· 8 min read·by Thiago Nogueira

The "best in your sector" list was written by a competitor

Whoever writes the category comparison decides who is in it and in what order. In the data we measured, it is almost always someone with something to sell at the end.

contentcomparisonsoftwarefurnituregeo

What showed up when we asked for a comparison

Measuring ten Brazilian markets, software gave the most direct result of all.

We measured the page-one positions for the sector's real buying questions. That produced 19 distinct domains across 20 positions, and the overwhelming majority was the same thing: a "the 12 best management systems" article on a vendor's blog.

In other words, the one answering "which ERP should I choose" for your customer is another ERP vendor.

In furniture the pattern repeated with an interesting variation. The theme that asks for brands returned five ranking articles and no brands. Two of those lists were hosted on the sites of retailers in the sector, ranking precisely the brands they resell.

Why this matters more than it looks

A ranking list is not just any article. It is the page an AI assistant uses to answer a comparison question, because it is the only one that arrives with the comparison already done.

When the assistant summarizes that page, it inherits three decisions somebody made: who is in the list, in what order, and by what criteria. None of the three is neutral when the writer sells something at the end.

And the customer sees none of it. They receive three names and a justification, with the same confidence they would receive a fact.

The answer is not to complain about the list

The common reaction is to treat this as unfair. It is not. It is the consequence of a difference in behaviour.

The vendor who wrote the list invested in producing content about the category. The brand that does not appear produced content about itself. Both are legitimate, except the question the customer asks is about the category.

The answer is to write the comparison too. With one important difference: if it is advertising in disguise, it does not work, not with readers and not with assistants.

Why the obvious reaction usually makes it worse

The first reaction of almost every brand is to publish its own version of the list. It almost always comes out wrong, in a predictable way.

The text becomes "why we are the best". An AI assistant reads that as a company page, because that is what it is, and goes on citing the competitor's list, which at least presents alternatives. The reader does the same thing, only faster.

A comparison that works has to admit there are cases where another option fits better, it needs stated criteria, and it has to be backed by data someone can verify. That is uncomfortable to write about your own product and it is exactly what separates a cited page from one more piece of sales material.

The most common mistake

Publishing "why we are the best" and calling it a comparison.

An AI assistant reads that as a company page, because that is what it is, and goes on citing the competitor's list, which at least presents alternatives. The reader does the same thing, only faster.

A useful comparison admits there are cases where another option fits better. That feels like a loss and it is precisely what makes the rest of the text credible.

Where we come in

Finding out who wrote the comparison that decides your category, and what it would take for your brand to be in it, is measurement before it is writing.

It is the first step of any engagement we run: seeing what surfaces today when someone asks about your category, who holds that space, and by what criteria. The first check on your brand is free.

The public data behind this study is on the software and furniture pages.

Frequently asked questions

Is it not risky to name competitors on my own page?
The comparison already exists and already names your competitors. The difference is that today it was written by someone with no interest in describing you well. Naming alternatives is what gives the rest of the text credibility.
How long should such a page be?
Long enough to answer the question with explicit criteria. In the results we measured, what repeats is not length but structure: stated criteria, named alternatives, and a recommendation per use profile.
Does this work for a small brand?
It works better on a specific slice. A comparison about one product type or one defined audience is defensible by whoever knows that slice, and that is where a smaller brand has a real edge over a big competitor’s generic list.

Liked this? Get the next one in your inbox.

One or two long-form articles on AI search visibility every month. Same depth as this piece. No filler.

We use double opt-in. By signing up you accept the privacy policy. Unsubscribe in one click from any email.

Want this analysis run on your brand?

Every Citorial plan ships an action plan based on the same techniques, and we execute it for you.

See pricing