PaperHorse tracks how ChatGPT, Google AI, Claude, Gemini, and Perplexity answer the questions people ask when looking for a business like yours. See where you’re recommended, who shows up instead, and what to do next.
Already have an account?
How it works
Enter your website URL. We review your site and build a profile of your brand, what you offer, and the customers you serve.
We ask ChatGPT, Google AI Mode & Overviews, Claude, Gemini and Perplexity the questions people are asking when looking for a business like yours.
Your report shows how often your brand is mentioned, which competitors appear, and how the answers differ by AI engine.
We identify the best opportunities to earn more AI recommendations, create the content to support them, and measure how your visibility changes over time.
Inside the product
These screenshots show what you’ll see as a customer. They use a fictional company called GlassWhistle.
01 · The visibility report
Your latest report shows how often your brand appears in answers to non-branded buying questions. It also shows how often you are named first, your average position when you appear, and which AI engines mention you most.
Organic discovery · non-branded questions
Measured only on questions that don’t name GlassWhistle. This shows whether AI surfaces you on its own.
How often you’re the top pick
8%
First named in 9 of 120 answers
Average position when recommended
2.4
Across 29 mentions. Lower is better.
02 · Brand recall & sentiment
Branded questions name your business directly. They measure whether AI mentions your brand in its answer, how accurately it describes you, and whether the response is positive, neutral, or negative.
Brand recall · branded questions
These questions already name GlassWhistle, so a mention is a given. This reflects brand recall, not whether AI surfaces you unaided.
Ranked first
65%
Top of the answer in 65% of branded questions
Positive sentiment
51%
Negative 5% · Neutral 20%
03 · Trends
Each point shows the percentage of tracked answers that named your brand in that report. See your current mention rate, the percentage-point change, and whether the rate is moving up or down over time.
Mention rate
24%
04 · Competitor report
The chart ranks your brand and competitors by share of voice, showing the percentage of brand mentions each received across your tracked questions. The arrows show how each brand's rank changed since the previous report.
Change is against your previous report, Sep 1.
05 · See the answers
Open any tracked question to compare the stored answers from ChatGPT, Google AI Mode & Overviews, Claude, Gemini and Perplexity. See whether your brand appeared, where it was positioned, how the mention was classified, and the complete answer text.
“For EU residency, GlassWhistle documents in-region processing and offers a Frankfurt deployment with customer-managed keys…”
View“Other options include GlassWhistle, though its EU regions are newer than the incumbents…”
View06 · Prioritized action plan
Your action plan puts the highest-priority recommendation at the top and explains why it matters. Each task shows its difficulty, estimated impact, and whether it involves your own content or an outside source.
Publish a data residency and sub-processor page
Why it matters. Data residency is cited in 19 of 30 answers, always from a competitor’s trust centre. You have no page stating yours.
07 · Content generation
The screenshot shows a website draft, but PaperHorse creates more than owned content. Depending on the opportunity, it can draft website pages, social posts, Reddit and other forum responses, influencer emails and DMs, and editorial pitches or contributed content for authoritative third-party sites.
Publish a guide on Data residency for enterprise CDPs
Data Residency for Enterprise CDPs: Regions, Keys, and Sub-processors
Procurement asks two questions: where does the data physically sit, and who else can touch it. Here is every region we process in, the keys you control, and our full sub-processor list.
Regions and in-region processing
EU workloads stay in Frankfurt end to end, including backups and search indexes
Who it’s for
You spent years on Google rankings. Now your customers are asking ChatGPT instead.
Your boss is going to ask about AI search visibility. Now you'll have the answer.
Your next opportunity might require new website content, a Reddit response, an influencer pitch, or a contributed article. We'll help you write it.
Use Agency Mode to manage separate clients from one centralized dashboard, with visibility reports and prioritized recommendations organized for each client.
Pricing
Choose based on reporting frequency, how many customer questions you want to measure, and whether you manage one brand, several brands, or separate clients.
Starter
Essential weekly AI visibility tracking.
Cancel anytime before day 8.
Standard
For brands ready to turn AI visibility insights into action.
Cancel anytime before day 8.
Pro
For brands making AI visibility an ongoing growth priority.
Cancel anytime before day 8.
Enterprise
For teams managing larger brands and audiences.
We set the brands up with you.
Agency
Running AI search reports for multiple client brands? We offer tailored volume pricing for agencies managing a book of business.
Tell us how many brands you manage and we’ll come back with a number.
Agency program →FAQ
Have another question? Ask us directly. contact us and a person answers.
Every plan checks ChatGPT, Google AI Overviews and Google AI Mode. Higher plans add Perplexity, Claude and Gemini, up to six engines in total. Your prompt set runs against each engine separately, so you can compare where your brand appears and where it is missing.
Your visibility score is the percentage of non-branded answers that mention your brand. These are questions where a customer describes what they need without naming you. For example, if your brand appears in 34 of 120 non-branded answers, your visibility score is 28%.
Yes. We list the competitors that appear in your tracked answers and rank them by how often they are mentioned. You can open an answer to see the exact context in which each competitor appeared.
Reports run weekly on Starter, twice a week on Standard, and every other day on Pro and Enterprise. We store every answer so you can inspect the evidence behind each report and compare results over time.
Yes. A payment card is required to start a 7-day Standard or Pro trial. Your card is not charged during the trial. Cancel before the trial ends to avoid the first subscription charge.
Yes. You can change or cancel your plan from your account. Cancellation takes effect at the end of your current paid billing period, so you keep access until that date. Once the subscription ends, your dashboard is no longer available.
Yes. Our agency program lets you manage multiple client brands under a single account with volume pricing. Contact us and we will help configure the right setup for your client roster.
PaperHorse turns the evidence it finds into a prioritized action plan designed to help your brand appear more often in AI answers. Every item explains what to do, why it matters, and which customer questions it may affect.
Actions can include improving a website page, publishing new content, responding to a relevant Reddit or social discussion, or pursuing coverage on a third-party site. Pro and Enterprise also include content drafts when writing is needed.
Your first visibility score is ready within minutes. After you make improvements, PaperHorse measures again weekly on Starter, twice a week on Standard, and every other day on Pro and Enterprise. Some movement may appear in the next measurement. Other changes take longer because AI platforms decide when to recrawl pages and update their sources. PaperHorse tracks the results over time so you can distinguish a single changed answer from a sustained improvement.