'What Difference Will Your Project Make?' The Grant Application Question That Sinks Most Bids

What funders are really asking

If a funding bid is on your plate this month, you don’t need a textbook on grant application questions — you need the four things an assessor is scoring when they read your answer to “what difference will your project make?”. I write these answers for my own charity under the same deadline pressure, so what follows is the working version, not theory. When a funder asks the question, they are asking you to demonstrate a causal chain from an evidenced need, through a specific activity, to a measurable change in the lives of the people or communities you serve. They are not asking what you will do; they are asking what will be different afterwards, for whom, and how you will know. Assessors read hundreds of these answers a year, and most restate the activity instead of the outcome.

This single question is disproportionately weighted in scoring because it tells a funder whether you understand your own theory of change, or whether you have simply designed a nice-sounding project. Grant panels are trained to separate outputs (what you did) from outcomes (what changed as a result), and applications that blur the two consistently score lower, even when the underlying project is sound. If you are building out your wider application, see our how to write a grant application pillar guide for the full process this question sits inside, and our grant funding handbook for everything around it, from finding funders to reporting.

The 4-part answer structure: need, change, evidence, measure

The strongest answers to this question follow a four-part structure: state the need with a figure, describe the specific change you expect, cite the evidence that this type of intervention works, and name how you will measure it. Missing any one of the four parts is the single most common reason this section loses marks, even in otherwise strong bids.

• Need: a specific, evidenced problem, not a general statement. “Isolation among older people” is weak; “38% of people over 75 in [area] report feeling lonely often or always” is strong, provided the figure is real and sourced.
• Change: what will be different for the people you work with, stated as a shift, not an activity. “Fewer lonely older people” is weak; “60 older people will report reduced loneliness on the UCLA Loneliness Scale” is strong.
• Evidence: why a funder should believe your activity produces that change. This can be your own past delivery data, a published evaluation, or sector research such as the gov.uk statistics library or a relevant What Works Centre review.
• Measure: the specific tool or method you will use to prove the change happened, whether that’s a validated scale, a pre/post survey, or administrative data — our outcome indicator picker suggests tested measures by outcome area if you’re starting blank.

This is exactly the structure a theory of change template is designed to force you into before you write a word of the application, which is why the two documents should be written together, not the theory of change retrofitted afterwards.

From experience: When I write the “what difference will your project make?” answer for my own charity’s applications, I draft the measure first — the tool or record we would actually use to prove the change — and only then write the promise. Working backwards stops you promising a change you can’t evidence, and it usually shrinks the claim into something an assessor will believe. A modest outcome you can prove beats an impressive one you can’t, almost every time.

Weak vs strong: three annotated examples

Below are three real-pattern examples (anonymised and reworded) showing the same project described as an output-led weak answer and an outcome-led strong answer, with the fix annotated. In each case, the strong version is not longer or more elaborate; it simply answers the actual question asked. (More annotated pairs live in our example grant application answers.)

Example 1: a youth mentoring project

Weak: “Our project will make a real difference to young people’s lives by delivering 15 mentoring sessions to 30 young people over 12 months, giving them the support they need to thrive.”

Strong: “70% of the 30 young people on our mentoring programme will show a measurable improvement in school attendance (tracked via school records) within two terms, based on outcomes from our 2024/25 cohort where 22 of 28 mentees improved attendance by an average of 11%.”

Annotation: the weak version describes delivery volume (an output). The strong version states a percentage, a timeframe, a measurement method and cites prior evidence, all four parts of the structure.

Example 2: a food security project

Weak: “We will provide emergency food parcels to families in crisis, ensuring no one goes hungry in our community.”

Strong: “150 families referred through our crisis pathway will avoid a food-related A&E or GP presentation during a food crisis, measured through referral partner follow-up at 6 weeks, addressing a need where [your local food insecurity figure, with source] of households in [area] report skipping meals.”

Annotation: “ensuring no one goes hungry” is an unmeasurable promise no charity can make. The strong version narrows the claim to something specific, plausible and checkable.

Example 3: a mental health drop-in

Weak: “Our drop-in sessions will improve wellbeing and reduce isolation for people experiencing mental health difficulties.”

Strong: “80% of regular attendees will report improved wellbeing on the Warwick-Edinburgh Mental Wellbeing Scale (WEMWBS) after 12 weeks of attendance, consistent with the 76% improvement rate recorded in our 2024 cohort.”

Annotation: naming the actual validated tool (WEMWBS) signals competence to an assessor in a way that “improve wellbeing” never can.

Linking to your theory of change

Your answer to “what difference will your project make?” should be a condensed, plain-English version of your theory of change, and the two must use identical numbers, timeframes and language, because assessors do cross-check. If your application says 60 people and your theory of change says 80, that inconsistency is one of the fastest ways to lose an assessor’s confidence in the rest of the bid.

Build the theory of change first using our free theory of change template, then lift the outcome statement directly into the application question, trimmed to fit the word count. This also speeds up every future application, because you are reusing a validated outcome statement rather than writing a fresh one under deadline pressure each time — that reuse is half the reason I bother keeping my own charity’s version current.

Phrases that undermine you

Certain stock phrases signal to an experienced assessor that an applicant is filling space rather than answering the question, and they should be cut on sight. Phrases like “make a real difference,” “empower individuals,” and “ensure the best possible outcomes” carry no information: they could describe almost any project, which is precisely why panels are trained to discount them. My own first drafts still contain a few of these — the discipline is in the edit, not the draft. See our full 30 phrases that weaken your grant application post for the complete list and stronger alternatives.

The fix is nearly always the same: replace the vague phrase with a number, a name, or a named tool. “Empower young people” becomes “30 young people will complete an accredited employability course”; “ensure the best possible outcomes” simply gets deleted, because the sentence around it already says what the outcome is once you’ve applied the 4-part structure above.

What to do next

Before your next application deadline, pull up your last three submitted bids and check every “difference” or “impact” answer against the need-change-evidence-measure structure above. Rewrite any that describe activities rather than outcomes, cross-check the numbers against your theory of change, and delete any of the weak stock phrases flagged in the 30 phrases post — our grant application pre-submit checklist folds all three checks into one pass.

If you want this checked automatically before you submit, CharityIQ’s AI grant writer is trained to flag output-dressed-as-outcome language and missing evidence in real time as you draft.

Want your outcome statements checked before you submit? Join the AI grant writer waitlist →