Your business is being considered.
AI suggests your business as an option for the tested customer questions. The next step is to identify where it is recommended and where other providers appear.
3 Prompts · 3 repetitions · 144 planned answers
Illustrative sample report. All businesses, answers and results are for demonstration only.
This shows your business visibility for the customer questions you selected. AI recommends it as a suitable choice in 64 of 144 clearly assessed answers.
3 customer questions without the business name · Switzerland, Germany, France, Italy · German, French, Italian, English · 3 planned repetitions per combination.
The number of AI answers described above that mention your business.
The number of answers that recommend your business as a suitable choice for the customer’s needs.
The number of answers that include a source link to your website.
AI suggests your business as an option for the tested customer questions. The next step is to identify where it is recommended and where other providers appear.
For these questions, answers name other providers without mentioning your business. First, check how relevant these needs are to your offer.
All selected countries · All selected languages · A snapshot of the selected prompts. Not a measurement of all ChatGPT users.
API-based sample, not a direct measurement of the ChatGPT app.
Your business appears as a possible choice. Original answers show which services and qualities AI associates with your name.
Missing mentions mean missing visibility for these questions, not poor business quality. There is no universal industry benchmark here.These businesses appear in answers to the same customer questions. We call them competitors here because they appear as alternatives for the tested need. This is not a complete list of your actual competitors.
Counts explicit suggestions of a business as a suitable option. A name alone does not count.
1 competitors rank ahead of your company and 3 behind. 0 are tied or cannot be compared.
Common basis for comparison: 144 answers. 0 answers that cannot be jointly evaluated are excluded here. Values may therefore differ from the overview.
The list may include platforms and directories alongside direct competitors. Their role is labelled where the stored evidence supports it. A mention alone does not establish direct competition.
Evidence for the selected metric:
Evidence for the selected metric:
Evidence for the selected metric:
Evidence for the selected metric:
Evidence for the selected metric:
Rank shows your position within this selection. Expand a competitor to see mentions, recommendations and supporting answers. This is not an industry rating.
Your brand’s share of recorded business mentions: 23 % (96 / 416), based on 144 answers with business analysis. Not market share.
Compare how often your business is mentioned, recommended and linked as a source across your selected markets and languages. The filters above apply here too.
Länder: The recommendation rate is the same across the evaluated groups.
Sprachen: The gap between the highest and lowest recommendation rate is 44 percentage points.
| Country | German | French | Italian | English |
|---|---|---|---|---|
| Switzerland | 67 %6 / 9 | 44 %4 / 9 | 22 %2 / 9 | 44 %4 / 9 |
| Germany | 67 %6 / 9 | 44 %4 / 9 | 22 %2 / 9 | 44 %4 / 9 |
| France | 67 %6 / 9 | 44 %4 / 9 | 22 %2 / 9 | 44 %4 / 9 |
| Italy | 67 %6 / 9 | 44 %4 / 9 | 22 %2 / 9 | 44 %4 / 9 |
Country bars combine the selected languages; language bars combine the selected countries. For a more focused country comparison, select a single language above. To compare languages, select a single country.
Where possible, compare the same questions and number of runs. Different questions, language mixes or missing answers may explain differences. These are observed rates within this audit, not statistically established country or language effects. Unclear and pending answers do not count as No. Prompts that name the brand are excluded.
A prompt is a customer question to AI. See where your business appears, who is named instead and what to investigate.
Most frequently recommended other providers: Example Provider A (48 / 48)
Open a run to see the full AI answer with all recorded providers and sources.
Most frequently recommended other providers: Example Provider A (36 / 48)
Open a run to see the full AI answer with all recorded providers and sources.
Most frequently recommended other providers: Example Provider A (36 / 48)
Open a run to see the full AI answer with all recorded providers and sources.
Start here: what services do customers seek, which competitors appear and what evidence does AI use?
Named instead of your business: Example Provider A (28 answers), Example Provider B (28 answers), Example Provider C (12 answers) and 1 other competitors.
