CLEAR Framework · Guide A
How to Pass the Five Hard Gates
A short, practical guide for anyone whose audit keeps stopping at the gate before it even gets to a score.
Why you are reading this
You've run one or more audits and hit a hard gate failure. That's normal, and it's fixable. But it's worth understanding why it happens before you resubmit, because a hard gate failure isn't a scoring problem. It's a structural one.
The CLEAR Framework checks five conditions before it scores anything on the 28-question, 56-point scorecard. If content fails any one of them, scoring stops immediately: no pillar score, no percentage, nothing. Credibility is the framework's non-negotiable foundation: high marks on Leverage, Empathy, Action, or Reach cannot compensate for a missing named source, an unsupported claim, overclaiming, misleading AI disclosure, or unreviewed high-risk advice. Fix the gate first. The scorecard only runs once all five pass.
The Five Hard Gates
No accountable source
What it checks: A named author, expert, or verifiable source is identifiable for any claims of expertise.
Why content usually trips it: The content reads as brand-voice copy with nobody attached to it: no byline, no named expert quoted, no “I built this / I ran this” first-hand framing.
How to fix it:
- -Add a real name to the content itself, or make sure the person publishing it is credited somewhere the audience can see, not just internally.
- -If it's expert commentary, name the expert and their relevant credential or track record, not just their job title.
- -Where possible, write in first person from someone who actually did the thing being described.
In the audit tool
In Step 2 of the audit, enter the publishing channel and byline. If a named author or publisher is supplied there, the body text doesn't have to repeat it to pass Gate 1. But if you leave that field blank and the body is also anonymous, this gate fails.
Unsupported significant claims
What it checks: Significant factual claims are backed by a named source, original data, or stated qualification.
Why content usually trips it: A stat, study, or result gets dropped in without attribution: “studies show,” “most companies find,” a number with no source behind it.
How to fix it:
- -Name the source for every material stat or study: publication, organisation, and ideally a date.
- -If the claim comes from your own data or experience, say so explicitly (“in our own testing,” “across the 40 clients we worked with in 2025”).
- -If you genuinely can't verify a claim, either cut it or qualify it honestly rather than stating it as fact.
Material overclaiming
What it checks: Outcomes, credentials, and capabilities asserted can credibly be demonstrated.
Why content usually trips it: Results or guarantees are stated more confidently than the evidence supports: “guaranteed results,” “the only framework that works,” credentials implied but not actually held.
How to fix it:
- -Match every claim of outcome or capability to what you can actually evidence or deliver.
- -Swap absolute language (“guaranteed,” “the best,” “always”) for language you can defend under scrutiny.
- -If a result varies by context, say so: a caveated true claim passes; an uncaveated inflated one doesn't.
Misleading AI disclosure
What it checks: AI involvement is disclosed (or clearly not required) given how the content is presented.
Why content usually trips it: AI was used to draft or substantially shape the content, but nothing in the piece signals that in a context where the audience would reasonably expect to know.
How to fix it:
- -If AI assisted the drafting, say so plainly, in whatever form fits the format: a line in the byline, a footnote, or a standard disclosure statement.
- -Match the disclosure to the actual level of AI involvement: a light editing pass is different from AI-drafted-and-lightly-reviewed.
- -If there was no AI involvement, this gate passes automatically: nothing to add.
In the audit tool
The “AI assisted drafting this content” toggle in Step 2 tells the auditor what to expect. Turn it on and the tool requires an explicit disclosure statement in the content to pass; turn it off and none is required. This must match reality: the toggle doesn't change what your content says, it changes what's expected of it. Toggling it off to skip the gate on genuinely AI-assisted content will pass this audit but misrepresents the content itself.
Unreviewed high-risk advice
What it checks: Any health, financial, legal, safety, or compliance advice has been reviewed by a qualified SME (or none is given).
Why content usually trips it: The content strays into advice with real consequences if wrong: money, health, legal standing, safety. Nobody qualified has checked it.
How to fix it:
- -Have a suitably qualified person review the content before it publishes, and be able to say who that was.
- -If the content is general commentary rather than individualised advice, make that framing explicit so it isn't read as advice it wasn't reviewed to give.
- -If there's no high-risk advice in the piece at all, this gate doesn't apply and passes automatically.
In the audit tool
The “Reviewed by a qualified SME” toggle in Step 2 feeds this gate directly. It only matters if the content actually contains health, financial, legal, safety, or compliance advice. For everything else, it's a non-issue.
What to do next
- 1Re-read the note under whichever gate or gates failed on your last audit: the tool tells you exactly which ones tripped.
- 2Fix the specific gap named above, not the whole piece: a hard gate fix is usually one sentence, one credit line, or one toggle, not a rewrite.
- 3Re-submit the same content. Once all five gates pass, the 28-question scorecard runs automatically and you'll get your Top Wins summary, free on every tier.
Hard gates exist because Credibility is the foundation the other four pillars sit on. Nothing else in your content, however useful, resonant, or well-distributed, can be evaluated fairly until the basics of who's accountable and what's actually true are settled. Clear that bar once and every future audit on similar content gets faster.
New to the tool? Start with Getting Started: Your First Audit. For how the weighting works once you're past the gates, see How scoring works.