Most businesses have no idea whether ChatGPT mentions them. The ones that do check usually do it inconsistently, using one engine, with no record of what changed. Here's a structured approach to monitoring your brand across all major AI engines, from the free manual method to a fully automated pipeline.

Can I track brand mentions in ChatGPT for free?

Yes. The manual approach costs nothing and takes about 30 minutes a week. It's not scalable, but it's the right starting point before you invest in tools.

Here's the manual process: open each AI engine and run a set of queries as if you were a potential customer. Use prompts like:

"Best [your category] in [your city]"
"Who are the top [your service] providers in [your market]?"
"Recommend a [your product type] for [customer use case]"

Run the same set of queries every week. Log the full response in a Google Sheet with columns for: date, engine, query, whether your brand appeared (Y/N), whether your website was cited (Y/N), which competitors appeared.

Do this for 4 weeks before drawing conclusions. One week of data is noise. Four weeks shows a pattern.

The engines to check manually:

chat.openai.com (ChatGPT)
chatgpt.com (same model, different URL, sometimes different responses)
perplexity.ai
gemini.google.com
bing.com/chat (Microsoft Copilot)
claude.ai

According to SparkToro and Datos, roughly 1 in 3 Google searches in late 2024 triggered an AI Overview. That number has continued growing. Monitoring these six engines covers the vast majority of AI-generated brand mentions.

What free tools exist for AI brand monitoring?

Three tools offer free or trial access: HubSpot AI Search Grader, ProductRank.ai, and Ahrefs Brand Radar. Each has limitations, but they're a better starting point than nothing.

HubSpot AI Search Grader is the most accessible free option. Enter your brand and a few keywords, and it gives you a visibility score across ChatGPT and Perplexity. It doesn't track over time (no historical data), but it's useful for a one-time benchmark. Free, no account required.

ProductRank.ai has a free tier that lets you track a handful of queries. It's lightweight but functional for solo founders or small teams running their own monitoring. The paid tier adds more queries and engines.

Ahrefs Brand Radar is included in Ahrefs plans, so it's "free" if you're already an Ahrefs subscriber. The limitation is that it focuses on Google AI Overviews rather than ChatGPT or Perplexity. If Google AI Overviews are your primary concern, this is convenient. If you care about chatbot mentions, you'll need something else alongside it.

Which paid tools are worth it?

Peec AI at $99+/month is the most comprehensive option for small agencies and consultants. Otterly at $49+/month is better for individual businesses monitoring themselves. SE Ranking's brand monitoring is worth it only if you're already a customer.

Otterly ($49+/month) tracks ChatGPT and Perplexity with a clean interface. It's built for businesses that want to do their own monitoring without a lot of setup. The reporting is basic but readable. Best for: solo founders, small marketing teams, businesses in one market.

Peec AI ($99+/month) tracks 5+ engines, has a more developed prompt library, and produces better reports for sharing with clients or stakeholders. It's more useful if you're managing AEO for multiple brands or need documentation-quality output. Best for: agencies, consultants, in-house teams that report to leadership.

SE Ranking ($44+/month, Brand Radar included) makes sense if you're already using SE Ranking for SEO. The AI visibility data sits in the same dashboard as your other SEO metrics. The coverage skews toward Google AI Overviews. Best for: existing SE Ranking customers who want some AI monitoring without a separate tool subscription.

Frase is primarily a content optimization tool, but it includes AI visibility features. If content production is your main workflow and monitoring is secondary, Frase covers both. Best for: content teams that want AEO insights baked into their writing workflow.

According to Semrush's 2025 AI report, 42% of marketers plan to increase investment in AI search monitoring tools in the next 12 months. The tooling market is moving fast. Pricing and feature sets will shift.

How do you build your own monitoring pipeline?

You need three API keys and n8n (or Make). Total cost: $10-15/month. This is what we built at Flowforge for client monitoring, and it runs every morning at 6am SGT without manual input.

Here's the stack:

OpenAI API: query ChatGPT programmatically. Cost: ~$20/month at low to moderate query volume.
Perplexity API: query Perplexity. Cost: ~$20/month.
SerpAPI: query Google AI Overviews and Bing Chat. Cost: ~$50/month.
n8n self-hosted: the orchestration layer. Free if you host it yourself.
Google Sheets: results storage and reporting. Free.

The workflow logic is straightforward. At 6am, n8n loops through a list of tracked queries, calls each API with the query text, parses the response for brand mentions and domain citations, writes a row to Google Sheets with date, query, engine, mention (Y/N), citation (Y/N), and the full response text.

You end up with a running log of every query result, which makes trend analysis easy. You can see exactly when your AI visibility changed and correlate it with content you published.

The upside over SaaS tools: full data ownership, lower per-client cost, and the ability to customize exactly what you track. The downside: setup time is 1-2 days the first time, and you need to maintain the pipeline when API formats change.

What metrics should I actually track?

Track four numbers: visibility rate, citation rate, share of voice, and sentiment. These give you enough signal to know if your AEO work is moving the needle.

Visibility rate: the percentage of your tracked queries where your brand appears in the AI response. Formula: (queries with brand mention / total queries) x 100. A starting baseline for most businesses without deliberate AEO work is 0-20%. A good 6-month target is 30-50% in your core queries.

Citation rate: the percentage of responses where your website domain is explicitly linked or cited. This matters more in Perplexity (which cites sources by default) than in ChatGPT. Track them separately.

Share of voice: how often your brand appears versus named competitors in the same response set. If ChatGPT mentions your top competitor in 60% of relevant queries and you appear in 15%, that gap is your target.

Sentiment: whether the mention is positive, neutral, or negative in context. Most brand mentions in AI responses are neutral (just listed), but watch for patterns in how you're described.

How do I track AI referral traffic in GA4?

Set up a custom traffic source filter in GA4 for the major AI engine domains. This connects AI mentions to actual website visits.

Add these as traffic sources to monitor in GA4:

chat.openai.com
chatgpt.com
perplexity.ai
gemini.google.com
bing.com/chat
claude.ai

In GA4, go to Reports > Acquisition > Traffic Acquisition and filter by session source. You can also create a custom channel group called "AI Engines" and include these domains as matching rules.

This matters because Perplexity in particular drives real traffic when it cites your domain. Users click the source links. Quantifying that traffic helps you calculate the revenue value of being cited.

How often should I check AI visibility?

Weekly is the minimum. Daily is better if you're in a competitive category or have recently published new content. AI engine responses can shift noticeably within days of Google indexing new pages.

The manual approach can realistically only run weekly. Tools like Peec AI and Otterly typically run daily or weekly checks depending on your plan. The self-built pipeline approach is easy to run daily because it's automated.

Don't check more than once a day. AI engines have some randomness in responses, so daily variation is normal. Weekly or monthly trends are what actually tell you something useful.

What's a good AI visibility rate to aim for?

In a non-competitive local category, 40-60% visibility across your core queries within 6 months is achievable. In competitive national or global categories, 20-30% is a more realistic 6-month target.

The right benchmark is always relative to your competitors, not an abstract percentage. If the top competitor in your category appears in 70% of relevant AI queries and you appear in 10%, the gap is the story. Close it systematically.

Most businesses that start deliberate AEO work move from near-zero visibility to measurable mentions within 60-90 days. The content that tends to move the needle fastest: FAQ-style pages, structured comparison content, and anything that directly answers the queries AI engines get asked most.

At Flowforge, we run this monitoring pipeline for clients and include the results in weekly automated reports. If you want to see your current AI visibility across ChatGPT, Perplexity, and Gemini before deciding what to do, the free audit gives you a starting point.