A versus page is a web page built to answer one question: which of two named products is the better choice. The format takes its name from the query pattern that produces it, "Brand A vs Brand B," and it exists because search demand concentrates at the moment a buyer has already narrowed a category to a shortlist. A category explainer serves someone who does not yet know what the options are. A versus page assumes that stage is finished and supplies a decision between two of them. It is the close relative of the listicle, the "best [category] in [year]" ranking, and both formats are usually commissioned from the same content teams under the same brief.

The commercial logic separates them cleanly. Listicles harvest demand from buyers assembling a shortlist. Versus pages harvest demand from buyers who already have one, which is why the queries carry low volume and high conversion intent.

How a versus page is assembled

The keyword layer comes first. Practitioners work from a modifier set built around "vs," "versus," "or," "compared to" and "alternative to," then multiply it across a competitive set. The arithmetic is unforgiving: twenty vendors generate 190 unique pairings, and search engines treat "A vs B" and "B vs A" as separate strings even though one ordering usually captures most of the volume. That combinatorial property turns a content plan into hundreds of near-identical pages, and it is the structural reason the format keeps colliding with spam policy.

Page anatomy has converged on a recognizable template: a verdict paragraph, a side-by-side feature table, a pricing comparison, segmentation by use case, and often a migration section. Structured data support is thin. Schema.org publishes no comparison type, so publishers fall back on Product, Review, ItemList and FAQPage markup, none of which describes a pairwise comparison directly. The table does the semantic work, which is one reason extraction systems find the format legible.

Three parties publish versus content, with sharply divergent incentives. First-party pages are written by one of the two vendors named, a structure in which the author is also a competitor. Second-party pages come from a rival targeting the same pair. Third-party pages come from affiliate publishers, review sites and software marketplaces such as G2 and Capterra, which monetize through referral fees. Only the third category is nominally disinterested, and affiliate economics complicate even that.

Where the format sits in the search stack

On a conventional results page, versus content competes for the standard organic slots plus People Also Ask placements. Inside generative surfaces the position changes. Comparison and ranking formats are over-represented in citations from large language models, a pattern Semrush documented while running its own visibility program, focusing on listicles because they frequently appear in LLM citations. An analysis of 23 factors influencing AI citation, published May 7, 2026 by Cyrus Shepard, found content length scored 6.7 with inconsistent evidence, while structure and extractability mattered more.

Compression is the other change. Productrise compared standard Google search against AI Mode across more than 100,000 identical shopping searches over 21 days in July 2026 and found AI Mode returned roughly 95 percent fewer product listings, averaging 4.3 listings per answer against 22.5 in standard search. Fewer slots raise the value of occupying one.

Google has built commercial surfaces around the comparison behavior itself. When the company announced shopping ads for AI Mode in February 2026, it cited commissioned research with 4,773 participants conducted between October and December 2025, concluding that AI Mode works better when users can easily compare brands and stores.

Origin and evolution

Editorial comparison predates search marketing, but the modern versus page descends from affiliate review publishing, which scaled after Amazon launched its associates program in 1996.

That changed on April 8, 2021, when Google shipped the product reviews update, an English-only ranking change intended to reward in-depth research over content that summarized a set of products thinly. The accompanying guidance made comparison an explicit quality criterion: Google's documentation on writing high quality reviews instructs publishers to explain what sets something apart from competitors and to cover comparable alternatives. A ranking system was now asking for versus content by name.

The system broadened on April 12, 2023, when the update dropped "product" from its title and began evaluating any content offering a recommendation, opinion or analysis. The November 8, 2023 iteration was the last announced publicly; the reviews system has since run on continuous improvement, removing the discrete update dates practitioners used to diagnose ranking losses.

Enforcement arrived from a different direction. On March 5, 2024, Google introduced spam policies covering scaled content abuse, expired domain abuse and site reputation abuse, the last effective May 5, 2024. A documentation refresh on September 25, 2024 renamed doorways to doorway abuse, aligning the older tactic with the newer vocabulary, a grouping covered in PPC Land's piece on doorway pages. The site reputation policy later drew regulatory attention when the European Commission prepared a Digital Markets Act probe in November 2025, and Google's Chief Scientist for Search Pandu Nayak called the investigation misguided.

The paid search mirror

Versus pages have a direct counterpart in bidding on competitor brand terms. Google's trademark policy is complaint-driven: restrictions on trademarked terms in ad text apply once an owner files a valid complaint. Exceptions exist for resellers and informational sites, but the policy explicitly excludes ads referring to a trademark for competitive purposes, which means a versus landing page can generally be bid against a rival's brand keyword while the rival's mark stays out of the creative.

Identity signaling tightened further when Google expanded its limited ad serving policy to Search from June 2026, noting that ads referencing other brands, and generic ads carrying no branding, may confuse users about who is advertising.

Rules governing comparative claims

Two regulatory frameworks apply, and neither was written for search. In the United States, the Federal Trade Commission's Statement of Policy Regarding Comparative Advertising, adopted August 13, 1979 and codified at 16 CFR 14.15, defines comparative advertising as comparing alternative brands on objectively measurable attributes or price while identifying the brand by name. The Commission encourages the practice and treats industry restraints on truthful comparison as unwarranted, but requires clarity and substantiation.

In the European Union, Directive 2006/114/EC of December 12, 2006 sets cumulative conditions in Article 4. Comparisons must not mislead, must address goods meeting the same needs, must compare material, relevant, verifiable and representative features, must not discredit a competitor's trade marks, and must not take unfair advantage of a rival's reputation. Verifiability is the condition most versus pages struggle with.

