AI Overviews can cite your listicle while recommending your competitors
Search Engine Land reports that Google AI Overviews cited self-promotional “best software” pages but excluded the brands behind them from recommendations in most cases. For GEO teams, the key lesson is simple: citation visibility and recommendation visibility are now different markets.
Editor’s Note by Dageno: Dageno AI’s editorial team structurally distilled Search Engine Land’s report on Lily Ray’s B2B software analysis to help GEO and AI visibility teams separate citation outcomes, recommendation outcomes and the evidence signals that make a brand safer for AI systems to recommend.
Understand the article in 60 seconds
- Lily Ray analyzed 100 B2B “best [category] software” queries across April 15, May 15 and June 8; 80 triggered AI Overviews, giving GEO teams a clear view of commercial-query answer behavior.
- Self-promotional listicles were cited 323 times, but in 224 cases Google cited the brand’s own page without recommending that brand; citations alone are not a win condition.
- The Oasis LMS example shows the mechanism: Google cited Oasis content but recommended Kajabi, Thinkific, LearnWorlds and Teachable, all brands named inside the source page.
- Third-party and user-generated sources such as Forbes, Reddit and YouTube appeared heavily in “best” responses, showing that independent proof can outweigh owned rankings.
A citation can be a receipt while the buying recommendation goes somewhere else
The article’s central mechanism is not that owned listicles stopped being visible. They still entered AI Overview citation sets. The problem is that Google used those pages as evidence about the category and then routed the recommendation to stronger brands with broader external corroboration.
For Dageno readers, this changes how “AI visibility” should be reported. A brand can win a citation and still lose the buyer moment if the answer names competitors as the better options. GEO measurement therefore needs two ledgers: source visibility and recommendation visibility.
Platform mechanism: Treat citations as retrieval evidence, not commercial share of voice. The business outcome is whether the AI answer selects, mentions and recommends the brand for the intent that matters.
Lily Ray’s B2B software data separates citation from recommendation
The source article, “Google AI Overviews cite self-serving listicles, but recommend competitors 69% of the time,” is best treated as a set of operating signals: fact, mechanism, impact, team response and metric.
100 commercial software prompts
The analysis covered 100 “best [category] software” prompts and checked Google AI Overview behavior on three dates. That matters because the study looked at bottom-of-funnel category questions, not generic informational searches.
323 self-promotional citations
Owned “best” pages were cited hundreds of times, proving that Google could retrieve those assets. The failure point was not pure crawling or indexing; it was recommendation confidence.
224 citation-without-recommendation cases
In 224 instances, a brand’s own page appeared as a source while the brand was not named as a recommended option. That is the clearest split between citation rate and answer-share.
External sources gained weight
Forbes, Reddit and YouTube appeared among highly cited domains for “best” responses. AI systems seem to prefer sources that provide distance, lived context or third-party comparison language.
AI Overviews are judging category authority, not just harvesting owned rankings
The system behavior matters because AI answers are assembled through retrieval, ranking, extraction and synthesis rather than a single static ranking position.
A self-ranked list can teach Google which competitors belong in a category. If the brand lacks independent support, the page may strengthen the competitor graph more than its own recommendation odds.
Ray’s analysis found better-known category leaders with stronger link profiles and wider third-party mentions were more likely to be recommended. GEO therefore depends on entity strength outside the owned site.
Search Engine Land also notes FTC risk when company-controlled content is presented as independent review material. Disclosures, actual experience and relationship clarity affect trust as well as legal exposure.
Commercial prompts now reward third-party proof and entity strength
Dageno’s GEO reading turns the article into practical visibility questions for AI Overviews, LLM answers, brand entities, third-party sources and competitor answer-share.
- AI Overview inclusion should be split into “source cited” and “brand recommended,” because those outcomes can move in opposite directions.
- LLM brand mention rate should be measured on non-branded category prompts, not only prompts that include the company name.
- Competitor answer-share must identify when owned assets help competitors become the named recommendation.
- Third-party source presence should include review sites, forums, YouTube explainers and editorial comparisons, not only backlinks.
- Commercial query coverage should compare best, alternative, comparison and “for [use case]” prompt families.
Owned “best” pages need proof, disclosure and external corroboration
The content response should produce evidence that machines can retrieve and humans can trust. That means specific facts, clear ownership, updated context and externally supported claims.
