Scores are never for sale — see how we're funded and the commercial independence clause below.
Four beacons
Assessed
Accountability & Transparency
40 points — filing history (15), trustee oversight (15, scaling up to a board of three or more), declared policies (10).
Assessed
Financial Health
30 points — reserves (15, full marks 3–24 months, tapering on both sides), income stability (10), liabilities to assets (5, scaling down as debt rises).
Assessed
Financial Efficiency
20 points — program expense ratio and fundraising cost, each on a graduated scale (not a single cutoff). Kept at 20% of the total: financial-ratio scoring alone is not a reliable effectiveness signal (see "why not just an overhead ratio?" below).
Assessed
Community Support
10 points — volunteer-to-staff ratio from Charity Commission workforce data, scaling continuously up to a 2:1 ratio.
Score to stars
| Score | Stars | Label |
|---|---|---|
| 90+ | ★★★★★ | Exceptional |
| 75+ | ★★★★☆ | Good |
| 60+ | ★★★☆☆ | Needs improvement |
| 1+ | ★★☆☆☆ | Poor |
| 0+ | ☆☆☆☆☆ | Not rated |
What we measure today
- Accountability & Transparency (40 pts): five-year filing history, trustee oversight (scaling up to a board of three or more, the minimum recommended by the Charity Governance Code), safeguarding / conflict / volunteer policies.
- Financial Health (30 pts): reserves (full marks 3–24 months, tapering on both sides), three-year income trend, liabilities-to-assets.
- Financial Efficiency (20 pts): program expense ratio and fundraising cost per £1 raised, each on a graduated scale rather than a single pass/fail cutoff.
- Community Support (10 pts): volunteer-to-staff ratio from regulator workforce data where disclosed (England & Wales annual-return fields; OSCR and CCNI records often omit this, so the beacon may be thinner).
Every filing figure traces back to the charity's own accounts, prepared under the Charity SORP accounting standard and submitted to the relevant UK regulator — the Charity Commission for England and Wales, OSCR in Scotland, or CCNI in Northern Ireland. Governance metrics reference the voluntary Charity Governance Code, which the Commission endorses but does not mandate. We don't score outcomes or real-world impact — public filings don't capture that, so we don't claim to measure it.
Policies & red flags
Statutory inquiry (kill switch)
If the relevant UK regulator has an open statutory inquiry, the charity scores 0/100 automatically with a regulatory warning banner — no pillar math is applied.
Red flags
Factual cards — statutory inquiry, fewer than three trustees, late filings, high fundraising spend, sharp income drops, very high reserves.
Give with Confidence
Score 75+ (about 4★ on the 5-star display), no serious red flags, filings up to date, no statutory inquiry.
What we exclude from a full score
A full Clarity Score needs detailed annual-return fields. Charities under £100,000 income, and many OSCR/CCNI records with thinner published returns, get a basic register profile instead of a score. Optional donor support never influences ratings. Basic profiles cover charities in our latest extracts from the three UK regulators; recent registrations may not appear until the next refresh.
Financial peer comparisons
These comparisons provide context alongside the Clarity Score and do not change it. We group active charities in our detailed dataset by regulator, primary cause category, latest annual income band and year in which the accounts end. Records with registration notices and the fictional demo are excluded. The cause category is a broad directory classification, not proof of an identical delivery model.
- We require known start and end dates covering 350–378 days (50–54 weeks). Short first accounts, extended periods and unknown period lengths are excluded. We do not annualise them or mix them with full years.
- The featured charity is excluded from its peers. A panel appears only when at least ten other charities qualify. Each measure needs ten disclosed peer values before we publish its median.
- Income and expenditure come from the same latest period. Charitable activities share divides the disclosed charitable expenditure by total expenditure for that period, using our existing disclosure checks. Missing or invalid values are excluded; reported zeros remain zeros.
- The median is the middle sorted value, or the mean of the middle two values. Coverage beside each median shows how many qualifying peers supply that measure. It is a dataset comparison, not a census of every UK charity.
Year-end months, service costs, restricted grants, accounting policies and grant-making versus direct delivery can still differ. A higher spending share or larger income is not evidence of greater effectiveness. We do not benchmark reserves here: readily comparable unrestricted reserve figures are not available across the cohort. See how to choose between similar charities.
