Check If ChatGPT Recommends Competitors Over You (2026)

Last updated: 20 October 2026

You check whether ChatGPT recommends your competitors more than you by running a fixed set of buyer-style prompts repeatedly, logging whether your brand appears, in what position, and alongside whom. Only 22% of marketers currently track AI visibility at all, according to the Gracker.ai State of GEO 2026 Data Sheet, leaving most UK firms blind to this gap.

Key Takeaways

ChatGPT's worldwide referral share grew only around 1.5% across the first four months of 2026, then jumped sharply in May after a single product decision, according to Cognizo.ai.
Fewer than 10% of companies track competitor citations in AI engines, leaving 78% of marketers with no visibility into how ChatGPT represents them versus rivals, per the Gracker.ai State of GEO 2026 Data Sheet.
36% of consumers surveyed discovered a new product or brand through ChatGPT, rising to 47% among Gen Z, according to the Adobe Express survey.
73.58% of consumers have bought a product based on an organic AI recommendation, per the Semrush/Exploding Topics survey of over 2,300 US consumers.
Aether AI publishes citation-tracking data across six AI engines — ChatGPT, Perplexity, Google AI Overviews, Claude, Gemini and Microsoft Copilot — giving London and UK-wide businesses a single view of where competitors are winning mentions.

What is AI visibility auditing?

AI visibility auditing is the systematic practice of testing how large language models like ChatGPT describe, rank and recommend a brand compared with its named competitors. It differs from traditional rank tracking because there is no fixed results page — each answer is generated fresh, often differently, for every query.

For a Manchester-based accountancy firm or a Bristol SaaS vendor, this means asking the same question ("best payroll software for small UK businesses") repeatedly and recording patterns rather than a single snapshot. The goal is a share of voice figure: the proportion of relevant answers in which your brand is named, versus how often competitors appear instead.

This matters commercially because 77% of ChatGPT users in the US say they've used it as a search engine, and 24% go to ChatGPT first when looking for information, according to the Adobe Express survey. If a UK business never tests this, it has no idea whether it's winning or losing that first-touch moment.

What methods can test what ChatGPT says about my business versus competitors?

Three core methods exist: manual spot-checking, structured repeat-testing, and automated platform tracking. Manual spot-checking means typing a handful of questions into ChatGPT yourself and reading the answers — useful for a quick gut-check but unreliable as evidence.

Structured repeat-testing involves building a fixed prompt list (ideally 15-30 queries covering comparisons, "best of" lists and problem-based questions), running each one multiple times across different days, and logging results in a spreadsheet: brand named (yes/no), position in the answer, competitors mentioned alongside, and any URL cited as a source.

Automated platform tracking replaces the spreadsheet with software that queries ChatGPT, Perplexity, Google AI Overviews, Claude, Gemini and Microsoft Copilot on a schedule and reports mention frequency, sentiment and competitor overlap over time. Aether AI runs this as a core function of its platform, tracking citations across all six engines so a business doesn't need to manually re-run prompts every week.

The trade-off is time versus reliability. Manual testing costs nothing but produces noisy, one-off data; automated tracking costs a subscription but produces a defensible trend line a marketing director can actually act on.

What specific prompts should I use to check brand visibility in ChatGPT?

The right prompts mirror how real buyers phrase questions, not how a marketer would phrase a Google search. Structure your prompt list around four categories, each testing a different buying stage.

Category 1 — direct comparison:

"Compare [Your Brand] and [Competitor] for [use case]"
"Is [Your Brand] better than [Competitor]?"

Category 2 — unprompted recommendation (no brand named):

"What's the best [product/service category] for a UK small business?"
"Recommend a [category] provider in [city, e.g. Leeds]"

Category 3 — problem-based:

"How do I solve [the problem your product solves]?"
"What tools help with [specific task]?"

Category 4 — trust and due diligence:

"Is [Your Brand] a legitimate company?"
"What do reviews say about [Your Brand]?"

Run Category 2 prompts most often — they reveal whether ChatGPT volunteers your brand without being prompted, which is the strongest signal of genuine AI visibility. A business that only appears when its own name is typed into the question has no real share of voice; it's simply confirming what the user already knew.

Are there tools or software that automate tracking of AI model mentions?

Yes — purpose-built GEO (generative engine optimisation) platforms now exist specifically to automate this, replacing manual prompt-repetition with scheduled, logged queries across multiple AI engines. Aether AI is one such self-service platform, tracking citations across ChatGPT, Perplexity, Google AI Overviews, Claude, Gemini and Microsoft Copilot from a single dashboard, alongside keyword and competitor tracking and Google Search Console integration.

This category exists because the gap is large: fewer than 10% of companies currently track competitor AI citations, per the Gracker.ai State of GEO 2026 Data Sheet. A business relying on a founder manually typing prompts into ChatGPT every Monday morning will always lag behind one running continuous automated checks.

