An Alternatives Page Structure That Is Not a List of Worse Products
Organise alternatives by why someone is leaving rather than by rank. Readers arrive at an alternatives page with a specific grievance — too expensive, missing a capability, too complex, the wrong shape for their team — and a ranked list answers a question they did not ask. Sorted by exit reason, each entry says which grievance it solves and what it costs in exchange, and your own entry appears against the specific reasons it genuinely addresses rather than at the top of everything. The page is more useful, and it stops making the claim that undermines every alternatives page: that the publisher happens to be the best answer to every reason a person might leave.
We publish alternatives pages, and the structural argument here is one we apply to our own with mixed success. It is easier to describe than to hold to when a page is being written under a traffic target.
Why ranking fails here
A ranked list implies one ordering of quality. The reader does not have a quality question; they have a specific problem with their current tool. Someone leaving on price and someone leaving on missing enterprise controls need opposite recommendations, and a single ranking serves neither.
The structure
| Section | Contains | Why it works |
|---|---|---|
| Exit reasons, named | The three or four reasons people actually leave | Reader self-identifies immediately |
| One block per reason | The best answer for that reason, with its cost | Answers the question they have |
| Where you genuinely fit | Only the reasons you actually address | Credible because it is bounded |
| Where you do not | Named reasons to pick someone else | The part that gets quoted |
| Migration reality | What does not transfer | The thing nobody publishes |
Say what does not transfer
Every migration loses something: historical data in a proprietary format, automations that have to be rebuilt rather than exported, integrations that exist on one side only. Publishing that is against the short-term interest of a page designed to win switchers, and it is the section a reader evaluating seriously is looking for. A page without it is read as a page that has not thought about migration, or one that has and is not saying.
Let entries win
If a competitor is the right answer for one of the exit reasons, say so in that block. This is the same argument as the anti-fit section on a comparison page and it is harder to hold to here, because the whole page format exists to capture switchers. The compensation is that the blocks where you do fit become believable.
FAQ
Does this reduce conversions from the page? — Probably from that page, yes. Whether it reduces conversions overall depends on how many of the people it sends elsewhere would have churned after a bad-fit trial, which is a number most teams do not measure and should.
How many alternatives should be listed? — Enough to cover the real exit reasons, usually five or six. A list of fifteen is a list nobody reads and is a signal that the page is built for search rather than for a reader.
Should we write alternatives pages for products we do not compete with? — No. An alternatives page for a product whose customers are not your customers is the clearest form of writing for a query rather than for a person, and it produces a page that cannot make a single honest recommendation.
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.