We sell AI visibility. AI doesn't recommend us yet.
Week 0 baseline · 4 November 2026
We pointed our own tracker at trendlyai.com. We asked ChatGPT, Perplexity, Claude and Google AI Overviews the 20 questions a buyer asks before picking a tool like ours. None of the 80 answers named us. This is our starting line, and in four weeks we publish the next scan against it.
0 of 80
general buyer answers name TrendlyAI
20 questions × 4 engines. trendlyai.com cited 0 times.
18 of 20
branded answers mention us
But only 8 of 20 got the facts right.
Week 4
Next update early December 2026
Same questions, same engines, same scoring.
Scanned 4 November 2026 · 120 answers stored · US location, web search on · Method
A note from the founder
Why publish a zero?
I build a tool that tells you whether AI assistants recommend your brand. It felt wrong to sell that without running it on myself first. So I did, and the answer is no: not once in 80 tries.
Most case studies only show the after picture. I'd rather show you the before, with the raw numbers and the embarrassing bits, and then show you what changes. If you run the free check on a young site, your Week 0 will probably look a lot like this one.
Baseline · Set 1
Engine by engine: zero everywhere.
ChatGPT0 of 20
answers name TrendlyAI
- Our site cited
- 0 of 20
- Named most
- Profound (7)
gpt-4o-mini, web search
Perplexity0 of 20
answers name TrendlyAI
- Our site cited
- 0 of 20
- Named most
- Semrush (13)
sonar
Claude0 of 20
answers name TrendlyAI
- Our site cited
- 0 of 20
- Named most
- Otterly.AI (12)
claude-haiku-4-5, web search
Google AI Overviews0 of 20
answers name TrendlyAI
- Our site cited
- 0 of 20
- Named most
- Semrush (13)
- Organic top 10
- 0 of 20
Google US results
ChatGPT chose to search the web on 12 of the 20 questions. Google showed an AI Overview for all 20.
Who AI names instead
The tools AI names instead. We weren't one of them.
- Otterly.AI41/80
- Semrush39/80
- Profound38/80
- Peec AI33/80
- Ahrefs26/80
- AthenaHQ19/80
- SE Ranking18/80
- Scrunch AI17/80
- Rankscale13/80
- TrendlyAI (us)0/80
Set 2 · reported separately
Where we already show up
18 of 20
answers mention TrendlyAI. 14 of 20 cite our site.
- ChatGPT5/5
- Perplexity5/5
- Claude5/5
- AI Overviews3/5
Questions: "What is TrendlyAI?", "TrendlyAI review", "Is TrendlyAI legit?", "TrendlyAI vs Profound", "TrendlyAI pricing".
0 of 20
answers mention TrendlyAI, even for questions that describe our exact features.
Named most instead: Otterly.AI (8 of 20), Rankscale (5 of 20).
For example: “Is there an AI visibility tracker under $20 a month?” Our Solo plan is $19.
A mention is not a correct answer.
We read all 20 branded answers by hand against our live site. Only 8 got the facts right.
- Accurate 8
- Partly wrong 5
- Wrong 4
- About another company 2
- No answer 1
ChatGPTGot it rightTrendlyAI review
“TrendlyAI is an AI visibility platform designed to help brands monitor and enhance their presence across AI-driven search engines like ChatGPT, Perplexity, Claude, and Google AI Overviews.”
PerplexityGot it rightTrendlyAI vs Profound
“TrendlyAI is the budget-friendly, action-oriented option; Profound is the enterprise-grade analytics platform.”
Branded answers
What AI gets wrong about us
Google AI OverviewsGot it wrongWhat is TrendlyAI?
“Trendlyne is an all-in-one financial technology and stock market analytics platform designed for retail investors, research analysts, and portfolio advisors.”
What's true: Trendlyne is a different company, an Indian stock analytics app. The overview never mentioned us. The same thing happened on "TrendlyAI review".
