Are 'X vs Y' comparison pages worth writing for AI search?
Comparison queries almost always trigger an AI Overview. A fair 'X vs Y' page can be the source it cites; a self-serving one rarely is. What to include.
The short answer
Yes, 'X vs Y' comparison pages are worth writing for AI search if they are fair. Seer Interactive found 95.4% of 'X vs Y' queries showed an AI Overview, so an engine usually writes the answer. A page with clear criteria, dated facts and honest trade-offs gives it something to cite. One claiming you win everything rarely gets quoted.
Key takeaways
- In Seer Interactive's panel of 53 brands (January 2025 to February 2026), 95.4% of 'X vs Y' informational queries showed an AI Overview.
- In the same study, brands cited in an AI Overview earned about 120% more organic clicks per impression than when they were not cited.
- A fair comparison page states its criteria, dates its facts and says where the other product is stronger.
- Self-serving comparison pages get cross-checked against review sites and documentation, and the engine quotes those instead.
- Write comparisons for the pairs buyers actually ask about, and recheck every fact on a schedule.
In this article
Why comparison queries matter more in AI search
Comparison queries are where AI answers appear most reliably. Seer Interactive, a search agency, tracked 53 brands from January 2025 to February 2026. It found 95.4% of "X vs Y" informational queries (267 of 280) showed an AI Overview.1 Across all informational queries in the same study the rate was 36%, against 8% for commercial and 5% for transactional queries.1
So when a buyer types "Acme Search Co. vs Contoso Growth", an engine writes the comparison before the buyer sees a single link. The question for a brand is no longer whether to rank for the comparison. It is whether the engine cites your page while writing it.
Being cited is worth a lot. In the same Seer study, brands cited in an AI Overview earned about 120% more organic clicks per impression than when they were not cited.1 Seer adds that cited pages still trailed queries with no AI Overview by 38% on informational searches, so citation narrows the loss rather than removing it.1
Comparison queries
Almost every 'X vs Y' search gets an AI answer
What does the research say about comparison pages?
The evidence is encouraging but correlational. Siege Media, a content agency, published research in June 2026 on which transactional page types on business-to-business sites go with traffic from AI assistants, with versus pages among the types it compares.2 Its write-up also discusses the part brand authority and category fit play, and it is worth reading in full before you plan a programme.
Correlation is not a promise. Sites with many comparison pages may also be the sites with strong brands, active review profiles and a content team that ships. Our view: comparison pages are worth writing because buyers ask comparison questions and engines answer them, not because a study found an association. If they also help AI traffic, good.
Engines also differ in how much room they leave. In Semrush's 2026 AI Visibility Index of 126 million US prompts (January to April 2026), ChatGPT cited about 15 sources per response and Gemini about 3.3 On an engine that cites few sources, a comparison page competes for a very small number of slots.
What should a fair comparison page include?
A fair page gives the engine and the buyer what they need to decide, including reasons to choose the other product. Six elements do most of the work.
- An answer first: one or two sentences saying who each product suits best.
- A criteria table with the dimensions buyers actually compare, such as price structure, integrations, deployment and support.
- Facts with sources: link to the competitor's own pricing and documentation for every claim about them.
- A date: "Checked on" with the day you verified the facts, and a commitment to recheck.
- Honest trade-offs: at least one row where the other product is stronger, said plainly.
- Who should not buy you: the use cases where the competitor is the better fit.
That last element feels risky, and it is the one that makes the page credible. A buyer who reads "if you need X, the other product handles it better" believes the rest of the table. So does a reviewer checking the page against independent sources.
Fair vs self-serving comparison pages
Self-serving
- Wins on every row
- Competitor facts without sources or dates
- Vague claims: "easier", "more powerful"
- No word on who should buy the other product
Fair
- Answer first: who each product suits
- Every competitor fact linked and dated
- Specific, checkable differences
- Plain trade-offs, including where you lose
Where self-serving comparison pages backfire
They backfire when an engine can check them, which is most of the time. Engines that cite several sources put your page next to review sites, the competitor's documentation and forum threads. If your page says the competitor lacks a feature their documentation describes, the answer may quote the documentation and leave you looking careless.
They also backfire with buyers. A comparison that reads like an advert sends the buyer to a third-party review, which is where they were going anyway. Our view: a self-serving comparison page is worse than none, because it hands the engine and the buyer a reason to distrust your other pages.
Outdated facts cause the same damage without any bad intent. Prices change, features ship and plans get renamed. A comparison page checked once and left for a year will contain errors, and engines will repeat them. Put a recheck date on every comparison page and keep it.
Which comparisons should you write first?
Write the comparisons buyers actually make, starting with the competitor that appears in most of your deals. Three sources tell you which pairs those are.
- Your sales team: which competitor comes up in most late-stage deals.
- Your prompt runs: which brands engines name next to yours when buyers ask for options.
- Search Console: which comparison queries already bring you impressions.
Then add "alternatives to" pages for the market leader and for your own product, written to the same standard. Our guide to the seven pages a SaaS site must get right puts comparisons and alternatives in context with pricing, integrations and security.
Ranking still helps but does not settle the question. Our explainer on how often top-ten pages are cited in AI Overviews covers that overlap. Technical extras matter less than content here: our review of schema, llms.txt and Reddit evidence weighs what markup can and cannot do for citations.
How to check whether it worked
Run the comparison prompts before you publish and again a month later, on several engines and several days. Record whether the engine names you, what it says about the difference, and which sources it cites. A comparison page has worked when the engine's description of the trade-off matches yours and cites your page or a source you can live with.
Our AI answer monitor runs these checks across ChatGPT, Gemini, Perplexity, Claude, Copilot and Google AI, using official APIs where they are offered and a licensed AI-answer data provider otherwise. In a Sigzen AI audit, each prompt runs on three separate days, because one answer is an anecdote, not a measurement.