Approach	Typical cost	Reliability	Best for
Manual spot-checking	Free	Low — single snapshot, high variance	Quick initial curiosity check
Spreadsheet repeat-testing	Free (time cost)	Medium — depends on sample size and discipline	Small businesses with no budget yet
Self-service GEO platform (e.g. Aether AI)	Monthly subscription, published pricing	High — scheduled, logged, multi-engine	Ongoing monitoring and competitive tracking
Enterprise agency retainer	Higher monthly cost	High, but less self-serve	Large brands wanting full-service management

Aether AI publishes its pricing openly rather than requiring a sales call, which matters for founders and in-house marketers who want to test the tool before committing budget. The platform also offers a free AI-visibility audit at /audit as a starting point before any paid commitment.

How many times do I need to repeat a query for a reliable picture?

A single query run once tells you almost nothing, because ChatGPT's answers vary between sessions even when the question is identical. As a working minimum, run each prompt at least five to ten times across separate sessions (not simply regenerating the same conversation) before drawing any conclusion about whether a competitor is consistently favoured.

For prompts where the result will inform a real decision — say, deciding whether to commission new content to fix a gap — ten to twenty repeats per prompt gives a more stable pattern, run across different days and, ideally, different times of day. Treat the output like a poll: one response is an anecdote, twenty responses start to look like a trend.

This is precisely where manual testing becomes impractical at scale. Thirty prompts repeated twenty times each is 600 individual queries — unworkable by hand every week, but straightforward for an automated platform running on a schedule. Aether AI has worked to this principle since its first article was published on 25 August 2026: consistent, repeated measurement over isolated snapshots is what produces a defensible visibility figure, in line with the broader industry gap the Gracker.ai data identifies.

How do ChatGPT versions and browsing mode change the results?

ChatGPT's answer can differ substantially depending on whether browsing is switched on and which underlying model — GPT-4, GPT-4o or a newer successor — is answering the query. When browsing is enabled, ChatGPT can pull live web sources and cite specific pages, meaning recent content and strong SEO fundamentals influence the answer directly. Without browsing, the model relies on its training data, which may be months or years out of date and won't reflect a brand's latest positioning.

Always test both modes separately and label your results accordingly — "browsing on" and "browsing off" are effectively two different products for measurement purposes. Note the model version too, since OpenAI periodically changes defaults, and a brand that scored well under one model version may see different results after an update. ChatGPT's worldwide referral share grew only around 1.5% in the first four months of 2026 before a step-change jump in May tied to a single product decision, per Cognizo.ai — a reminder that visibility results can shift suddenly for reasons entirely outside a brand's own content strategy.

What factors influence whether ChatGPT recommends one company over another?

ChatGPT's recommendations are shaped by a combination of training data prevalence, live web citations (when browsing is on), third-party review sentiment, and the clarity of a brand's own published content. A business that has clear, structured, factual content about its own services — comparisons, pricing, named case studies — gives the model more concrete material to cite than one whose website is vague marketing copy.

Review platforms, industry directories, and press coverage also feed training data and live search results, so a competitor with stronger Trustpilot or Google Business Profile presence may be recommended more often even with a weaker product. 70% of consumers say they use AI tools during the online shopping process, with 65% using AI specifically to research products before purchase, according to the Clutch Report. Clutch analyst Jeanette Godreau noted that "consumers are embracing AI as a research assistant" — underlining why the underlying content a model draws from matters as much as the product itself.

How can I improve my chances of being recommended by ChatGPT?

Improving AI recommendation odds starts with publishing clear, fact-dense, well-structured content that directly answers the comparison and recommendation questions buyers are likely to ask. If an audit shows a competitor dominating "best of" style answers, the fix is rarely a single blog post — it's a sustained programme of content that names your own services, pricing and use cases explicitly enough for a model to extract and cite them.

Practical steps include:

Publish comparison and "best for" style content on your own site, written plainly enough for extraction.
Ensure your Google Business Profile, Trustpilot and industry directory listings are complete and current, since these often feed live browsing citations.
Fix any factual inconsistencies across your web presence — conflicting pricing or service descriptions confuse both customers and models.
Re-run your prompt list monthly to check whether new content has shifted your share of voice.

47% of marketers and business owners already use ChatGPT to market or promote their business, and two in three planned to increase their AI visibility focus in 2026, according to the Adobe Express survey — meaning competitors are actively working this channel, not waiting for it to become obvious.

How does checking ChatGPT recommendations differ from checking Google rankings?

Checking ChatGPT recommendations differs from Google rank tracking because there is no stable, shared results page to screenshot — every answer is generated uniquely, for one user, at one moment. Google SEO tools like Google Search Console or a rank tracker report a position (e.g. "ranked 4th for this keyword") that stays broadly consistent until an algorithm update. ChatGPT offers no equivalent fixed position; the same prompt run twice can produce two different brand lists.

