A sample of the work

See the edit.
Inspect the evidence.

A useful handover connects what we found to something your team can change. Start with an illustrative page edit, then explore the dated record from our own audit.

Jump to the real audit record

A page edit,
in context.

Illustrative deliverablePage + FAQ draft

Buyer question

Which call-recording software fits B2B sales teams?

Current page wording

Software for modern teams.

Proposed page wording

Call-recording software for B2B sales teams.

FAQ draft

Which CRM integrations does the product support?

Source and next action

Start with your product and pricing pages.

Our own baseline.

CiteLantern was not organically named in the captured text.

We asked five AI engines the questions a SaaS buyer would ask about our own category. One run per question; limits below.

Captured 26 Oct and 5 Nov 2026

Founder: Nazli Ergin · Page last updated 11 November 2026

Answers naming CiteLantern
0
of 13 captured; one capture was partial
Answers captured
13
of 15 asked; Gemini returned errors on two questions
Run per question
1
A snapshot, not a visibility rate
Did the engine name CiteLantern?Three buyer questions, five engines, one run. Tap or select any cell or question row to inspect who got named.
QuestionChatGPTPerplexityGeminiClaudeGoogle AI Overviews
Q1 Which companies help B2B SaaS get recommended by ChatGPT and other AI search engines?
Q2 How do I find out why ChatGPT recommends my competitor instead of my software?
Q3 Best AI search visibility service for a small B2B SaaS company?
Q1 selected
  • ChatGPT: named Omniscient Digital, Omnius, Graphite +8 more
  • Perplexity: named Omniscient Digital, Quoleady, Arobis AI +2 more
  • Gemini: no answer captured (engine error)
  • Claude: named Omniscient Digital, Animalz, Graphite +26 more
  • Google AI Overviews: named DerivateX, Skale, DoodleWeb +12 more

Not namedNo answer (engine error)Dashed: partial capture

Names seen in the captured textCaptures naming each product, of 13. Not a market ranking.
  • Peec AI8
  • Profound7
  • Otterly.AI7
  • Omniscient Digital4
  • Scrunch AI4
  • AthenaHQ4
  • Semrush4
  • CiteLantern0

What this baseline shows.

The captured answers mixed software tools and service providers. They are not all direct commercial alternatives to an agency. This limits what we can infer from this category test.

In the text we captured, CiteLantern was not organically named. The answers included other products and sources such as review roundups, vendor-written lists and community threads. These observations do not tell us why an engine selected any name.

What we saw
The captured answers included established tools and service providers.
What to investigate
Compare the sources cited in the answers with our own public information.
Limit
One run per question cannot establish stable visibility or the effect of any change.

Why AI names a competitor, and what to check.

This is the question we put to ourselves: how do you find out why an AI engine recommends a competitor instead of your product? The four steps below are the order we work in. They are proposed steps, not measured results.

1. Read the answers
Ask the same buyer question in each engine and note every name and source. Here: 13 captured answers, none naming CiteLantern.
2. Sort the sources
Group what the answers draw on by type, such as review roundups, vendor-written lists, community threads and company sites. Compare that with what your own public pages say.
3. Find the gaps on your pages
Check that your product and pricing pages say who the product is for, what it does and which buyer question it answers, in plain words. The page edit above shows one example.
4. Edit, then re-run
Your team publishes the edits. We re-run the same questions on a later date and report what changed. One run is a snapshot, so we do not claim a cause.

How we ran it.

Date and language
26 October 2026, English. ChatGPT and Perplexity question 3 re-run on 5 November 2026.
Questions
The three above, identical in every engine, one per fresh conversation, no follow-ups.
Engines
ChatGPT (temporary chat), Perplexity, Gemini, Claude (private chat) and Google AI Overviews.
Captured material
Available on request; one Perplexity response (question 2) is incomplete.
Limits we want you to know about
  • Answers can vary by account context and time. ChatGPT and Perplexity question 3 were re-run on 5 November.
  • In one ChatGPT answer the tool pre-filled citelantern.com as the user's own site, from that account's context; we did not count it as an engine naming us.
  • Gemini failed on two questions after retries.
  • The Perplexity capture for question 2 was partial. A missing mention in partial text does not establish what the full answer said.
  • One run is a snapshot, not a ranking.

Want this for your product?

Send your website, plus the questions your buyers ask if you have them. You get a short written check by email, and a fixed quote if there is a gap worth closing.