that states the message. "Community policing forums can rebuild trust in insecure areas" tells the reader more than "An assessment of community policing".
  2. Key messages. Three to five short points at the top. Many readers will stop here, so make them complete. Write this section last.
  3. The problem and why it matters now. Who is affected, how badly, and what happens if nothing changes.
  4. What the evidence shows. The main findings, with a sentence or two on where they come from.
  5. Policy options. Two or three realistic options, with their benefits, costs and risks.
  6. Recommendations. What should be done, by whom and when.
  7. Sources and contact. Key references, the author, the date and who to call for more.

Use evidence a busy reader can trust

  • Lead with the finding, not the method. Summarise the method briefly and put the detail in an annex.
  • Give a source and a date for every figure.
  • Be honest about limits. Saying what the evidence cannot show makes the rest more credible.
  • Use local evidence where you have it. Decision-makers trust data from their own context.
  • Use one or two clear charts rather than a large table.
  • Present the evidence fairly, including findings that complicate your case.

Write recommendations people can act on

Recommendations are the part of the brief that decision-makers act on, so make each one specific, feasible and linked to the evidence. Name who should act, what they should do and by when. Keep them few: the ODI guide suggests they are usually limited to three, and it separates implications (what could happen) from recommendations (what should happen).

Compare these two:

  • Weak: "Stakeholders should strengthen emergency response."
  • Stronger: "The state emergency management agency should publish a flood response plan for its highest-risk local government areas before the next rainy season, funded from its existing budget."

The second one tells a specific reader what to do and makes it possible to check later whether it happened.

Make it easy to read

  • Use plain language and short paragraphs. Define every acronym at first use.
  • Write headings that say something: "Most reported incidents go unrecorded" rather than "Findings".
  • Prefer the active voice: "The ministry should publish…" not "It is recommended that publication be considered…".
  • Keep bullet lists short; IDRC suggests no more than five to seven points.
  • Leave white space. A crowded page looks like hard work.

Common mistakes to avoid

  • Cutting down a full report. IDRC advises drafting a new, purpose-driven brief instead.
  • Burying the ask. If the recommendation first appears on page four, many readers will never see it.
  • Too many recommendations. Ten recommendations usually means none is a priority.
  • Too much method and jargon. Decision-makers need to know the evidence is sound, not every step of the analysis.
  • Recommendations the evidence does not support. Overstating your case damages trust in everything else.
  • Ignoring cost and feasibility. An unaffordable or politically impossible option will be set aside.
  • Poor timing. A brief that arrives after the budget is set or the decision is made has little effect.
  • No date, author or sources. Readers need to know how current and how reliable the brief is.

Test it before you send it

Try to explain the brief in 20 seconds. If you cannot, the message is not yet clear. Then ask a colleague who knows nothing about the topic to read it and tell you what they think the reader should do. If their answer matches your purpose sentence, the brief is working.

Finally, check every figure, name and reference, and proofread in separate passes. Our editing and proofreading checklist covers this step. If your brief draws on survey data, our guide to choosing a data analyst may also help.

Quick checklist

  • One issue, one named reader, one clear ask
  • Two pages or close to it
  • Title and key messages state the conclusion
  • Every figure sourced and dated
  • Limits of the evidence stated
  • A few specific, feasible recommendations
  • Plain language, clear headings, few acronyms
  • Tested on a non-specialist reader and proofread

Sources

Writing a policy brief on governance, security or public policy? Eliakim Anzizi, PhD in Defence and Strategic Studies, helps organisations turn research into clear briefs with practical recommendations. See our policy analysis and report writing services, then contact us or message us on WhatsApp.