Testing whether structured entity density and explicit answer formatting on asquaresolutions.com key pages increases AI citation frequency. Baseline established, implementation in progress.
Hypothesis
If we increase entity density and format content for direct answerability on 5 target pages, then citation frequency in Perplexity and ChatGPT search will increase within 30 days.
Does restructuring existing content for higher entity density and direct answerability change how often AI search systems cite those pages? This experiment tests that question against real pages on asquaresolutions.com — a small business services site in the trust verification and scam detection space.
The experiment is in progress. Baseline has been captured. The first two target pages have been optimized. Measurement window runs through mid-June 2026.
If we increase structured entity density (named entities, explicit definitions, Q&A formatting) and direct answerability signals on 5 key pages of asquaresolutions.com,
Then those pages will be cited more frequently in Perplexity and ChatGPT responses to relevant queries within 30 days of the changes going live,
Because AI retrieval systems prioritize content that directly answers a query with minimal inference burden — pages that contain the answer in an extractable, attributable form are more likely to be surfaced and cited than pages that imply the same information through narrative prose.
The mechanism being tested is not SEO ranking — it is retrieval signal density. A page that scores well on traditional SEO may still be poorly optimized for AI citation if its content requires synthesis to extract a direct answer.
Baseline was established between 2026-05-10 and 2026-05-17. The measurement method: run 8 target queries through Perplexity, ChatGPT (GPT-4o web search), and Gemini with web access. Record whether any asquaresolutions.com pages appear in cited sources. Record the query and the citation position when they appear.
| AI System | Queries run | asquaresolutions.com cited | Citation rate |
|---|---|---|---|
| Perplexity | 8 | 1 | 12.5% |
| ChatGPT (web) | 8 | 0 | 0% |
| Gemini (web) | 8 | 1 | 12.5% |
Queries that produced citations:
Queries that produced no citation:
Current content structure observations (pre-optimization):
The target pages are primarily written in narrative marketing prose. Key characteristics that likely reduce AI citation frequency:
ℹBaseline citation rate is low but not zero
The two citations that did appear in the baseline are encouraging — the domain is not being ignored by AI retrieval systems. The hypothesis is that the citations are driven by domain authority signals rather than content answerability. The experiment is testing whether we can increase the citation rate by improving the content signal.
The following content changes are being applied to each target page:
Entity density changes:
Answerability changes:
Structured data:
application/ld+json) to each target page using the same Q&A pairs from the FAQ sectionWhat is not changing:
Five pages selected based on search query volume in Google Search Console and relevance to queries where AI citation is plausible:
TrustSeal landing page (/trustseal) — Primary target. Targets "website trust checker" and "website safety verification" queries. Already receiving 2 AI citations in baseline.
ScamCheck landing page (/scamcheck) — Targets "scam website detector" and "check if website is scam" queries. Zero baseline citations despite high query relevance.
"How to Tell If a Website Is Safe" blog post — Informational content targeting how-to queries. Currently written as narrative advice. High answerability potential with Q&A restructuring.
Website Trust Score explainer (/trust-score-explained) — Definition-heavy page explaining how trust scores work. Ideal for entity density treatment — the entire page is about defining a concept.
Free vs Pro feature comparison page — Lower priority for AI citation, included as a partial control. Minimal changes: only FAQ schema added, no content restructuring.
⚠Page 5 is a partial control, not a true control
Page 5 receives FAQ schema but no content restructuring. This is not a clean control condition — it tests schema markup in isolation. A true control would require leaving a comparable page completely unchanged, which is being maintained by tracking citation rates on three other asquaresolutions.com pages that are receiving no changes in this period.
Measurement cadence: Weekly, for 4 weeks (2026-05-25, 2026-06-01, 2026-06-08, 2026-06-15)
Queries per session: 8 target queries, same queries as baseline, run through Perplexity, ChatGPT (web search on), and Gemini
What is recorded per query:
Limitations acknowledged:
Confounding factors to monitor:
Status: In progress. Baseline captured. Pages 1 and 3 optimized.
| Page | Baseline captured | Content restructured | FAQ schema added | First post-change measurement |
|---|---|---|---|---|
| TrustSeal landing | Yes | Yes | Yes | 2026-05-25 |
| ScamCheck landing | Yes | No | No | — |
| "How to Tell If a Website Is Safe" | Yes | Yes | Yes | 2026-05-25 |
| Trust Score explainer | Yes | No | No | — |
| Feature comparison (partial control) | Yes | No | Yes | 2026-05-25 |
Pages 2 and 4 are scheduled for optimization this week. The measurement clock starts from when each page's changes are deployed to WordPress — the first measurement for all pages will be 2026-05-25 to allow at least 7 days post-change before measuring.
ℹEarly observation from Page 1 optimization
After restructuring the TrustSeal landing page (entity definitions added, first-paragraph answerability rewrite, FAQ section added), an informal test query run on 2026-05-17 showed Perplexity citing the page for "how to check website trust score" — a query that produced no citation in the baseline. This is a single data point, not a result, but it suggests the direction is correct. The formal measurement on 2026-05-25 will establish whether this holds.
The experiment log will be updated with measurement results as each weekly session completes.