GreenPrecheck is a paste-box for environmental marketing copy. You drop in a claim like "eco-friendly", "100% natural", or "carbon neutral", and it returns a scored heuristic pre-check: is the wording vague, absolute, carbon-related, or recyclable-shaped in a way that tends to draw scrutiny. I named it after the thing it is most useful for — making you delete a word before a regulator or a customer does.

The problem is unqualified absolutes

The FTC Green Guides are mostly about qualification. "Recyclable" is fine if the thing is actually recyclable in most communities; "green" with no backing is the kind of claim the Guides warn against. The detection logic is a tiered word list. Tier one: absolute or near-absolute terms ("100%", "zero", "all natural") that need substantiation. Tier two: vague benefit words ("eco", "earth-friendly", "sustainable") that mean little without context. Tier three: carbon and recyclable phrasing that triggers a specific substantiation question.

I run the pasted text through all three tiers and report which fired, with the reason. A model, where present, only rewrites the findings into a checklist. The matching itself is rules, because the question "did this word appear" is exactly the kind of enumeration a model should not be trusted with.

What the report gives you

Three things come back. Vague-claim flags: the soft words that read as environmental benefit without saying anything checkable. Carbon and recyclable cues: phrasing that implies a footprint claim and therefore needs numbers behind it. A fix checklist: concrete rewrites that keep the marketing honest — replace "eco-friendly" with the specific property ("made from 30% recycled resin"), drop the absolute, or add the qualifier.

The score is a prompt, not a pass. A low score means "nothing obvious fired"; it does not mean the claim is substantiated.

A note on why vagueness gets flagged at all: a claim with no checkable meaning is the one regulators pursue most, because it tilts the buying decision without costing the seller anything. "Sustainable" with no backing is worse than a specific, modest claim that a customer could verify. The tool is built to push copy toward the specific version, where the marketing still works and the substantiation question is at least answerable. The GreenPrecheck page is explicit about this, because the tool's value is in catching the lazy wording, not in blessing the careful wording.

Where it stops

It cannot tell whether your underlying claim is true. If you write "made from 30% recycled resin", the tool will not verify the resin. It does not know your supply chain, your certifications, or whether "carbon neutral" is backed by removal or just offsets. It reads text, not invoices.

It also follows the US FTC Green Guides and the style of the EU EmpCo direction, but it is not a substitute for either regime's formal review. And it does not model every sector-specific rule — a "compostable" claim for packaging has its own tests the tool does not run.

So it is decision-support. The report is the thing you hand your copywriter so the conversation is "here are the three words to fix" instead of "is this legal, maybe?". If a claim is going on a product that ships at scale, the flagged items deserve a real substantiation check. The GreenPrecheck scan just makes sure you are not publishing the obviously weak version first.

FAQ

What kinds of claims does it flag?
Vague benefit words ("eco", "sustainable"), absolute terms ("100%", "zero"), and carbon or recyclable phrasing that implies a footprint claim needing backing.

Does it verify my environmental claims are true?
No. It reads the wording and flags risky patterns. It cannot see your supply chain or certifications.

Is this a certification?
No. It is a heuristic pre-check for decision-support, not a green certification or legal sign-off.

Does it cover EU rules?
It follows the FTC Green Guides and the style of EU EmpCo language, but it does not perform a formal regulatory review under either framework.