Start by checking what a rating actually measures, because the four best-known systems measure four different things and only one of them produces a score you can compare across organizations. A Candid seal reports how much information a nonprofit disclosed. A BBB Wise Giving Alliance report says whether 20 standards were met. CharityWatch grades financial efficiency and says so explicitly. Charity Navigator produces a weighted score across four categories. Treating any of these as a verdict on whether the work is any good will mislead you.
Step one: identify which system you are looking at
Charity Navigator
Charity Navigator’s current methodology, updated in March 2026, scores organizations across four beacons: Impact and Measurement, Accountability and Finance, Leadership and Planning, and Culture and Compensation. Metrics are calculated on a zero to one scale, each assigned a weight, and the star rating comes from a weighted average of eligible metric scores.
The published bands are straightforward. A score of 90 or above earns four stars and the label “Great.” Seventy-five to 89 earns three stars, “Good.” Sixty to 74 earns two stars, “Needs improvement.” Fifty to 59 earns one star, “Poor.” Below 50 earns zero stars.
The system has changed twice recently, which matters if you are reading older commentary. Charity Navigator unified the Encompass system with the legacy star rating on November 2, 2022. It then issued the methodology revision described above, which it summarizes as moving scoring from participation to quality and which added metrics covering impact reporting, workplace policies, and board-approved budgets.
Candid, formerly GuideStar
Candid states the point plainly: a seal is not a rating. It indicates the type and amount of information a nonprofit has shared. Bronze requires basic identifying and contact information. Silver adds program descriptions and the population served. Gold requires financials, either an audited statement or a recent Form 990, plus leadership demographics. Platinum requires goals, strategies, and at least one self-selected quantitative metric, plus board demographics.
Two details are worth carrying. Candid confirms a nonprofit can earn Gold without sharing an audited financial statement. And the Platinum metric is chosen and reported by the organization itself, with no outcome standard attached.
BBB Wise Giving Alliance
BBB Wise Giving Alliance evaluates against 20 Standards for Charity Accountability grouped into governance and oversight, measuring effectiveness, finances, and solicitations and informational materials. Each standard receives one of three findings: met, not met, or unable to verify. Organizations meeting all 20 are called BBB Accredited Charities.
The organization states it does not rank or grade charities against one another. Two of its standards are numeric: spend at least 65 percent of total expenses on program activities, and spend no more than 35 percent of related contributions on fundraising.
CharityWatch
CharityWatch assigns letter grades from A+ to F based on two calculations averaged together: program percentage, and cost to raise $100. It considers a charity highly efficient at a program percentage of 75 or above and a cost to raise $100 of $25 or less.
Its distinguishing move is how it handles joint costs. Accounting rules let charities report portions of solicitation costs as program expenses. CharityWatch states that it adjusts those solicitation expenses out of reported program expenses and into fundraising before calculating ratios. It also compares fundraising expense to related contributions rather than total revenue, which produces a very different number for organizations with large non-donation income.
CharityWatch is explicit about its scope: the letter grade rests solely on financial efficiency measurements.
Step two: stop treating overhead as a quality measure
This is the correction the raters themselves have asked donors to make. In June 2013, the chief executives of GuideStar, Charity Navigator, and BBB Wise Giving Alliance published a joint open letter to American donors titled The Overhead Myth. Jacob Harold, Ken Berger, and Art Taylor signed it.
Their argument was that the share of expenses going to administration and fundraising is a poor measure of a charity’s performance. The letter asked donors to weigh transparency, governance, leadership, and results instead, and made the case that many charities should spend more on overhead, since those costs include training, planning, evaluation, and internal systems.
The letter did allow that overhead has some diagnostic value at the extremes, where it can help identify fraud or poor financial management. It argued against overhead as the sole or predominant measure rather than against looking at it at all. All three organizations still publish expense ratios today.
The practical version: a program ratio of 45 percent is worth a question. The difference between 78 percent and 84 percent tells you nearly nothing.
Step three: check the date on the underlying data
Ratings lag, sometimes badly, and the lag is disclosed if you look for it.
Charity Navigator’s own methodology guide notes that Form 990 filings may be delayed up to a year after the fiscal period ends, so the data may not reflect current financial position. Its program expense and fundraising efficiency metrics are three-year averages, meaning a rating published today can reflect fiscal years well in the past. Self-reported portal data stays valid for four years.
CharityWatch updates ratings generally every other fiscal year. BBB Wise Giving Alliance updates reports every two years. The IRS itself warns on its Tax Exempt Organization Search page that data updates are delayed and that it is still processing paper-filed 990 series returns received in 2021 and later.
A rating is a photograph, not a live feed. Check what year it is a photograph of.
Step four: understand who is missing
Coverage gaps are structural, not accidental. Charity Navigator requires at least three electronically filed Form 990s within the past six years, and states that the most recent filing cannot be a Form 990-PF, 990-N, or 990-EZ. Organizations that file only the short forms are listed but not rated. Private foundations are not evaluated.
The IRS lets organizations with gross receipts normally at or under $50,000 satisfy their annual filing requirement with Form 990-N, the e-Postcard, and permits Form 990-EZ below higher thresholds. Every one of those organizations is invisible to the rating systems that depend on the full return.
Scale tells the rest, using each organization’s own current published count. CharityWatch states it rates over 600 charities. BBB Wise Giving Alliance reports on more than 1,500 nationally soliciting charities, plus roughly 5,300 covered by regional BBB reports. Charity Navigator states it has rated more than 245,000. Candid says it holds profiles on roughly 1.9 million organizations, which spans the whole nonprofit universe rather than public charities alone, so the four figures are not directly comparable.
An unrated organization is usually a small or non-participating one, not a suspect one. Charity Navigator’s non-financial beacons draw on a portal a nonprofit must choose to complete, BBB accreditation requires enrolling, and Candid seals require claiming and updating a profile.
Step five: read the Form 990 yourself
The Form 990 is where the raters get their numbers, and it holds more than any of them summarize. Part IX separates program, management, and fundraising expenses. Part X shows assets and liabilities. Page 6 asks whether there was a significant diversion of assets and whether a records retention policy exists. Page 12 covers whether financial statements were audited and whether an audit oversight committee exists.
Know one limit. The IRS requires exempt organizations to make returns available for three years including schedules and attachments, but organizations other than private foundations need not disclose the names and addresses of contributors. Donor lists are not public.
Use Tax Exempt Organization Search to confirm an organization’s eligibility to receive deductible contributions and to pull its filings and determination letter directly.
What no rating tells you
None of these systems verifies that a program worked for the person it served. Charity Navigator’s own limitations section acknowledges that self-reported data may introduce reporting bias and that strong reported data collection practices do not guarantee effective execution. Its impact reporting tool is a pilot, and impact is not required until 2027.
Ratings answer whether an organization is well run and forthcoming. Whether its approach is the right one for the problem is a judgment they cannot make for you.
For donors weighing organizations working on economic hardship specifically, a plain-language comparison of groups working on poverty covers the questions worth asking beyond the score, though the filings and the search tool above remain where the verifiable answers live.