ClaudeGot it wrongTrendlyAI review
“TrendlyAI is an artificial intelligence-driven content analysis platform that assists users in identifying trending topics, evaluating sentiment, and forecasting content performance”
What's true: That was our product before 2026. Today TrendlyAI tracks whether AI assistants recommend your brand. The old pages are gone, but a third-party listing still has the old text.
ChatGPTGot it wrongTrendlyAI vs Profound
“Primarily on market trends and analytics. It helps businesses track social media trends, market sentiments, and competitor analysis.”
What's true: ChatGPT answered from memory here, without searching, and described the old product.
ClaudeGot it wrongTrendlyAI pricing
“TrendlyAI offers two main plans: Solo at $19/month and Pro at $49/month, with a 7-day free trial available.”
What's true: There are three plans: Solo $19 a month (1 brand), Pro $49 (3 brands) and Agency $149 (10 client brands), all with weekly scans on 4 engines.
Sources AI trusts
The sites AI reads in our category, and whether we're on them.
| Domain | Answers citing it (of 80) | Cited by | Mentions TrendlyAI? |
|---|---|---|---|
| youtube.com | 15 | AI Overviews 15 | Not checked (videos) |
| sona.com | 10 | Perplexity 10 | No |
| rankability.com | 8 | Claude 5, Perplexity 3 | No |
| semrush.com | 7 | Perplexity 3, ChatGPT 2, Claude 1, AI Overviews 1 | No |
| visible.seranking.com | 7 | Claude 5, Perplexity 2 | No |
| accessibilitychecker.org | 6 | Perplexity 3, AI Overviews 3 | No |
| blog.hubspot.com | 6 | Perplexity 4, ChatGPT 1, Claude 1 | No |
| orchly.ai | 6 | ChatGPT 2, Perplexity 2, Claude 2 | No |
| reddit.com | 6 | AI Overviews 6 | Not checked (blocked) |
| tools8020.com | 6 | ChatGPT 4, Perplexity 2 | No |
Each engine leans on different sources: every YouTube and Reddit citation came from Google AI Overviews, every sona.com citation from Perplexity. Getting on one list does not help everywhere.
The plan
What we're changing, and what we expect to happen
- 1
Comparison and alternatives pages
ShippedWe published 31 comparison, alternatives and review pages between 29 October and 3 November, on top of earlier pages like TrendlyAI vs Otterly. So far AI cites them only when someone asks about TrendlyAI by name.
- Hypothesis
- Engines that search the web will start citing these pages for "alternatives" questions.
- We'll watch
- The 3 alternatives questions in the baseline (Profound, Peec AI, Otterly).
- 2
Getting onto the lists AI already cites
In progressWe are pitching the independent best-of articles in the sources table above, starting with the ones more than one engine uses.
- Hypothesis
- One or two inclusions on a list AI already trusts will move the number more than ten posts on our own blog.
- We'll watch
- Whether any of the 10 most cited domains mentions TrendlyAI.
- 3
Reviews from real users
PlannedWe have very few public reviews, and some third-party listings still describe our old product.
- Hypothesis
- Reviews from real users give AI a third-party source to trust, and answers to "TrendlyAI review" get more accurate.
- We'll watch
- Accuracy of the 5 branded questions.
- 4
Publishing original data
ShippedWe published a study on how often ChatGPT, Perplexity and Claude agree on the top brand. This case study is the second one.
- Hypothesis
- Original numbers get cited and linked by the articles AI reads.
- We'll watch
- Citations of blog.trendlyai.com on general questions.
- 5
Fixing the wrong facts
PlannedStructured Organization data that links our profiles to trendlyai.com, an About page and llms.txt that say plainly we are not Trendlyne and list current pricing, and updates to the third-party listings that still carry the old description.
- Hypothesis
- Clear, consistent facts in places AI reads will stop the Trendlyne mix-up and the old product description.
- We'll watch
- The 12 branded answers that were partly wrong, wrong, about another company or empty.
Method
How we measured it
- Date
- All 120 answers collected on 4 November 2026.