Rebuild listicles as transparent buyer guides
Keep category education, but disclose relationships, separate evidence from claims and avoid presenting owned rankings as independent review journalism.
Create third-party proof paths
Pursue customer reviews, partner explanations, analyst mentions, community discussions and video walkthroughs that make the brand credible beyond its own domain.
Give AI extractable product evidence
Publish feature matrices, use-case pages, integration facts, pricing context, customer-fit descriptions and limitations in language an answer engine can quote accurately.
Audit competitor leakage
If a page names several competitors but provides weak proof for the owned brand, it may act as a competitor recommendation asset. Rewrite around evidence, not vanity ranking.
GEO, SEO and brand teams must move from self-ranking to evidence design
The finding is cross-functional. If only SEO reacts, the brand may fix rankings while leaving answer framing, source quality, paid learning and attribution untouched.
| Team / owner | What changes | What to do next | Metric or signal to monitor |
|---|---|---|---|
| GEO team | Visibility is no longer equal to being cited. | Track prompt families where the brand is cited, mentioned and recommended separately. | Citation rate, recommendation rate, competitor answer-share. |
| SEO team | Ranking a “best” page is not enough if the answer recommends rivals. | Map category pages to independent corroboration and strengthen internal entity links. | AIO inclusion, non-branded prompt visibility, source diversity. |
| Content strategy | Generic self-rankings create weak evidence and possible trust risk. | Convert listicles into transparent decision guides with proof, methodology and limitations. | Extractable facts, disclosure coverage, content-to-proof links. |
| Brand / PR | AI systems look for external confirmation of category leadership. | Secure credible third-party mentions, reviews, comparisons and customer narratives. | Third-party citation share, sentiment, review-source presence. |
| Analytics | Organic traffic declines may hide answer-level visibility shifts. | Report source citations and recommendation outcomes by prompt cluster and date. | Visibility trend since core updates, answer-share delta, landing-page assist. |
| Legal / compliance | Review-style pages can create FTC exposure if independence is implied. | Check disclosures, material relationships and evidence quality before scaling “best” content. | Disclosure completeness, review-basis documentation, risk flags. |
Measure citation share and recommendation share as separate outcomes
The measurement layer should explain what changed, why it changed and which team can act. These signals are designed to expose GEO problems before they appear as lost revenue.
Shows retrieval, but not commercial success; owned pages may be used as evidence for competitors.
Measures whether the brand becomes a named option in the answer, the closest proxy to AI answer demand capture.
Reveals when your content helps stronger competitors occupy the recommendation slot.
Counts credible external sources that reinforce the brand’s category position.
Identifies whether leaks happen in “best,” “alternative,” “comparison,” or use-case prompts.
Signals that a self-serving listicle strategy is leaking visibility
The risk signals below are not generic performance warnings. Each one points to a specific visibility failure that can be investigated and assigned to an owner.
- Your “best” page is cited but the brand is absent from recommended options.
- Competitors named on your page repeatedly appear above you in AI Overview recommendations.
- Review language lacks methodology, disclosures or evidence of real product use.
- Traffic declines cluster around pages built from AI-generated comparisons and self-ranked lists.
- Reddit, YouTube or editorial sources define the category better than your own site does.
The next rebuild should start with independent proof paths
The fastest path is to turn the article’s evidence into a repeatable operating rhythm: monitor, diagnose, publish, validate and attribute.
Separate source wins from recommendation wins
Build dashboards that split citations, mentions and recommendations by prompt family.
Rewrite buyer guides around verifiable evidence
Replace self-congratulating rankings with criteria, proof, limitations and clear disclosures.
Strengthen the external corroboration layer
Use PR, customer advocacy and partner content to create independent evidence paths.
Repair competitor leakage
Find pages where your copy teaches AI systems more about competitors than about your own fit.
Tie answer-share to pipeline signals
Connect AI recommendation movement to branded search lift, assisted conversions and lead quality.
Turn AI visibility signals into execution.
Dageno AI is a complete execution platform from data monitoring → strategy → content generation → result attribution. It helps teams monitor AI answer surfaces, identify strategy gaps, generate content actions and attribute results to visibility and business impact.
Alexa for Shopping