How to Write a Comparison Page That Names Its Own Anti-Fit
Put a named anti-fit in every comparison page: a specific set of buyers who should choose the competitor, stated in their terms and placed before your own case rather than after it. Two reasons, and only one of them is about assistants. The durable reason is that a page which never concedes anything carries no information — every claim on it is equally unfalsifiable, so a reader discounts all of them. The narrower AEO reason is that an assistant composing a balanced answer needs a source for the other side, and a page that supplies both sides can be cited for either. What we cannot tell you is how much the second effect is worth; treat the first as sufficient justification on its own.
We publish comparison pages and we have a commercial interest in you believing they work. The argument below is deliberately narrow, and the part we can actually demonstrate is smaller than the part we cannot.
What an anti-fit section is
Not a hedge, and not a list of features you lack. A named buyer with a named situation who should choose the other product. 'If your business is a newsletter with paid subscriptions, buy Kit and stop reading' is an anti-fit. 'Every tool has trade-offs' is nothing.
Where it goes
| Placement | Reads as | Use when |
|---|---|---|
| In the opening summary | Confident and useful | The anti-fit covers a large share of readers |
| Immediately after the comparison table | Honest framing of the data | The table makes you look better than you are |
| A named section of its own | Deliberate, easy to quote | Default choice |
| Last paragraph, after the call to action | A disclaimer nobody reads | Never |
The test
Would a competitor's sales team be comfortable sending this page to a prospect? If yes, it is a genuine anti-fit. If they would need to add context first, it is a hedge dressed as one. That test is harsher than it sounds and most comparison pages fail it, including ones written in good faith.
What this costs
Some readers will leave and buy the competitor. That is not a side effect to be minimised — it is the mechanism. Those readers were going to discover the mismatch during a trial or after a migration, and finding out on the comparison page is cheaper for them and for your support queue. The cost is real and it lands on this quarter's numbers; the benefit lands later and is harder to attribute.
What we cannot show you
Whether anti-fit sections measurably increase citations. We have not run an experiment that would separate that from everything else changing on our pages, and we are not going to imply we have. The claim we will make is about the reader: a page that concedes nothing is discounted entirely, and a page that concedes precisely is trusted on its remaining claims.
FAQ
Does this apply to alternatives pages too? — More so. An alternatives page where every alternative is worse than you is transparently not an alternatives page, and readers arriving on it are specifically looking for the honest version.
How specific should the anti-fit be? — Specific enough that a reader can self-identify in one sentence. 'Large enterprises' is not specific. 'Teams who need client accounts and rebilling' is.
What if we genuinely are better for everyone? — You are not, and a page claiming it will be read as marketing regardless of whether it happens to be true. If you cannot name a buyer who should choose otherwise, you have not understood the competitor's customers.
Put this to work on your own website.
Fenn finds what your customers ask, drafts the articles and site fixes, and measures what ChatGPT, Claude, Gemini, Perplexity and Grok say about you — with every change waiting for your approval.