- Engines
- ChatGPT (gpt-4o-mini), Perplexity (sonar), Claude (claude-haiku-4-5), Google AI Overviews. US location, web search on.
- Set 1, baseline
- 20 general buyer questions × 4 engines = 80 answers. No question names TrendlyAI. 0 errors.
- Set 2, reported separately
- 5 branded + 5 niche questions × 4 engines = 40 answers. Never added to Set 1.
- Counted as a mention
- The answer text contains “TrendlyAI”, “Trendly AI” or trendlyai.com.
- Counted as a citation
- trendlyai.com or blog.trendlyai.com among the answer's cited URLs.
- Samples
- One answer per question per engine. No system prompt; the question was the whole message.
- No re-rolls
- We didn't edit, drop or re-run any question to change a number.
Limits
- One answer per question is a snapshot. AI answers change between runs, so the trend over several scans matters more.
- API answers can differ from the consumer apps (logged-in ChatGPT, Perplexity Pro, claude.ai).
- Brand names were extracted by a model and normalised, so small naming errors are possible.
- Many Google organic results came back as redirect links, so we matched on titles as well as domains.
See all 30 questions +
Set 1, general (20)
- What are the best AI visibility tools?
- Best AI search visibility tools 2026
- How can I track my brand in ChatGPT?
- What tools can monitor brand mentions in Perplexity?
- What are the best Profound alternatives?
- What are the best Peec AI alternatives?
- What are the best Otterly alternatives?
- What is a cheap AI visibility tracker for a small business?
- Best AI visibility tools for agencies
- Which tools offer white-label AI visibility reports?
- Is there a free AI visibility checker?
- How do I check if ChatGPT recommends my brand?
- Best GEO tools for SaaS companies
- What is the best AI search rank tracker?
- How can I improve my brand's visibility in AI answers?
- What are the best AEO tools?
- Best LLM brand monitoring software
- What is a good Google AI Overviews tracking tool?
- How can I track my brand's visibility in Claude?
- What is the best AI share of voice tool?
Set 2, branded (5)
- What is TrendlyAI?
- TrendlyAI review
- Is TrendlyAI legit?
- TrendlyAI vs Profound
- TrendlyAI pricing
Set 2, niche (5)
- What is the cheapest AI visibility tracker with white-label reports?
- Is there an AI visibility checker with a free trial and no credit card required?
- Which AI visibility tool tracks results by country and language?
- Is there an AI visibility tracker under $20 a month?
- Is there a tool that drafts fixes for AI visibility gaps?
How TrendlyAI scores visibility in general is on the methodology page.
Every scan, in order
Same 20 general questions on the same 4 engines each time. We add a card here after every scan.
Week 0 baseline · 4 November 2026
0 of 80
general answers name TrendlyAI
Week 4
? of 80
Results in early December 2026. If the number goes down, we publish that too.
What does your Week 0 look like?
Run the same kind of check on your own domain: 5 buyer questions on ChatGPT and Perplexity, a 0 to 100 score in about 30 seconds, no signup.
Run your free checkOr see a sample report. Paid plans track all 4 engines every week, with a 7-day trial and no card.
Questions about this case study
Why publish a low score?
Because a real starting point is more useful than a polished claim. It also shows exactly what our tool measures, on a brand we know inside out.
Why are the branded questions reported separately?
Because they are much easier. Almost any brand gets named when the question includes its name. Adding them to the 80 general answers would make the baseline look better than it is.
Will you cherry-pick the Week 4 results?
No. Week 4 uses the same 30 questions, the same 4 engines and the same scoring. We will not edit, drop or re-run a question to improve the number, and if it goes down we will publish that too.
Is this the same check customers get?
It uses the same engines and the same kind of buyer questions. Paid plans run tracked questions every week, so you see a trend instead of one snapshot. The free check runs 5 questions on ChatGPT and Perplexity.
Can I run this on my own brand?
Yes. The free check runs a short version on ChatGPT and Perplexity in about 30 seconds, with no signup.