Rules
Building nonprofit content plans around a sign-off list
A content sign-off list turns program evidence into donor-facing claims, and it fails when one story is stretched into a broad impact claim.
What to take away
- Every donor-facing claim needs a source recordthe program data behind it, the date, and the limit of what it proves.
- One story becomes a broad claim only when the program data covers the whole group, not the person in the photo.
- Sign-off fails most often on donor intent: a restricted gift described as general support.
- Track task completion, qualified support and correction quality, not impressions.
- When a claim fails review, correct the published version and note who owns the fix.
A content sign-off list is the record that says who approved a claim, on what evidence, and when it expires. Without it, the marketing lead becomes the only person who knows why a sentence exists.
Where the evidence comes from
Program staff hold the numbers. Marketing holds the audience. The sign-off list is where the two meet, and it should name the program report, its date, and the person who can confirm it.
Donor intent is the check most often skipped. A gift restricted to one site cannot be described as funding the whole program, however the story reads. Record the restriction beside the claim.
Mission content built from program evidence is slower to write and much harder to challenge. That trade is the point.
One story, one claim
A single success story proves that something happened once. Turning it into a broad impact claim needs the denominator: how many people entered the program, how many finished, and over what period.
If the program data does not cover the group, write the narrow version. "One participant completed the training" is honest. "Our training works" needs the cohort behind it.
The FTC advertising substantiation policy requires a reasonable basis for objective claims before they go out. Apply it to the exact promise and audience in the draft, not to the program in general.
What to record at sign-off
| Field | What goes in it | Who confirms |
|---|---|---|
| Claim | The sentence as published | Marketing lead |
| Source | Program report name and date | Program manager |
| Limit | What the data does not show | Program manager |
| Restriction | Donor designation on the funds | Development |
| Correction owner | Person who fixes the published page | Marketing lead |
A blank field blocks release. An assumed approval is not an approval, and the person who assumed it will not remember doing so in eight months.
Sign-off list fields
- Claimsentence as published
- Sourceprogram report name and date
- Limitwhat data does not show
- Restrictiondonor designation on funds
- Correction ownerperson who fixes page
Test the claim before you scale it
- Pick a defined group and a comparison basis.
- Track task completion, qualified support and correction quality.
- Walk one normal path and one failure path.
- Keep the inputs, the version and the decision.
- Set a stop rule and a review date.
The GAO data reliability guide ties data quality to intended use, so the same figure can be sound for one claim and unsound for another. Record the source date and its limits beside the decision.
Testing a claim before scaling
- Pick defined group and comparison basis
- Track completion, support and correction quality
- Walk one normal and one failure path
- Keep inputs, version and decision
- Set stop rule and review date
Early numbers are diagnostic. Expansion should wait for a result the current team, rights, systems and budget can reproduce.
Worked example: case 119
Case 119 is a claim that failed review after publication. A newsletter said a program "serves hundreds of families" on the strength of one site's intake sheet.
The audit recovered every approval. The program manager had signed off on the site figure. Nobody had checked whether the other two sites reported at all. The published number was one site's count presented as the program's.
The fix was narrow: correct the sentence, name the source as a single site, and add the missing sites to the next quarterly report. The correction owner was the marketing lead, and the review date was set for the following quarter.
When AI touches the workflow
The NIST AI RMF Playbook lists voluntary actions under govern, map, measure and manage. Use it when a tool drafts or summarizes content, and keep the program facts, the test and the decision with your own team.
The playbook is not certification and not a ranking. It is a set of prompts, and the sign-off list still carries the names.
Portability and vendor dependence
The GOV.UK open standards guidance links open standards to interoperability and reuse. It governs UK public-service work, so read it as a portability prompt rather than a rule for your charity.
The practical test: if your email platform or CRM changed tomorrow, could you export the content record and the email marketing checklist that goes with it? If not, the sign-off list lives in a vendor's format, not yours.
For the wider program this list sits inside, see nonprofit content marketing. The recurring questions teams ask about ownership, evidence and review are collected in common nonprofit content marketing questions, and the drafting side is covered in how to improve nonprofit content marketing.
Common questions
Who owns the sign-off list?
The marketing lead owns the document; the program manager owns the facts in it. Name both, because a list with one owner tends to drift toward whichever of the two is easier to reach.
How long does an approval last?
Set an expiry date on every claim. Program data ages, staff leave, and a two-year-old approval is a guess wearing a signature.
What happens when a claim fails review?
Correct the published version and record the correction owner and date. Leaving the old sentence live while the file says otherwise is the failure mode case 119 describes.
