3DPano for Business
Interactive sample · read only

The Hotel Chelsea – real sample audit

Navigate the same overview, visibility, competition, sources, visual, graph, evidence and implementation structure used by the customer portal.

Vispertise™ audit demo

The Hotel Chelsea

New York · Hotel · Sep 17, 2026, 04:09

Deep analysis

Visibility

78.7

How often the place is found when the question does not name it. Relevant audiences and AI providers have equal weight, as in Overview.

/ 100 points

Mention visibility (MV)

20% weight · 15.7 points in the overall score
90

How strongly the answers recommend the place when it is mentioned.

/ 100 points

Recommendation strength (RS)

15% weight · 13.5 points in the overall score
96

Accuracy = correct / (correct + incorrect). Verifiability = (correct + incorrect) / all eligible claims. Unverifiable claims are not counted as incorrect. Accuracy is relative to the available references, not independent fact certification. Percentages are equally weighted by persona, then provider; the summed counts separately describe the evidence base. Repeated claims are not necessarily distinct facts.

/ 100 points

Factual accuracy (ER)

15% weight · 14.4 points in the overall score

The hotel appeared in 88% of the measured decision situations. Couples, families, accessibility needs and events generally surfaced relevant offers. Extended business stays were the main gap: only a quarter of answers requesting residential suites and a 24-hour gym mentioned the hotel.

Room-size comparisons were usually present but recommendation certainty was weak. Conflicting capacities, missing venue dimensions and room-block limits, and unverified doorway and shower clearances constrain decisions. Connect the existing offer to explicit, comparable practical facts.

Same score as in Overview →
Partial measurement

Factual accuracy and verifiability

This checks how accurately AI answers describe your organisation. We compare claims with references retained during the audit: accuracy measures how many verifiable claims were correct, while verifiability measures how much of the full claim set we could assess. Read the two together: high accuracy can still leave many claims unresolved. Clearer, verifiable sources may be needed for those claims.

Weighted proportion · 0–100%

Factual accuracy96.0%
Factual accuracy: 96.0%
Verifiability46.8%
Verifiability: 46.8%

Source collection is partial. The score describes retained, verified evidence and does not establish complete website coverage. Unverifiable claims are reported separately.

Measurement basis

What is the score based on?

We show separately whether every planned answer was retained and how many received a verified evaluation.

Measurement completeness100%

160 / 160 planned answers were retained.

Evaluation coverage100%

160 / 160 answers were included in evaluation.

Excluded answers are not scored as zero; they are removed from the evaluation sample. The fixed pillar weights do not change.

Sample and limitations

40 questions · 160 answers · repetitions: 1. This is a limited question sample and one audit observation; it does not establish stability over time.

Claims: 124 correct, 5 incorrect, 136 unverifiable · Answers: 20 · Questions: 5.

Measurable persona–provider groups: 15/16.

Accuracy = correct / (correct + incorrect). Verifiability = (correct + incorrect) / all eligible claims. Unverifiable claims are not counted as incorrect.

Accuracy is relative to the available references, not independent fact certification. Percentages are equally weighted by persona, then provider; the summed counts separately describe the evidence base. Repeated claims are not necessarily distinct facts.

Visibility

Audience visibility at a glance

This view includes every tested audience and question: evidence-selected relevant audiences plus the fixed standard benchmark panel. The primary GEO score remains relevant-scope.

Filter this view
Benchmark LLMs

58 of 72 selected answers mention the place (81%).

Anonymous discovery answers only. Changes with the filters; the stored audit result does not.

5Audiences tested
18Tested questions
14Question families

Question-family results

11 Strong 2 Partial 1 Needs improvement

Audience ranking

Open an audience to see the combined HU + EN question-family evidence.

01
Emily Carter 4 questions · 3 families · Relevant audience
94%15/16
Strong

Synthetic persona: 34 years old · International leisure traveler planning a couples' city break · Chelsea · EN · Visiting New York from the United Kingdom for a short couples' break.

Choose a distinctive Chelsea stay that balances room comfort, neighborhood access and wellness.

EN15/1694% Open all audience evidence →
02
Danielle Brooks 4 questions · 3 families · Relevant audience
88%14/16
Strong

Synthetic persona: 42 years old · Parent organizing a family trip and wedding-guest accommodation · Chelsea · EN · Organizing rooms for family members attending a wedding in New York.

Keep relatives together in suitable multi-bedroom layouts while simplifying wedding logistics.

EN14/1688% Open all audience evidence →
03
Samantha Ortiz 4 questions · 3 families · Relevant audience
88%14/16
Strong

Synthetic persona: 39 years old · Corporate and wedding event producer · Chelsea · EN · Sourcing New York spaces for corporate events, private dinners and wedding functions.

Select spaces whose format, capacity and private-event features match each client brief.

EN14/1688% Open all audience evidence →
04
Robert Ellis 3 questions · 2 families · Relevant audience
67%8/12
Strong

Synthetic persona: 58 years old · Older adult traveler with mobility and hearing access requirements · Chelsea · EN · Assessing an accessible New York hotel stay before making a reservation.

Confirm that both the guest room and public areas meet specific mobility and communication needs.

EN8/1267% Open all audience evidence →
05
Marcus Reed 3 questions · 3 families · Relevant audience
58%7/12
Mixed

Synthetic persona: 46 years old · Business consultant planning an extended work stay · Chelsea · EN · Planning a multi-night work assignment in Chelsea.

Find a work-friendly hotel that supports a longer stay without sacrificing exercise or residential amenities.

EN7/1258% Open all audience evidence →
Decision guide

What the result means for the business

These are not more scores: they show which audience performs weakest and best, and where improvement is most attainable.

Weakest-performing audience

Marcus Reed

7/12 58% Mention rate

This is the weakest result among the measured audiences, so its decision situations are the clearest place to investigate and improve first.

2 Strong 0 Partial 1 Needs improvement
Open evidence →
Best-performing audience

Emily Carter

15/16 94% Mention rate

This is the strongest result among the measured audiences and shows which positioning to preserve and reuse consistently elsewhere.

3 Strong 0 Partial 0 Needs improvement
Open evidence →

Cookie settings

Required for secure sign-in and essential service functions. These are always active.

Allow our own visitor measurement and Google Analytics to help us understand use of the website and customer portal.

Cookie notice