
Dr Sana Mirza, a veterinary researcher in Faisalabad, had finished a study on mastitis control in small dairy herds when the funding agency asked for one more document: a research impact statement of no more than 300 words. She stared at the blank form for an hour. Her paper described what she had found. The form wanted something else entirely — what had changed because of it.
That confusion is common, and it is the reason most impact statements fail. An impact statement is not a summary of your research. It is a short, evidence-backed account of a change outside your laboratory or library that happened because of your research — a policy rewritten, a practice adopted, a cost cut, a disease reduced.
This guide explains what it is, who reads it, and how to write one a busy panel member can believe in five minutes.
An impact statement is a short narrative, usually 150 to 500 words, that answers one question: what difference did this research make beyond academia? It names the change, says who benefited, and proves it with evidence a stranger can check. It is written for people who will never read your paper, and it is judged on the change, not on the cleverness of the science.
The structure is always the same, whatever the field. First, the problem before your work, with a number if you have one. Second, what your research contributed, in plain language. Third, the change itself: who did what differently, and when. Fourth, the evidence that proves it happened.
Compare the two versions. "Our research improved dairy farming in Punjab" is a claim: no change, no beneficiary, no proof. "Following our 2023 trial on 48 farms, the district livestock department adopted our mastitis screening protocol in 2024, and recorded cases fell by 31 percent within one season" is a case a panel can score.
Three terms sound alike and are constantly confused, so keep them apart before you write a word. The journal impact factor is a number attached to a journal, calculated from how often the journal's recent articles are cited on average. Our explainer on the journal impact factor covers how it is calculated.
A high-impact journal is a publishing choice: a journal whose articles are widely read and cited in your field. Choosing one is a strategy about visibility and audience.
An impact statement is the narrative document you write yourself, and the only one of the three that is about your work's effect on people, policy, or practice. When a form asks for impact, write the story of the change — never paste a journal's impact factor and stop there.
Your reader is almost never a specialist in your topic. On a funding panel, your statement may be read by a chemist judging a veterinary proposal, or an economist judging a linguistics project. Write for that reader: plain words, short sentences, no unexplained abbreviations.
What these readers score is narrower than most writers assume. In the United Kingdom's Research Excellence Framework, the national assessment of university research, impact counts for 25 percent of the overall score, judged on reach and significance. The 2021 exercise received 6,781 impact case studies from UK universities, each one a story of a change with evidence behind it.
Funders work the same way. The US National Science Foundation judges every proposal on two criteria, intellectual merit and broader impacts, and the UK's UKRI councils ask applicants to show potential impact in the application, after dropping the separate "Pathways to Impact" attachment in 2020.
Almost every real-world change falls into one of three kinds, and naming the kind first makes the rest of the statement easier to write. The first is policy impact: your research changed a rule, a guideline, or a decision. For example, a study showing a cheaper vaccine schedule protects calves just as well leads a livestock department to rewrite its guidance.
The second is social impact: your research changed what people know, how they behave, or how a community lives. For example, a handwashing study leads a district education office to install soap and water points in 60 schools, and pupil diarrhoea cases fall over the following year.
The third is industry and economic impact: your research changed how a business works, what it spends, or what it sells. For example, a feeding trial shows a local by-product can replace a third of imported concentrate without reducing milk yield; three feed mills adopt the ration, and feed costs fall by 12 percent.
The single rule that separates statements that score well from statements that do not is simple: every claim of change must carry proof that an independent reader can check. The proof does not need to be elaborate; it needs to be specific and verifiable.
Strong evidence comes in a few reliable forms. A policy document, guideline, or law that cites your work or names your team. A dated letter from the organisation that adopted your method, saying what it did and when. A before-and-after measurement — cases, costs, yields, or waiting times — with the dates and the sample behind it.
Weak evidence is easy to spot. "Our work was widely praised" proves nothing, because praise is not a change. "Our paper has 200 citations" proves other researchers read it, which is academic influence, not real-world impact. A letter from the end user — the department, the clinic, the company — carries far more weight than a testimonial from a co-author. Keep the document, the date, and the number with every claim.
The statement below is a sample, written to show the structure. The numbers are illustrative, not from a real study — replace every fact with your own.
"Mastitis costs small dairy farmers in Punjab an estimated 15 percent of milk yield, and most cases are detected only after the animal is visibly ill. Our 2022–2024 trial tested a low-cost screening protocol — a weekly strip test plus a simple record card — on 48 farms in Faisalabad district, and showed that early detection cut clinical cases by 31 percent within one season. In March 2024, the district livestock department adopted the protocol in its extension advice to all 214 registered dairy farms in the district, and trained 36 field officers to deliver it. The department's own records show recorded mastitis cases fell from 412 in the 2023 season to 284 in 2024 on participating farms. The protocol costs farmers less than 200 rupees per animal per month, and the department has now requested funding to extend it to two neighbouring districts."
Notice what each sentence does. The first states the problem with a number. The second states what the research contributed and found. The third names the adopter, the date, and the scale. The fourth gives the before-and-after measurement. The fifth states the cost and the next step.
Compare a weak version of the same work: "Our research on mastitis is highly significant and has been published in a leading journal. It will benefit farmers across Pakistan." No date, no adopter, no number, no evidence — it claims the whole country and proves nothing.
