Exa ATLAS and Tavily Test Different Kinds of Search Completeness
Exa ATLAS and Tavily’s company benchmark use different tasks and scores. A worked example shows why their results cannot form one search leaderboard.
Latest research
Completed tests, case studies, and evidence reviews
Case studies, experiments and analysis of original research.
-
-
CITECHOICE Study: More Citation Credit, Uncertain First-Citation Gains
A controlled AI-search replay found more citation markers for structured source text, but no clear gain in whether a source was cited at all. Here is what publishers should measure.
-
ChatGPT and Gemini Shared No Source Domains in 76.7% of Product Checks
A product-advice preprint found no shared displayed source domains in 76.7% of matched ChatGPT and Gemini checks. What the test means for AI visibility audits.
-
AuthGR: What Authority-Aware Generative Retrieval Changes
An ACL 2026 industry paper connects authority-aware retrieval with stronger ranking judgments and higher click behavior inside a commercial search system.
-
13.4 Billion Web Questions Show FAQ Frequency Is Not Search Demand
The most repeated questions on the web increasingly come from templates and commerce pages. Publishers need behavioral demand evidence before creating another FAQ.
-
AI Answer Citations Can Fail at Four Separate Layers
Four recent papers expose separate failure points between retrieval and a trustworthy AI citation. SearchEngineAnswer reproduced PageRecall’s public tier-0 checks and built a claim-level audit ledger.
-
AI Search Engines Cited Different Websites in a 960-Answer Study.
A 960-answer experiment and a separate B2B panel show why AI citations vary by engine, prompt type and measurement unit. Their percentages should not be pooled.
-
CiteShade Raised Wrong Answers From 1% to 68%
A multi-source RAG attack made incorrect answers cite trusted evidence. Source-deletion tests reveal which document actually drove the claim.
-
CITECHOICE Finds Structure Shifted Citation Credit, Not Source Admission
A causal replay found that structured rendering increased citation count, while its effect on whether a source was cited at all remained inconclusive.
From this desk
Further reading
Guides and practical resources for the questions behind the news.
Read the flagship case study
How SearchEngineAnswer went from zero to early search visibility
This first-person build record connects publishing decisions with indexing evidence, Search Console observations, limitations, and the next measurement window.
Case study
Connect decisions with observed outcomes
Follow the site build, indexing evidence, changes, and limits instead of treating one visibility event as a universal rule.
Experiments
Change one material variable
Preserve the baseline, observation window, confounders, and decision rule so a result remains inspectable.
Evidence
Keep every claim close to support
Classify facts, observations, interpretations, recommendations, and opinions before drafting conclusions.
A useful standard
How to read SearchEngineAnswer research
A useful study makes its conditions visible. Treat each result as evidence within a declared sample and environment, not as a universal platform rule.
Method before conclusion
The question, inputs, environment, measures, and stopping rule should be understandable before the result is interpreted.
Failures stay in the record
Timeouts, absent results, rejected cases, and contradictory observations are data, not material to hide.
Limits remain visible
A measured association or one platform observation does not automatically establish causation or a permanent ranking rule.
References and next steps