How to build a grant assessment framework your trustees will actually use

Six trustees around a table. Three of them have read the assessment papers. Two skimmed them on the train in. One has not looked at them at all, but she knows the CEO of the charity under discussion personally, worked alongside her at another organisation a decade ago, always thought highly of her. The conversation that follows is not a bad conversation. But it is not the same conversation that would happen if any combination of those six people sat down in a different order, on a different day, with a different mix of prior knowledge in the room.

A grant assessment framework does not replace that conversation. It does not eliminate the value of experience, relationships, or instinct. What it does is give the conversation somewhere to land. It makes sure that when trustees disagree about a charity, they are disagreeing about the same thing, with the same information in front of them.

What a framework actually is

There is a tendency to think of a grant assessment framework as a scoring spreadsheet. It can be that. But the spreadsheet is an output of the framework, not the framework itself.

At its core, a framework is an agreed answer to three questions. What does this foundation care about? How much does each thing it cares about matter relative to the others? And what would a charity that meets those standards actually look like in practice?

The third question is the one most frameworks skip, and it is the one that creates the most inconsistency. Trustees can agree that financial sustainability matters and then score the same charity completely differently because they hold different pictures of what financial sustainability looks like for a small grassroots organisation compared to an established national charity.

Start with strategy, not criteria

The most common mistake in framework design is starting with criteria. Trustees sit down and generate a list of things charities should have, financial health, good governance, clear theory of change, and then try to turn those into a scoring tool. The result is usually a generic list that could describe almost any grant-making programme anywhere.

Before designing assessment criteria, a foundation needs to be able to answer these questions honestly. What change do we want to see in the world, and where? What stage of organisation do we want to fund, early, established, or scaling? How much risk are we willing to carry? Are we trying to fund proven approaches, or are we deliberately seeking out unproven ones with higher potential?

The answers to those questions should drive the criteria. A foundation committed to funding early-stage grassroots organisations needs different assessment criteria than one focused on scaling established charities to reach more people. Reserves requirements, governance expectations, and evidence standards are all different in those two contexts. A framework that does not reflect those differences is not really a framework. It is a box-ticking exercise.

Choosing criteria

Most grant assessment frameworks for UK trusts and foundations work across five broad areas: financial health, governance quality, impact evidence, leadership and culture, and strategic fit with the foundation's priorities.

The content of each area, and how heavily it is weighted, should reflect what the foundation actually cares about. A foundation that funds organisations working with particularly vulnerable beneficiaries might weight safeguarding governance much more heavily than one focused on environmental causes. A foundation that explicitly funds organisations in early-stage development might apply lower reserves expectations than one that funds established delivery organisations.

Make the weightings explicit before you apply them to any charity. If you discover mid-assessment that you want to weight one criterion more heavily because a particular charity scores well on it, you have found a bias worth noticing.

Picking a scoring approach

Three approaches come up most often, and each suits a different kind of foundation.

Binary scoring (yes or no against each criterion) is the simplest approach and the hardest to argue with. Either the safeguarding policy exists and is current, or it does not. Either the reserves cover three months of operating costs, or they do not. The limitation is that it cannot capture degrees of strength. A charity with excellent impact evidence and a charity with thin but honest impact reporting both score the same yes on "impact evidence present."

Rated scoring (usually 1 to 5 on each criterion) allows for nuance and produces an overall score that can be compared across charities. The risk is false precision. A score of 3.4 versus 3.6 is not a meaningful difference. Rated scoring works best when the rating scale is explicitly defined for each criterion, not left to individual interpretation.

Tiered assessment (Tier 1, Tier 2, or does not meet threshold) sits between the two. It is more informative than binary and less granular than rated. It lends itself naturally to a shortlisting process: Tier 1 charities go to deeper assessment, Tier 2 charities go on a watch list or a future round, the rest are declined with a clear reason.

The right choice depends less on which approach is theoretically better and more on which one a small volunteer board with limited time will actually use consistently. A framework that takes four hours to apply per charity will not be applied consistently. A framework that takes forty-five minutes and produces a clear recommendation will.

Making it usable

The graveyard of grant-making is full of frameworks that were beautifully designed and never used. They usually share one characteristic: they were designed by someone who cared deeply about getting it right, rather than by the people who would have to use it in a room with six other opinions and a two-hour window.

A useful framework has no more criteria than trustees can hold in their heads during a meeting. Somewhere between five and eight is right for most foundations. It has clear definitions that mean the same thing to different people. It produces an output that is a useful input to discussion, not a conclusion that overrides discussion.

It also has a way of recording the conversation, not just the score. A trustee meeting that ends with a score of 3.4 and no written rationale has not produced a decision record. A meeting that ends with a brief note explaining why the governance concerns outweighed the programme strength has.

Recording the decision

Grant decision records serve two purposes. They create accountability in the present: trustees who know they will have to write down their reasoning tend to reason more carefully. And they create institutional memory for the future: a foundation that can look back at five years of decision records and see what it funded, why, and what happened, learns things that no framework can teach.

A decision record does not need to be long. It needs to say what criteria were applied, what evidence was reviewed, what questions arose and how they were answered, and what the outcome was and why. "We funded them because they scored highest" is not a decision record. "We funded them because the reserves risk was manageable given the strong statutory income pipeline, and the impact evidence, while self-reported, was clearly structured and honest about limitations" is.

cleargiving.io lets foundations configure their own scoring framework, criteria, weights, and tiers, so every assessment reflects the foundation's actual priorities rather than a generic template. Each assessment produces a full decision record that trustees can review before a meeting. You can find out more about how it works or read about the methodology behind the scoring. If you are building a framework from scratch, the charitable trusts page has more on what good practice looks like for smaller foundations. To see it in action, request access here.