Step 1: Gather your evidence before you write a word. List every change you can prove: the adopter's name, the date, the document, the before-and-after numbers. Email the end user now if a letter is missing — a dated letter takes a week to obtain and is worth more than any paragraph you can write.
Step 2: Choose one change to lead with. If your research changed three things, pick the one with the strongest evidence and the clearest beneficiary, and build the statement around it. The others get one sentence each at the end.
Step 3: Draft in the four-part order. Problem, contribution, change, evidence — one or two sentences for each part, in that order. Use plain verbs — adopted, cut, trained, replaced, rewrote — and real numbers wherever you have them. Keep the science to one sentence.
Step 4: Cut every sentence that does not prove the change. Strike any sentence about your publications, your journal, your awards, or your methods. Strike adjectives such as "groundbreaking", "novel", and "significant" — they are claims, and the evidence should do the claiming.
Step 5: Test it on a non-specialist. Give the draft to a colleague from another field and ask two questions: what changed, and how do you know? If they cannot answer both in their own words, fix the sentence they stumbled on. Then check the word limit and match the funder's wording in your first line.
Strong statements use concrete verbs and named actors: "the ministry adopted", "the clinic trained", "farmers saved". They give numbers with dates: "cases fell from 412 to 284 between the 2023 and 2024 seasons". They name the beneficiary in the first sentence, and they keep sentences short — one change per sentence.
Weak statements hide behind vague language. "Our work has the potential to transform the field" says nothing about what actually changed, and the word "potential" tells the reader the change has not happened. "Stakeholders were engaged" does not say who, or what they did afterwards. When you catch one of these phrases, replace it with the fact it is hiding: who, what, when, and how much.
Write about the change in the past tense when it has happened, and in the future tense only when it has not — and label the future clearly. "The department adopted the protocol in March 2024" is a fact. "The protocol will be adopted nationally" is a hope, and a panel will treat it as one.
1. Confusing outputs with outcomes. A paper, a talk, a dataset, and a citation count are outputs — things you produced. An outcome is what changed because of them. For every output you mention, add what happened next.
2. Claiming impact without evidence. A statement that says "our findings influenced national policy" without naming the policy, the date, or the document is an assertion, not a case. If the influence is real, the document exists — find it and cite it.
3. Writing for specialists. Abbreviations, method names, and field jargon exclude the very people who score the statement. Spell out every term on first use. A panel member who has to guess what an abbreviation means will not guess in your favour.
4. Overclaiming the scale. Claiming national impact from a study on 48 farms, or global impact from a single-city pilot, damages credibility. State the reach you can prove — a district, a hospital network, three feed mills — and let the panel judge its significance.
5. Ignoring the word limit. Funders and panels enforce limits strictly, and an over-long statement may be cut off mid-sentence or returned unread. Write to the limit the form states, and if the form gives no limit, keep the statement under 400 words.
| Type of impact | What the change looks like | Evidence that proves it |
|---|---|---|
| Policy | A guideline, rule, or decision is rewritten | The dated document citing your work |
| Social and public | Behaviour or services change for a community | Before-and-after figures and service records |
| Industry and economic | A business adopts a method or cuts a cost | Adoption records, sales, and cost figures |
| Health | A clinic changes practice or outcomes | Clinical records and outcome measurements |
| Environment | Land, water, or animal management changes | Monitoring data and management records |
Keep the source document, its date, and the figure with every claim.
These authoritative external resources explain the standards and tools mentioned in this guide:
What is the difference between an impact statement and an impact factor?
An impact statement is a narrative you write about what your research changed in the real world, such as a policy, a practice, or a product. An impact factor is a number calculated for a journal, based on how often its articles are cited. One is your story with evidence; the other is a journal-level metric you do not control.
How long should a research impact statement be?
Follow the limit in the call or form, because funders and panels enforce it. A typical statement runs between 150 and 500 words, with many grant forms asking for about one page. If no limit is given, aim for 250 to 400 words: long enough for one clear change and its evidence, short enough that every sentence earns its place.
What counts as evidence of impact?
Evidence is anything an independent reader can check: a policy document that cites your work, a letter from an organisation that adopted your method, usage figures, sales or licensing records, media coverage, or a before-and-after measurement. A claim without a source is an opinion. Keep the document, the date, and the number with every claim you make.
Can I write about impact that has not happened yet?
Yes, but label it honestly as expected or potential impact, and show why it is likely. Name the partner who will use the result, the step that will happen next, and the date it is planned for. Funders accept a credible plan for future impact; they reject a future benefit described as though it already happened.
Should I mention the journal impact factor in my impact statement?
Only if the form asks for publication details, and even then keep it to one line. The impact factor describes the journal, not what your work changed, so it cannot carry an impact statement. Panels want to know who used your findings and what happened next, not where the paper was published.
How do I write an impact statement if my research is basic or theoretical?
Trace the chain from your result to the next user. A new theorem may feed a method, a method may feed a tool, and a tool may reach a clinic or a classroom. Name the next link, who holds it, and what changed because of it. Long-term impact still counts when the chain is concrete and each step is documented.
Submit your manuscript ← Back to GAP Insights
Useful publishers, databases, funders, tools and institutions for researchers: