The Line Between Claim and Suggestion

Every investigative draft has sentences that feel true but rest on thinner ground than the writer realizes. The gap usually isn't dishonesty — it's momentum. A reporter who has spent weeks with a subject starts to write with the confidence of someone who knows more than the documents actually show, and a claim slips past that a fresh reader would flag immediately.

The fix is a habit rather than a talent: reading each factual sentence and asking whether a skeptical reader, with access to nothing but the sources cited, would accept it as written. If the honest answer is "probably, with a caveat," the sentence needs a caveat. If the answer is "only if they already agreed with me," the sentence needs to come out or be rebuilt from something firmer.

Sorting claims from suggestions

A claim is something the available evidence directly supports — a document says X, a filing shows Y. A suggestion is an inference drawn from that evidence, which may be reasonable but isn't the same thing. Conflating the two is the most common way a draft ends up overstating its case, often without the writer intending it.

A simple check before publication

Go through the draft and underline every sentence that asserts a fact about a named person or organization. For each one, write down — even briefly — the specific source that supports it. Sentences where you can't do this quickly are usually the ones that need softening, additional sourcing, or removal. Pair this with a fairness pass: for anything unflattering about a named party, check whether they've had a real chance to respond.

If you can't point to the specific line in a document that supports a sentence, the sentence is probably doing more work than the evidence can carry.

None of this slows a story down as much as it feels like it will in the moment. Most drafts only have a handful of genuinely weak sentences, and finding them early is far cheaper than finding them after publication.