Disambiguation

Listicle. A ranked set of many options rather than a pairing of two. Listicles absorb category-level demand; versus pages absorb pair-level demand.

Alternatives page. Targets "alternatives to [Brand]" queries, a one-to-many structure that names a single incumbent and several replacements. It shares the versus page's buying-stage intent with the listicle's shape.

Comparative advertising. A paid claim inside an advertisement, regulated by the FTC and Directive 2006/114/EC. Versus content is editorial output that may carry no advertising claim at all, though the distinction thins when the publisher is one of the compared parties.

Comparison shopping service. In the European regulatory sense, an intermediary listing merchant product ads, a category shaped by the 2017 Google Shopping decision and described in PPC Land's coverage of the CSS Summit. The overlap is lexical only.

Limitations and disputes

The format's central weakness is that the author is frequently one of the compared parties. Lily Ray, vice president of SEO strategy and research at Amsive, flagged FTC requirements for transparency and substantiation when a company claims competitive superiority, arguing that testing every competitor and disclosing the methodology is the threshold and that most publishers do not clear it.

Generative surfaces amplified the problem before policy caught up. Self-promotional listicles were documented inside AI Overviews in May 2025, where publishing an article naming a company as best in category was sufficient to have the claim repeated as though independently sourced. Ray described the tactic as remarkably simple. The mechanism generalizes: AI systems treat citation presence as a quality signal regardless of the citing source's interest, and an Ahrefs experiment with a fabricated brand tested how far that extends.

Correction came in January 2026. Ray observed on January 20 that dozens of companies took site-level hits, almost all of them holding large volumes of listicles, with one publisher running 2,000 articles claiming first position in its category. SISTRIX reported that affected sites lost up to 49 percent of visibility, sharing characteristics beyond the listicles themselves: scaled output, AI-generated text without editorial oversight, artificial freshness updates and schema misuse. Whether this represented a discrete signal aimed at self-ranking comparison content or a side effect of the December 2025 core update remains unresolved, and Google has published no policy naming the tactic.

Publishers dispute the enforcement in the opposite direction. At Google's October 2024 web creator summit, independent operators asked whether adding pros-and-cons boxes to more than 1,000 articles would trigger penalties. Others argue enforcement lands unevenly: fabricated earnings stories persisted on Google Finance through two 2026 spam updates.

Recent developments

Google formally extended its spam policies to AI Overviews and AI Mode on May 15, 2026, roughly two years after AI Overviews launched at scale, closing a documentation gap measured in years. On July 24, 2026 it added a guideline prohibiting fake and undisclosed incentivized reviews, enforceable through manual action, reaching coordinated promotional posting on third-party platforms as Reddit's role in product research draws scrutiny. A third spam update of 2026 reached every language on August 18. Policy text and detection capability remain separate questions.

Timeline

  • 1996: Amazon launches its associates program, seeding affiliate review and comparison publishing
  • August 13, 1979: FTC adopts its Statement of Policy Regarding Comparative Advertising, codified at 16 CFR 14.15
  • December 12, 2006: EU adopts Directive 2006/114/EC on misleading and comparative advertising
  • April 8, 2021: Google ships the product reviews update, rewarding comparison against alternatives
  • August 25, 2022: Helpful content update targets content written primarily for search engines
  • April 12, 2023: Reviews update broadens beyond products to any recommendation or analysis content
  • November 8, 2023: Final publicly announced reviews update; the system moves to continuous improvement
  • March 5, 2024: Google announces scaled content abuse, expired domain abuse and site reputation abuse policies
  • May 5, 2024: Site reputation abuse policy takes effect; manual enforcement begins
  • September 25, 2024: Doorways renamed doorway abuse in a documentation refresh
  • May 14, 2025: Analysis published documenting self-promotional listicles cited inside AI Overviews
  • November 13, 2025: European Commission opens a Digital Markets Act investigation into the site reputation abuse policy
  • January 20, 2026: Wave of site-level impacts observed across publishers holding large listicle inventories
  • February 2026: Shopping ads launch inside Google AI Mode
  • May 15, 2026: Google confirms spam policies apply to AI Overviews and AI Mode
  • July 2026: Productrise measures 95 percent fewer product listings in AI Mode than standard search
  • July 24, 2026: Google adds a guideline prohibiting fake and undisclosed incentivized reviews
  • August 18, 2026: Third spam update of 2026 rolls out to every language

Summary

Who. Three groups publish versus content: vendors writing about themselves and a named rival, competitors targeting the same pairing, and third-party publishers including affiliates and software marketplaces. Google defines the ranking and spam policies that govern the format, the FTC and the European Commission set the substantiation rules for comparative claims, and SEO practitioners including Lily Ray and analytics firms such as SISTRIX have documented its failure modes.

What. A page built to resolve a head-to-head query between two named products, typically combining a verdict, a feature table and pricing comparison, sitting one funnel stage below the listicle and targeting low-volume, high-intent queries.

When. The format predates search policy but entered it on April 8, 2021 with Google's product reviews update, broadened on April 12, 2023, moved to continuous evaluation on November 8, 2023, and came under enforcement pressure through the March 5, 2024 spam policies and the January 20, 2026 wave of site-level impacts.

Where. Organic search results, People Also Ask panels, AI Overviews, AI Mode and large language model citations, alongside a paid counterpart in competitor brand-term bidding governed by Google's complaint-driven trademark policy.

Why. Comparison queries arrive from buyers at the decision point, which makes the format disproportionately valuable per visit and disproportionately vulnerable to manipulation. The unresolved tension is structural: Google's own reviews guidance asks publishers to compare against competitors, while its enforcement systems and the FTC's substantiation standard treat self-serving comparisons as a quality and legal risk.