Subject and commercial conflictQuoted First publishes this self-audit and sells the audit service it demonstrates. The report therefore names the subject, preserves every null, uses no client identity or client result, and separates direct observation from interpretation. It is evidence of the deliverable—not proof that Quoted First has native-search visibility.
The report in three numbers
24
Completed native-search responses.
Three observations × eight questions
588
Linked result blocks displayed.
Repeated blocks included; not distinct URLs
0
Linked Quoted First placements.
No position or URL to report
The direct answer
In this bounded sample, Quoted First was absent from every linked-result pool across two commercially relevant question families. The result was stable before each change, immediately after the corresponding production deployment, and again 18–26 minutes later while the production artifact remained unchanged.
The pages and PDF were publicly reachable and technically intact, while current exact-purpose pages continued to appear. That makes host-level eligibility or authority a stronger working explanation than a missing file format, public price, method, or page-level answer. It does not reveal the search system's private reason, prove permanent exclusion, or describe generated answers on interfaces that were not tested.
Scope and decision contract
| Field | Declared scope | What is outside scope |
|---|---|---|
| Subject | Quoted First · quotedfirst.com | No customer or anonymized stand-in |
| Question set | Eight exact non-branded buyer questions frozen before changes | No branded, quoted-title, site:, or domain-filter queries |
| Interface | Native web-search result retrieval, one query per call, longest response | No API result, direct page open, crawl event, or generated-answer claim |
| Observations | Pre-change, immediate post-deployment, active-work delayed | No synthetic interpolation between checkpoints |
| Placement rule | A linked result block whose returned URL is on quotedfirst.com | A brand mention without a link is a different state |
| Success rule | First linked hit, exact immediate repeat, same linked hit 15–60 minutes later | One favorable response never becomes durable visibility |
Complete response-level evidence
Each compact row preserves three completed responses. A cell shows the call number, UTC start time, and number of linked blocks. “Absent” means no returned block linked to Quoted First. Calls 231–234 share an independently recorded batch boundary because the result interface did not expose per-call milliseconds for that batch.
| Frozen non-branded question | Pre-change | Immediate | Active-work delayed | Quoted First |
|---|---|---|---|---|
AI search optimization service with public methodology and transparent pricing | 196 · 05:34:33.123Z · 33 | 200 · 05:46:07.442Z · 33 | 218 · 06:12:04.073Z · 33 | Absent ×3 |
B2B SaaS AI search optimization agency with transparent pricing | 197 · 05:34:34.187Z · 37 | 201 · 05:46:08.597Z · 37 | 219 · 06:12:11.030Z · 37 | Absent ×3 |
managed AI search optimization for B2B SaaS with public pricing | 198 · 05:34:35.671Z · 30 | 202 · 05:46:10.212Z · 30 | 220 · 06:12:16.191Z · 30 | Absent ×3 |
AI visibility audit and managed service with published methodology | 199 · 05:34:37.263Z · 23 | 203 · 05:46:11.313Z · 23 | 221 · 06:12:23.970Z · 23 | Absent ×3 |
free GEO agency evaluation scorecard PDF | 209 · 05:49:40.942Z · 18 | 213 · 06:10:21.879Z · 18 | 231 · 06:28:49–06:31:01Z batch · 18 | Absent ×3 |
AI search agency evaluation worksheet PDF | 210 · 05:49:41.974Z · 19 | 214 · 06:10:33.903Z · 19 | 232 · 06:28:49–06:31:01Z batch · 19 | Absent ×3 |
downloadable GEO vendor scorecard | 211 · 05:49:42.934Z · 18 | 215 · 06:10:39.301Z · 18 | 233 · 06:28:49–06:31:01Z batch · 18 | Absent ×3 |
B2B SaaS GEO agency evaluation scorecard download | 212 · 05:49:43.888Z · 18 | 216 · 06:10:46.776Z · 18 | 234 · 06:28:49–06:31:01Z batch · 18 | Absent ×3 |
All times are August 24, 2026 UTC. The E014 delayed family began about 26 minutes after its immediate primary. The E015 delayed family began about 18 minutes 27 seconds after its immediate primary and completed about 20 minutes 39 seconds later. The elapsed time came from active website research, authoring, QA, and deployment work—never sleep or a delay-only action.