Commercial independence
CharityCompare sells memberships and advertising. These relationships are administered by a separate commercial process from the scoring pipeline, which runs entirely on UK regulator filings. Scores are regenerated from filings on a fixed schedule; no staff member can adjust an individual score. Paid placements are labelled “Featured partner” and are excluded from all ranked lists, leaderboards and comparison tools. If a charity cancels a paid plan, its score does not change — because it was never connected.
A worked Clarity Score example
This is a fictional calculation, generated by the same formula as our real ratings. It is not a charity, review or donation recommendation.
Example inputs
Three years each with £1 million income, £800,000 expenditure, £640,000 charitable-activities spending and £80,000 fundraising costs. Three trustees; all three assessed policies disclosed; no late filings or statutory inquiry. Reserves £400,000, assets £600,000, liabilities £60,000, 10 staff and 20 volunteers.
Programme-spending share: £640,000 ÷ £800,000 = 80%. Fundraising cost relative to all income: £80,000 ÷ £1 million = 8p per £1. This is not a cost per £1 of donations because the denominator includes other income.
Calculated score: 92/100
Accountability & Transparency: 40/40 points
- Filing history (5 years): 15/15
15/15 pts
No recorded lateness in available history = 15 pts · one late filing = 5 pts · two or more or unknown status = 0 pts. Missing data is not proof of late filing.
- Trustee oversight: 15/15
3 trustees · 15/15 pts
Three or more trustees on the register = 15 pts, scaled down for fewer (governance red flag below three).
- Declared policies: 10/10
10/10 pts
Safeguarding (4) · conflict of interest (3) · volunteer management (3) — from Charity Commission annual return.
Financial Health: 24/30 points
- Reserves (months of expenditure): 15/15
6 months · 15/15 pts
3–24 months = full marks, tapering on both sides · 0 months or 4+ years = 0 pts. Reserves are not necessarily cash; needs vary by operating model.
- Income stability / growth: 5/10
5/10 pts · 3-year average
+20% or more over the window = 10 pts · flat = 5 pts · −20% or worse = 0 pts, scaled between.
- Liabilities to assets: 4/5
10% · 4/5 pts
0% liabilities-to-assets = 5 pts, scaling down to 0 pts at 60%+ debt.
Financial Efficiency: 18/20 points
- Program expense ratio: 8/10
80% · 8/10 pts (3-year average)
90%+ on charitable activities = full marks, scaling down to 0 pts at 50% or below.
- Fundraising efficiency: 10/10
8p per £1 income · 10/10 pts (3-year average, disclosed years)
≤10p to raise £1 = full marks, scaling down to 0 pts at 40p or above (fundraising cost ÷ income).
Community Support: 10/10 points
- Volunteer-to-staff ratio: 10/10
20 volunteers / 10 staff · 10/10 pts
Volunteers at 2x staff or more = full marks · no volunteers = 0 pts, scaled between.
What happens when fields are missing?
Keeping the same financial years but omitting governance, workforce, assets and reserves inputs gives 23/100 under V1.1. Several undisclosed inputs earn zero points. V1.1 does not reweight the total to compensate. Unknown filing status earns no filing points. These are limitations of the formula, not findings that a charity lacks policies, is poorly governed or has no volunteers. Compare disclosure coverage as well as scores.
Our formula uses published proxies and chosen thresholds. It cannot establish programme effectiveness, fraud risk, service quality or the best use of your donation. A statutory inquiry triggers V1.1’s zero-score rule; an inquiry is not itself a finding of wrongdoing. Historical or exceptional register records are excluded from current promotional rankings.
V1.1 excludes missing fundraising-cost fields from the scoring average. An explicit reported zero is retained; no disclosed costs means no fundraising points. A fourfold income discontinuity, a gap in annual periods or an unusually short/long period receives no trend points pending comparability review. These rules make missingness and reporting boundaries visible, but do not turn financial ratios into impact evaluations.
See disclosure coverage in the refreshed dataset. Independent external assessment of the scoring design remains pending.
Methodology changelog · Scoring guide
Contains public sector information licensed under the Open Government Licence v3.0. Sources: Charity Commission for England and Wales, Office of the Scottish Charity Regulator and Charity Commission for Northern Ireland. Open Government Licence v3.0 .