Everyone says to build a “[your brand] alternatives” page. The evidence points at a different page.

/ 8 min read / By Faz

Every guide I could find on this gives the same advice. Buyers ask AI for alternatives to your product, the engine answers whether you like it or not, so publish your own page and shape the answer. It is a neat argument, and I expected to agree with it. Then I looked for the evidence behind it and found reasoning, a few opinions, and no test.

So I ran one. It is small: four well-known software companies, a set of questions on Google AI Mode in November, several repeat runs where the answers moved around. For each company I asked for its alternatives, read every source I could see, and checked whether it publishes a page about its own alternatives. I have left the companies unnamed, because nothing here is a criticism of their teams.

Where I landed: a self-written alternatives page is rarely the first thing to build. A head-to-head page against the rival you actually lose to usually is.

The sentence that matters is the first one

Before looking at any company’s page, read how these answers open. They did not start with a list. They started by explaining why people leave you.

For a leading contract platform, the opening said companies look elsewhere to avoid unlisted premium pricing, complex setups, and strict seat licensing for AI features. For a well-known work management tool, it said key features are heavily paywalled. For an automation platform, it pointed to steep task-based pricing.

That sentence is the most damaging thing in the answer, and it is what a page about your own alternatives ought to influence. So the useful question is who wrote it.

For the contract platform, the opening drew on a competitor’s “best alternatives” page. Of the eight sources shown, at least three came from competitors, and none came from the company. That competitor did not even make the recommendation list, despite ranking itself first on its own page. Its reason for leaving made the answer, and its product did not.

For the work management tool, the sources included its own subreddit, a thread in a rival’s subreddit, an independent review, and comparison pages published by two competitors.

Your exit story is being written right now by people with a reason to write it. That part of the guides is correct. Their remedy is where it gets shaky.

Two companies did publish the page

The work management company has exactly what the guides recommend: a current page comparing fifteen alternatives to itself. It deserves some credit, too. It says plainly which rivals suit which teams. It also lists itself first, and it is the only entry of the fifteen with no considerations section.

I ran the question eight times, in five phrasings, and never saw that page among the citations. That needs a caveat: five of those runs showed no citations at all, one list was collapsed, and one was cut off when I read it. The one complete list had five sources, four from competitors and one a YouTube video. So this is not proof the page was never used. But the answers still raised the paywalled features, the seat minimums, and rising costs, and nothing I could read traced those back to the company.

The second company, in SEO software, has a page with “alternatives” in the title that is really a list of reasons to stay. That is a reasonable thing for a marketing team to publish. It is also more than two years old, and it was not among the seven sources in the run I read, which went straight to the rivals.

A page ranking yourself first in a list of your own replacements is a sales page in the format of a neutral one. We made the same point about lists written by vendors. It is harder to pull off when the vendor is the very product the reader is thinking of leaving.

The same company’s head-to-head page was the top source

Here is the result that changed my mind. I then asked the same engine to compare that work management company with its best-known rival. The first source in the list was the company’s own head-to-head page. That is one run, so hold it loosely.

Same company, same blog, similar polish. One page topped the list behind the answer, and the other was nowhere I could see.

The contract platform showed the same pattern from the other side. It has no page about its own alternatives that I could find, and it did not appear in the sources for that query. But it does publish a page on a specific rival. When I asked for that head-to-head, the company’s page was cited for several of the points in its favour, the rival’s page was cited too, and the engine picked the company for most sales-driven and fast-growing organisations while giving the rival a clear win at the very large enterprise end.

The automation platform was the most useful case. It also has no page about its own alternatives that I could find, but it has a comparison hub, and its page on an enterprise rival showed up in the sources for the alternatives question. The answer then described that rival much as the page does, as a tool for IT teams, in a separate enterprise section with its own price tier. I cannot prove the page caused that, since other sources were cited for the same section. But notice what it would mean if it did: the company would not have argued its way out of the alternatives answer. It would have helped put a rival in a box where that rival was no threat to most readers.

Four companies is not a study. But it’s a consistent direction: a page about one specific rival gets read, and a page about everyone gets passed over. A head-to-head page answers a question the engine can use. Your list of alternatives to yourself competes with a dozen lists that have no stake in keeping you.

What to do, in order

Read your own opening sentence first. Ask for alternatives to your product several times, in a few phrasings, because the list moves. On the work management tool, the shortlist changed on almost every run and only two rivals appeared every time. Write down the reasons given for leaving.

Trace each reason to its source. If it is true, no page will fix it. Publish the price, explain the limit, or change the plan. If it is stale or wrong, go to the page repeating it and get that corrected. Your own page will not outvote eight others.

Then build head-to-head pages for the rivals named most often, and only those. Write them the way that gets cited, by conceding where the rival wins. That post covers how to win the query. The point here is that these pages come first.

Build the page about your own alternatives last, and only if you can pass this test. Would you publish it without yourself at number one? Would you give your own entry the same weaknesses section as everyone else? Would you state plainly who should leave, and to where? If not, you are building a sales page that buyers and engines will both discount, and your time is better spent elsewhere.

If you can pass that test and people are searching your name plus “alternatives”, it may be worth doing. Its job is narrow: stating honest reasons to leave, so that the true version of that first sentence exists somewhere. I should be clear that I did not find a page like this in the wild, so I cannot show you one working. That is the argument for trying it, and also the reason to keep your expectations modest.

What I got wrong

I read the first run for the work management tool, saw three sources, and started drawing conclusions as if those were all of them. They were only the ones displayed before the “show all” button. When I re-ran it, I found more sources, including an independent review with hands-on testing, and a different shortlist of rivals. Later runs showed no sources at all.

That matters twice over. A collapsed citation list makes it easy to declare a page absent when it is sitting below the fold, which is why every “not cited” claim above is limited to what I could actually read. And a single run gives you one answer out of many. If I had stopped at the first run, I would have told you confidently who your rivals are, and been wrong about half of them.

Where this leaves you

The guides are right that the engine answers this question without you, and right that you should care. The evidence I could gather says to spend your effort somewhere else first: read why the engine says people leave you, fix what is fixable at the source, and build honest pages against the two or three rivals who actually take your deals. A page listing your own alternatives comes after that, and only if it can tell the truth, including the part where you are the wrong choice.

If you want to see what the engines say when your buyers ask how to leave you, and who wrote it, that is part of the engagement, and it runs on a documented method.

Want this run for your B2B SaaS?

Founding pricing for the first 5 clients. Methodology fully public. Month-to-month, cancel anytime.

Apply to work together

Leave a comment

Your email address will not be published. Required fields are marked *