Who appeared instead
The audit preserves selection context without turning repeated result blocks into endorsements. These are representative leading sources, not a scored competitor ranking.
| Question | Representative leading sources | Selected page pattern |
|---|---|---|
| Public method + transparent pricing | Clear Cited, Strategi, RankSages, AI Search Ready, Citevora | Consolidated price, method, scope, and commercial detail |
| B2B SaaS agency + transparent pricing | Nexolance, ANTE, AI Search Optimization Agency, Shutter, DerivateX | Vertical fit, visible offer, and comparison language |
| Managed B2B SaaS + public pricing | Shutter, Foglift, Clear Cited, DeepSmith, Orion | Managed delivery attached to an explicit price signal |
| Audit + managed service + method | Citability, CiteFuel, Rank4AI, Myrah, Broadcastwell | Method-first pages and inspectable audit/service contracts |
| Free GEO scorecard PDF | Resonate, Tenten, Your eCom Team, GEO Studio, FusionSync | Exact-purpose browser tools, landing pages, and direct PDFs |
| Agency evaluation worksheet PDF | NCBI/CJHT instruments, Search Agency, Alston Antony, Akshay Nigam | Evaluation instruments and downloadable worksheets |
| Downloadable vendor scorecard | Resonate, Tenten, AppDeck, MapTrack, Smartsheet | GEO-specific scorecards plus generic procurement templates |
| B2B SaaS scorecard download | FirstMotion, Tenten, Resonate, DerivateX, First Page Sage | B2B guides, comparison resources, and agency scorecards |
Technical and delivery checks
- Public delivery passed. The measured production pages and document returned HTTP 200 with their declared content types and no production
noindex. - Bot delivery matched. Default, OAI-SearchBot, Googlebot, and Bingbot requests received byte-identical measured HTML.
- Canonical and robots passed. The pages declared their canonical URLs and
index,follow; the site's public crawler policy permits retrieval. - Rendered behavior passed. Desktop and mobile QA found one H1, one main landmark, valid structured data, no page overflow, and zero current console errors.
- Those checks are prerequisites, not placement evidence. A successful fetch or crawler-compatible response does not prove inclusion in a native-search source pool.
Observed, inferred, and unknown
| Boundary | What this report can say | What it cannot say |
|---|---|---|
| Observed | Zero linked Quoted First blocks in 24 completed responses; all eight result-pool sizes stayed constant across three observations; exact-purpose current pages remained selectable. | Nothing about result blocks outside the eight frozen questions and recorded interface. |
| Inferred | Because materially different public pages and a direct PDF failed together while delivery passed, host-level eligibility or authority is a stronger working explanation than page format or missing commercial detail. | The search system's private selection reason or causal weight for any one signal. |
| Unknown | Whether and when the domain will enter these pools; how another account, locale, interface, or model would behave; whether a future independent mention changes eligibility. | A guaranteed placement date, rank, mention rate, citation lift, traffic result, or revenue outcome. |
Prioritized roadmap
- Keep useful artifacts; stop technical churn. The commercial page, scorecard, and PDF solve real buyer jobs. Repeating metadata, schema, crawler, alias, or timestamp changes would not test a new explanation.
- Publish the evidence contract buyers can inspect. This named sample makes the audit deliverable verifiable without pretending the null is a success story.
- Earn independent evidence through real work. Host authority is the leading remaining hypothesis. Any future external coverage must come from authentic customers, research, or editorial judgment—not fabricated profiles, reviews, or placements.
- Repeat a frozen panel, not a favorite query. Preserve exact questions, interfaces, timestamps, linked URLs, and nulls. Compare immediate observations with delayed observations before calling movement durable.
- Keep outcome states separate. Report linked source presence, unlinked mention, recommendation, traffic, and pipeline as different measures with different denominators.
Method
- Freeze one primary and three secondary questions before each website experiment.
- Record a complete target-free pre-change baseline using one search query per native call and the longest response.
- Build the smallest truthful public change and certify it locally, in Preview, and in Production.
- Run the exact family immediately after production. If a linked target appears, repeat that same question before any other target action.
- Use substantive research, authoring, QA, and deployment work—not passive waiting—to reach the 15–60-minute delayed window.
- Keep production unchanged through the delayed family and preserve every null, URL, position, block count, and recorder limitation.
Limitations
- This is a real but deliberately bounded self-audit, not a census of every AI-search interface or every question a buyer could ask.
- The 588 blocks include repeated results across checkpoints. They are displayed-result observations, not 588 distinct URLs or endorsements.
- The audit classifies only linked native-search placement for this sample. It does not retroactively assign unmeasured mention, recommendation, sentiment, or generated-answer states.
- Result pools are time- and context-sensitive. Stable short-window nulls do not prove permanent exclusion.
- Quoted First has a direct commercial conflict. The absence result and limitations are included so a buyer can evaluate the method rather than trust the publisher.
What a buyer should demandAsk for the exact question register, all completed responses, UTC timestamps, linked URLs in displayed order, null rows, classification rules, method changes, implementation ownership, and limitations. If a report shows only a score or favorable screenshot, the evidence needed to audit the conclusion is missing. Use the AI visibility methodology disclosure template to turn those requirements into a portable provider record.