Your next step: Open the answers and compare the services described with your offering and review the 2 linked source pages.
View competitors, answers and sources for this question →Named instead of your business: Example Provider A (8 answers), Example Provider C (8 answers), Example Provider D (8 answers).
Your next step: Open the answers and compare the services described with your offering and review the 2 linked source pages.
View competitors, answers and sources for this question →Ordered by the number of answers missing your business. This does not estimate demand or revenue potential. It concerns mentions, not list positions.
These filters apply to the prompt table. CSV export includes all answers for the selected countries and languages, including branded prompts.
| Prompt | Results | Most recommended competitor |
|---|---|---|
| I need solar panels for my house. Which providers in my region would you recommend?Location · Without brand name | 44 recommended4 mentioned only Mention: 48 of 48 answers that can be assessedOut of 48 planned answers | Example Provider A · 48 / 48 |
| I want solar panels and a heat pump from one provider. Which companies handle planning and installation?Service / product · Without brand name | 20 recommended20 mentioned only8 not mentioned Mention: 40 of 48 answers that can be assessedOut of 48 planned answers | Example Provider A · 36 / 48 |
| I am looking for a solar installer with transparent quotes and reliable service. Which companies would you recommend?Purchase decision · Without brand name | 8 mentioned only40 not mentioned Mention: 8 of 48 answers that can be assessedOut of 48 planned answers | Example Provider A · 36 / 48 |
Green: recommended · Yellow: mentioned only · Red: not mentioned · Grey: pending / not evaluable. Competitors are compared using documented recommendations in the same selection; ties remain visible.
| Prompt | German | French | Italian | English |
|---|---|---|---|---|
| 1. I need solar panels for my house. Which providers in my region would you recommend? | 12 / 12 | 12 / 12 | 12 / 12 | 12 / 12 |
| 2. I want solar panels and a heat pump from one provider. Which companies handle planning and installation? | 12 / 12 | 12 / 12 | 4 / 12 | 12 / 12 |
| 3. I am looking for a solar installer with transparent quotes and reliable service. Which companies would you recommend? | 8 / 12 | 0 / 12 | 0 / 12 | 0 / 12 |
Example Company AG: 44 of 48 Empfehlungen. Example Provider A: 48 of 48.
Next check: Check statements about your business for accuracy and keep the services and sources up to date.
Example Company AG: 20 of 48 Empfehlungen. Example Provider A: 36 of 48.
Next check: Compare answers with and without your brand: which requirements and businesses differ?
Example Company AG: 0 of 48 Empfehlungen. Example Provider A: 36 of 48.
Next check: Compare answers with and without your brand: which requirements and businesses differ?
Citations and negative mentions are also shown separately for each answer.
All recorded sources for this selection: pages, frequency, business attribution and links to the related answers.
All recorded sources for the selected countries and languages, including questions naming your company. Cited means the AI linked to the page in an answer. Other search source means the API returned it as a research source, but it was not recorded as a citation in an answer.
Each page or domain is counted once per answer. A source link does not prove it caused a recommendation. For 144 answers, no separate search-source list was saved; only recorded citations can be shown.
Attribution: No verified link to your company or a recorded competitor. This does not prove independence.
Example Company AG is named in 96 of 144 answers citing this source and absent in 48.
What you can do next: Review the linked pages and their role. Frequent citation alone does not establish quality or trustworthiness.
Page · content type unconfirmed Cited in 144 answers
Attribution: No verified link to your company or a recorded competitor. This does not prove independence.
Example Company AG is named in 48 of 80 answers citing this source and absent in 32.
What you can do next: Review the linked pages and their role. Frequent citation alone does not establish quality or trustworthiness.
Likely directory Cited in 80 answers
Attribution: Domain from your company profile.
Example Company AG is named in 32 of 32 answers citing this source and absent in 0.
What you can do next: Check that your cited pages describe your services and company clearly and accurately.
