A claim examined alone cannot be assessed
Every claim a brand makes is a claim against alternatives. Premium against what. Fast compared to whom. Trusted relative to which other option a buyer was considering at the same moment. An identity examined in isolation can pass every other test. Internally consistent, honestly claimed, well produced, and still interchangeable with three competitors.
Table stakes and differentiators are not the same thing
The most common strategic error we find, and it is expensive because it consumes the most valuable space a business has. Some claims must be made because their absence is disqualifying. Licensed. Insured. Established. Responsive. Nobody wins on these, though everybody loses without them. Others actually distinguish. A brand leading with table stakes is spending its homepage saying what every competitor says, then wondering why nothing lands. The test is simple and rarely applied: take each claim and ask whether a named competitor could say the same thing truthfully. If four of them could, it is a category requirement wearing the clothes of a position.
What competitors structurally cannot say
The most defensible claims available to any business, because a competitor cannot adopt them by deciding to.
Size, in either direction
A ten-person firm can promise things a national chain cannot deliver, while the reverse is equally true.
Ownership and model
Independent against owned. Direct against reseller. Manufacturer against distributor.
Location
Where it genuinely matters to a buyer rather than being asserted for its own sake.
History
Thirty years cannot be acquired.
Specialization
The deliberate refusal to serve everybody. Hardest of all for a rival to adopt, since doing so means giving up revenue they currently take.
These are the claims worth building on, though most businesses hold at least one without ever using it.
Vocabulary, not only positioning
Where every business in a category uses the same words, the words have stopped carrying meaning and nobody notices because everybody is inside the convention. We compare the language across the competitive set. Which terms run universal, which are contested, and which sit unclaimed. Unclaimed vocabulary is territory, and taking it is covered in niche targeted content under organic search marketing. Then the same comparison against what customers actually say. A business describing itself in terms its buyers never use has a translation problem no amount of design will fix.
How the set gets chosen
The competitor set is agreed with you before anything is examined, and it is named in the report with the date. Your own view of who you compete with is one input. The machines name a second set when asked for alternatives, and the two frequently disagree, which is a finding before any comparison begins. Customers naming who else they considered supply a third. A relative finding is only as good as the set it was measured against, and that set moves.
What the comparison produces
Each claim marked as table stakes, differentiating, or contested. The claims only you can make. The vocabulary nobody has taken. And where a competitor is genuinely stronger, said plainly, since an audit finding no competitor doing anything better has not looked.
Where competitive work connects
The claims themselves are verified in claim verification. How the machines rank you against the same set is covered in AI brand perception. Taking unclaimed vocabulary is covered under organic search marketing.
How the engagement runs
Screen shared sessions against real competitor material rather than assumptions about it.
"If your positioning statement would be true of your three closest competitors, it is not a position and it is costing you the best space on your site."