What a disclosure pre-check cannot see

It is worth being precise about what this class of tool does, because the gap is where expectations break — and for creators, a broken expectation is measured in reach lost and deals paused after the post has already been seen by thousands of people.

What it catches

Patterns and placement that published guidance can be quoted against: disclosure-presence — whether any unambiguous connection-disclosing phrase or hashtag ("paid partnership", "#ad", "I earn a commission") appears at all, or only vague relationship labels the guides treat as insufficient, extracted from the actual text; clarity — a bare mid-sentence "ad" or decoder-jargon like "sp"/"collab" flagged as too easy to miss; placement — the first clear disclosure token's position measured against a heuristic for the platform's fold (Instagram captions fold at roughly the first ~125 characters; TikTok and Shorts fold early too), and a "#ad" dropped deep into a hashtag block; and material-connection specificity — an affiliate shortlink or "link in bio" CTA with no commission sentence, and copy that understates the connection you declared to the checker. These are checkable against the pasted copy, so they are checked against the pasted copy — deterministically, with quoted evidence for every finding, and the same result every run. The reference lists are derived from the factors documented in the Endorsement Guides (16 CFR Part 255) and its sections on disclosure of material connections and when and how a disclosure must be clear and conspicuous, snapshot-dated in the engine version rather than claimed to be live-synced.

What it cannot catch

• A clean label and a hidden deal. A post can carry a perfect "Paid partnership" line and still mislead if the actual arrangement was worse than disclosed — say a flat fee plus commission while the caption only mentions the product. No wording check verifies the contract behind the words; §255.1 cares about the real connection, which lives outside the paste.
• The relationship itself. Whether you were paid, gifted, employed, or just a fan is a records question. None of it lives in the pasted text, and none of it is looked up: EndorsePrecheck never connects to your accounts, brand deals, or any affiliate dashboard. An honest pre-check lists verification as an advisory and explicitly refuses to claim it checked — pretending to inspect ledgers you cannot see is how tools become noise.
• The exact fold on a device. "Below the fold" is a heuristic, not a measurement. Truncation length varies by app version, screen size, and whether the reader follows you; the tool flags a disclosure sitting far from the front of a long caption, not a guaranteed-invisible one. It never inspects any app and says so.
• The platform label you did set. Instagram, TikTok and YouTube each have paid-partnership and promotional-content toggles that are separate from caption text. A paste cannot see whether you switched one on. The check reminds you the toggle exists and that the guides treat it as helpful, not as a substitute for in-text wording — it never claims to verify its state.
• The reviewer's private mind. Enforcement — by an agency, a marketplace, or a platform's ad review — runs on discretionary models nobody publishes. The guides describe the neighborhood; the exact weighting of intent, materiality and remedy is not a formula. The same buried "#ad" draws a warning for a first-timer and a strike for a repeat account. Nobody — least of all a word-list engine — gets to promise what the final reviewer will do.

Why the boundary matters

A tool that claimed compliance would be claiming something nobody outside the regulator can deliver, and creators would find that out at the worst moment — mid-campaign, with the post live. Stating the limit up front is what lets the check get trusted for the things it genuinely knows: that the sentence matches a published factor, that the disclosure sits late in the caption, that the fix is a moved line or a plainer label, not a dice-roll. The honest claim is narrower and more useful than the fake one: a scored pre-check moves the odds, per post, before publishing. The human stays exactly where a disclosure pre-check should leave them — holding the context the machine cannot see: the contract, the audience, the label, the deal.

How to use the boundary productively

A pre-check is the first layer of a disclosure-governance stack, not the whole stack. Layered deliberately, each layer covers what the one before it cannot: the mechanical scan catches published-factor patterns and placement problems at draft speed; the verification pass — run by you, from your own deals, because this tool deliberately never does it — confirms every stated connection matches the real one; the toggle pass confirms the platform's paid-partnership or promotional label is set where the platform offers it; the legal pass owns jurisdiction interpretation, including the EU, UK and Australian regimes this build flags only as context advisories — their primary texts were unreachable from the environment that produced the rule set, and the exclusion is documented in the sources log. Buying these layers separately is how most serious teams already operate — the mistake is treating any single layer as covering the others. The scan's job is to make sure you never bring a known, mechanical, avoidable problem into the expensive layers.

The correct mental model

Treat a pre-check the way pilots treat the before-takeoff checklist. It does not make the flight safe; it catches, mechanically and repeatably, the handful of items whose omission turns a routine departure into an incident. The judgment — is this connection real, is this label conspicuous enough for this audience, is this the right deal to post today — remains entirely human. A disclosure pre-check is an endorsement-factor scan with memory: every draft gets the same published patterns and the same placement heuristics applied, every time, at paste speed, and every finding cites the sentence it came from. It is decision support, not a decision. Anyone selling the decision is selling something that does not exist. The creators who get the most from a bounded tool keep the boundary visible: scan early, escalate what the scan flags, and hold the judgment the scan deliberately declines to make.

Heuristic pre-check only — not an endorsement-disclosure certification or legal advice. Challenge and enforcement outcomes depend on regulators, marketplaces, and platforms; whether a material connection is real cannot be measured from the pasted text; regulated-disclosure questions belong to a lawyer.