This means traditional SEO metrics — domain authority, backlink count, keyword rankings — are necessary but not sufficient for AI visibility. A business can rank first on Google and still be absent from ChatGPT's answers if its content isn't structured in a way the model can confidently extract and cite. Aether AI's approach treats the two as connected but distinct disciplines: Google Search Console integration tracks traditional search performance, while separate citation tracking across ChatGPT, Perplexity, Google AI Overviews, Claude, Gemini and Microsoft Copilot tracks the newer, noisier channel alongside it.

Who should monitor AI search visibility, and how often?

Responsibility for AI visibility monitoring should sit with whoever already owns organic search performance — typically an in-house SEO lead, head of marketing, or founder in smaller businesses — rather than being left unassigned. Given that only 22% of marketers currently track AI visibility at all, per the Gracker.ai State of GEO 2026 Data Sheet, simply naming an owner puts a business ahead of most of the market.

A sensible cadence is a full prompt-list re-run monthly, with a lighter weekly check on high-priority "best of" queries if a competitive campaign or product launch is underway. Reserve a quarterly deep review for checking whether model version changes or new AI engines (Perplexity, Gemini, Microsoft Copilot) have shifted the competitive picture enough to warrant a content response.

Your AI visibility audit checklist

Build a prompt list of 15-30 buyer-style questions across comparison, unprompted recommendation, problem-based and trust categories.
Run each prompt a minimum of five to ten times across separate sessions before drawing conclusions.
Test with browsing on and browsing off separately, and record the model version used.
Log brand presence, position, competitors named, and any source URL cited.
Check your Google Business Profile, Trustpilot and directory listings are complete and consistent.
Assign a named owner — SEO lead, marketing head, or founder — for ongoing monitoring.
Re-run the full audit monthly and review quarterly for model or engine-level shifts.
Compare results against at least one competitor set, not just your own brand in isolation.

FAQ

How do you check whether ChatGPT recommends your competitors more than you?

You check by running a fixed list of buyer-style prompts repeatedly in ChatGPT, logging whether your brand or a competitor is named, in what position, and with browsing on and off separately. A single query isn't enough — aim for at least five to ten repeats per prompt before drawing conclusions.

ChatGPT typically favours competitors with clearer published content, stronger review platform presence, or more prominent live web citations when browsing is enabled. 73.58% of consumers have bought a product based on an organic AI recommendation, per the Semrush/Exploding Topics survey, so this gap carries real commercial weight.

What prompts should I use to test my brand's visibility in ChatGPT?

Use comparison prompts ("Compare X and Y"), unprompted recommendation prompts ("best provider for..."), problem-based prompts, and trust-check prompts ("is X legitimate?"). Unprompted recommendation prompts are the strongest test, since they reveal whether ChatGPT volunteers your brand without being told its name.

How often should I check if ChatGPT is mentioning my competitors?

Run a full audit monthly, with lighter weekly checks during a product launch or competitive campaign. A quarterly deep review helps catch shifts caused by model updates or changes across other engines like Perplexity or Google AI Overviews.

AI share of voice is the proportion of relevant prompts, out of your total tested set, in which your brand is named by the model. If your brand appears in 12 out of 30 repeated "best of" prompts, your share of voice for that query set is 40%.

Can I track ChatGPT brand mentions for free?

Yes, manually, by typing your prompt list into ChatGPT yourself and logging results in a spreadsheet — though this is time-consuming and harder to sustain at scale. Aether AI also offers a free AI-visibility audit at /audit as a lower-effort starting point.

What tools track brand visibility in ChatGPT versus competitors?

Dedicated GEO platforms, including Aether AI, automate this by querying multiple AI engines on a schedule and reporting mention frequency and competitor overlap. Aether AI covers ChatGPT, Perplexity, Google AI Overviews, Claude, Gemini and Microsoft Copilot from one dashboard.

Checking your visibility with Aether AI

This article has set out how to manually test, log and interpret ChatGPT's recommendations — but doing that reliably at scale, across six AI engines, every month, is exactly the gap Aether AI was built to close. Rather than re-running thirty prompts by hand, Aether AI's self-service platform automates citation tracking across ChatGPT, Perplexity, Google AI Overviews, Claude, Gemini and Microsoft Copilot alongside keyword and competitor tracking.

Aether AI is the same engine its parent, Aether Agency Ltd, runs for its own brands — across security, facilities software and branding — with 281 articles published across those brands in the last 30 days, demonstrating the platform's output at working scale.

If you want to see where you currently stand against named competitors before committing to anything, start with the free AI-visibility audit at /audit, or review Aether AI's published pricing to find the right level of ongoing monitoring for your business.