How does our website undersell what we built?

It happens one review pass at a time. The company hedges every claim until nothing on the page is worth acting on. Legal softens it, clinical caution softens it again, and what survives is a sentence like "designed to help support improved workflow efficiency." The fix is precision rather than volume. A narrow claim you can defend outsells a broad claim you have to qualify.

If your team knows the product is the best in its category and the homepage makes it sound like a pilot project, this is what's happening.

Why do websites hedge so much?

Three forces push the same direction, and every one of them is reasonable on its own.

Regulatory caution comes first. Teams working near FDA pathways, clinical decision support, or protected health data learn early that an unqualified claim creates exposure. That instinct is correct in a submission and corrosive on a homepage, where it gets applied to sentences that carry no regulatory weight at all.

Clinical culture pushes the same way. People trained in medicine are taught to state findings with appropriate uncertainty, and that habit is a virtue in a journal. On a website it reads as a company that isn't sure its own product works.

Legal review finishes the job. Each pass removes a little more specificity, because specificity is what creates liability. Nobody in the chain makes a bad decision. The page still ends up saying nothing.

The compounding effect is what makes it invisible from inside. No single review meeting produces a weak page. Twelve of them across two years do, and by then the softened version is the house style that new hires copy.

What does underselling actually cost?

We scored 302 healthcare software homepages this month on the 19-signal Brand Signal Score and published the results as the Health Tech Messaging Index. Two signals show the hedging tax directly.

Authority and Credibility averaged 0.84 out of 2. Owned Language and Category averaged 1.04. These are companies with real evidence and real differentiation who score in the bottom half on sounding like they have either, because the language that would carry it got sanded off.

The band distribution is the blunt version. Of 302 companies, 127 landed in the Invisible band and only 7 reached Magnetic. Just over two percent of an entire vertical has a homepage that states a position clearly enough to be repeated.

That number is an opportunity rather than an indictment. In a vertical where almost nobody states a position, the first company that does is the one buyers remember and the one AI engines have something to quote. The bar is far lower than it looks from inside a review cycle.

How do we make strong claims we can still defend?

Narrow the claim until it's provable, then state it without qualifiers. Precision is what lets you drop the hedge, because a specific claim carries its own evidence.

"Designed to help reduce documentation burden" is vague and hedged, and it protects nobody in particular. "Clinicians at three health systems closed charts 22 minutes earlier per shift" is narrower, stronger, and easier to defend, because it describes what happened rather than what the product intends. Legal usually prefers the second one once someone shows it to them, since it is a statement of fact with a source.

Where a number doesn't exist yet, name the mechanism instead of the outcome. Say what the system does, on what data, with which human in which loop. That's concrete, it's defensible, and it answers the compliance question and the value question at the same time. It also keeps you clear of AI-Parmesan, the habit of reaching for "AI-powered" when the specific claim feels too risky to make.

Isn't a bolder website just overclaiming?

Overclaiming and hedging are the same failure wearing different clothes. Both substitute a general statement for a specific one. "Transforms care delivery" and "designed to help support care delivery" are equally unfalsifiable, and a buying committee discounts both on sight.

The way out of that pair is specificity, which is the one direction neither of them goes. A claim narrow enough to be checked is the only kind that reads as confident, and it's the only kind an AI engine can attribute and repeat when a buyer asks who does this well.

“Overclaiming and hedging are the same failure in different clothes. Both trade a specific sentence for a general one, and a buying committee discounts them both on sight.”

... Greg Rosner

How do we find the hedges on our own site?

Three passes, and you can do all of them in an hour.

• Search the site for the hedge vocabulary: designed to, may help, can support, aims to, intended to, helps enable. Every hit is a sentence that lost its verb somewhere in review.
• Take your five strongest claims and ask what evidence sits behind each one. Where evidence exists, rewrite the claim to describe what happened and cite it. Where it doesn't, replace the outcome claim with the mechanism.
• Run the Three Questions Test on the top of the page: why change, why change now, why change with you. Hedged pages usually fail the third question, because a qualified claim gives a buyer no reason to prefer you over the alternative they already have.

That third failure is worth sitting with. A hedge doesn't only weaken the claim. It hands the decision back to the status quo, which is the competitor most health tech companies never name and the one that wins most often.

What do we change first?

Start with the hero section, because it sets the register for everything under it and it is the shortest thing to fix. Take the single strongest claim the company can support, write it as a statement of what happened, and put it at the top without a qualifier in front of it.

Then change how review works rather than only what it reviews. Bring legal the evidence alongside the claim instead of the claim alone. Most softening happens because a reviewer sees an assertion with no visible support and does the only safe thing available to them. Given a source, the same reviewer will usually approve stronger language than the team expected.

The buying committee reads confidence as competence in a category where the purchase carries clinical risk. A page that commits to something specific is doing the reassuring work the hedges were meant to do, and doing it better.

Why does the hedging come back after every rewrite?

Because review is a process and the copy is an artifact, so the process wins every time it runs. Unless somebody has written down which claims the company stands behind and what evidence supports each one, every reviewer is deciding from scratch, and deciding from scratch under liability always trends toward softer.

That's the job a Magnetic Messaging Framework does here. It documents the claims, the proof behind them, and the language the company owns, so legal reviews against a standard instead of an instinct. The page stops getting quietly weakened, and the same document feeds the deck, the sales conversation, and the AI tools your team drafts with.