How to write a lay summary for a research proposal or grant application

A proposal lay summary is not a short abstract. Reviewers, panel members from other fields and sometimes the public read it to decide what you are asking money for. It is written before the work exists, so the usual advice for finished papers needs three changes: tense, claims and impact.

Always follow the funder's own instructions first. Word limits, headings and tense differ between funders, and the call text wins over anything here. This page is a method, not a rule book for any single funder.

1. Write in the tense of the work, not the result

A proposal describes work you will do, so use "we will" and "we aim to find out whether". Do not write "we found" or "this shows". A lay summary of a proposal that sounds like a finished result is the commonest way to overclaim by accident. If the portal asks for a summary of a finished project (a final outcomes report, for example), switch to past tense and report only what you actually found.

2. Lead with the problem, then the question

Start with the thing in the world that is not understood, in words a neighbour would use. Then give your question as one plain sentence. Compare:

Too technical: We will characterise the thermal degradation kinetics of lithium-ion separators under cycling.

Plainer: Batteries can fail when their inner layers get too hot. We will test how those layers age with use, to understand why some fail early.

This pair is an invented illustration, not from a real proposal. The plainer version drops the technical detail on purpose; the full detail belongs in your scientific abstract.

3. Say what you will do in two or three steps

Name the method by what it does, not by its name: "we will ask 300 adults to keep a sleep diary for four weeks" beats "we will conduct a longitudinal self-report study". Keep your real numbers (sample size, duration, number of sites). If a number is still provisional, say "about" or "up to" rather than rounding it to something safer-sounding.

4. Be careful with impact

Many funders ask who could benefit. This is where proposal summaries drift into promises. Use the strength your plan supports:

• "could help" or "may inform" for results that depend on things outside your control;
• a named group (for example "clinic staff", "small farms", "teachers of Year 5") instead of "society" or "millions of people";
• no new statistics. If the impact figure is not in your proposal body with a source, it does not belong in the summary.

If you cannot honestly say who benefits yet, say what the project will make possible next. A modest honest sentence reads better to reviewers than a broad one they will doubt.

5. Fit the word limit by cutting in this order

1. Background beyond one sentence.
2. Method detail that a reader cannot picture anyway.
3. Second and third outcomes, keeping the main aim.
4. Jargon you had to define (replace it instead of defining it, if the meaning survives).

Count the words the way the portal does. Some portals count hyphenated words as two, and some count headings. Aim a few words under the limit so a counting difference does not block submission.

6. Check it before you submit

• Is every tense consistent with work that has not happened yet?
• Does every number match your proposal body?
• Are acronyms spelled out or removed?
• Does the average sentence stay near 15 to 18 words? Paste it into the free lay summary checker to see its word count, an estimated reading grade (grade 7 to 9 is a common target for a general adult reader), long sentences, jargon and hype words. It runs in your browser and sends nothing anywhere.
• Could someone outside your field say back what you will do and why?

A worked example (invented)

The project below is made up for demonstration. It is about 90 words.

Many rural clinics keep patient appointments on paper, and missed appointments are common. We will work with six clinics in one region to try a simple text-message reminder and compare missed appointments over six months with six similar clinics that keep their usual system. We aim to find out whether reminders reduce missed appointments, and by how much. If they do, clinic managers could use the results to decide whether a low-cost reminder system is worth setting up. We do not yet know whether the results will apply outside this region.

Notice the tense ("will work", "aim to find out"), the real numbers (six clinics, six months), a named group (clinic managers) and an honest limit in the last sentence.

Want a first draft from your own text?

If you already have a proposal abstract and a word limit, the $7 AI-written lay summary draft turns it into a summary that keeps your numbers and hedges, lists the jargon it replaced, and lists what your text does not say (such as who benefits) as blanks for you to fill. It is AI-written, so you must read and check it, and it does not write your proposal. A free chatbot can also do a first pass; this page's checks apply to either. See also the general lay summary guide, fill-in templates and before-and-after examples.