Likely service / product page Cited in 32 answers
Page types are inferred from URLs and recorded titles, not a content review. Full text, publishers, freshness and mentions on third-party pages have not been separately checked. Unrecorded internal model sources are unavailable.
How well does your website load, and how easily can AI agents understand and use it?
Illustrative results for Beispielunternehmen AG — not an actual Google measurement.
How quickly the page loads and responds to input. A higher score means better measured performance.
72/ 100How accessible the page is for people with disabilities. Automated tests do not replace manual checks.
88/ 100Whether the page follows key technical quality and security practices, such as secure connections and functioning browser features.
96/ 100Whether basic technical requirements for search engines are met. This score is not a Google ranking.
92/ 100How well AI agents can understand and operate the page. This experimental test does not measure ChatGPT recommendations.
1 / 2 checks passed
Tests technical usability for AI agents. This is not a ranking or a guarantee of ChatGPT recommendations. Inapplicable and untestable checks do not count as failures.
In this example, the contact button lacks a clear accessible name, making its purpose harder for an AI agent to identify.
Next step: Give controls clear accessible names and correct HTML roles. Inspect the accessibility tree and listed elements with a keyboard and screen reader.
In this example, controls stay in position while the page loads.
This optional technology is not used on the example page. This does not count as a failure.
This optional technology is not used on the example page. This does not count as a failure.
This optional technology is not used on the example page. This does not count as a failure.
This optional technology is not used on the example page. This does not count as a failure.
This optional technology is not used on the example page. This does not count as a failure.
The example hero image is too large. A smaller compressed image can speed up visible page loading.
Next step: Fix the reported loading problems, optimise affected images and scripts, then retest the same URL on mobile.
Light text on a light background is hard to read in this example. Darker text improves readability.
Next step: Correct the semantics of the listed elements, then manually test with keyboard and screen reader.
A mobile lab measurement of one page, not a complete website scan. Technical quality does not measure AI recommendations. Automated tests do not establish full accessibility.
Prioritised by observed gaps. Suggestions to investigate, not guarantees of increased visibility.
Observed in this audit: 40 answers mention competitors without naming your company. Your mentions: 8/48. Recommendations of your company: 0/48.
Companies appearing instead: Example Provider A, Example Provider B, Example Provider C, Example Provider D
These observations identify checks to make, not confirmed defects on your website. A later change in results does not prove that an edit caused it.
Observed in this audit: 8 answers mention competitors without naming your company. Your mentions: 40/48. Recommendations of your company: 20/48.
Companies appearing instead: Example Provider A, Example Provider C, Example Provider D
These observations identify checks to make, not confirmed defects on your website. A later change in results does not prove that an edit caused it.
These findings complement the content actions. They improve the technical foundations, without guaranteeing AI recommendations.
The example hero image is too large. A smaller compressed image can speed up visible page loading.
Next step: Fix the reported loading problems, optimise affected images and scripts, then retest the same URL on mobile.
Go to technical checks →Light text on a light background is hard to read in this example. Darker text improves readability.
Next step: Correct the semantics of the listed elements, then manually test with keyboard and screen reader.
Go to technical checks →In this example, the contact button lacks a clear accessible name, making its purpose harder for an AI agent to identify.
Next step: Give controls clear accessible names and correct HTML roles. Inspect the accessibility tree and listed elements with a keyboard and screen reader.
Go to technical checks →Report analysis v3: mentions, recommendations and citations are evaluated independently. Unknown is not zero. Mentions use the name and verified aliases in the answer text; URLs alone do not count. Older audits did not measure negative mentions separately.
Modell: fiktiv · Messprotokoll: Fiktive Demonstration. Country is an approximate API location setting. The instruction requests suitable options, sources and a short answer, which influences open questions. Translations may alter nuances.
The bootstrap interval uses repetitions within the selected questions, not a representative sample of all user requests. Small samples can produce misleadingly narrow intervals. Sources do not prove causality. Original answers remain accessible and unchanged.