Individuals flip to ChatGPT for researching, constructing instruments, and—sure—even shopping for.
However many manufacturers do not know what’s occurring inside ChatGPT. They’re exhibiting up in 1000’s of responses with out understanding which queries point out them, what’s being stated, or how they stack up towards the competitors.
You’ll be able to construct fundamental vibe-coded instruments to observe your ChatGPT visibility, however monitoring a small variety of prompts is like attempting to foretell the climate by looking of your window each as soon as in a whereas.
The truth is AI prompts, responses, and citations shift consistently. You want 1000’s of information factors to chop by way of the variance if you wish to see precise tendencies.
There’s a simple repair. You’ll be able to observe your ChatGPT visibility throughout 13.5 million current prompts inside Ahrefs Model Radar database, and even arrange your individual customized queries to observe what issues most.
Listed below are seven steps for utilizing Model Radar to measure, monitor, and enhance your AI visibility.
Model Radar helps you to observe as much as ten opponents alongside your model to see precisely the way you stack up in ChatGPT responses.
Simply enter your opponents manually or hit “Counsel extra” for AI suggestions.
It’s necessary so as to add your opponents’ domains or URLs in setup, if you wish to observe their web page visibility and citations in ChatGPT later down the line.
To refine your evaluation, you can too embody identify variations—that is particularly helpful when a competitor’s model identify can also be a generally used phrase.
When you’re totally arrange, you’ll see side-by-side metrics for each model, together with:
- Mentions: The overall occasions every model seems in AI responses
- Citations: The variety of responses that cite the model URL no less than as soon as
- Impressions: Weighted mentions primarily based on search demand
- AI Share of Voice: The share of complete impressions vs. the competitors
From there, you may run a competitor evaluation throughout the complete Model Radar database, or alternatively slender your focus by market or area of interest.
For instance, for Ahrefs we would zero-in on the subject of “Backlinks”, because it’s one in all our core subjects and is closely embedded in our product.
In the event you’re questioning: “Who’s extra seen in ChatGPT: me, or my greatest competitor?” you will discover the reply, and even see the way it adjustments over time, with a report referred to as “AI Share of Voice” (AI SoV).
Right here’s how you can verify it:
- Enter your model identify into the Model Radar dashboard
- Drop in your opponents, or hit “Counsel extra” for AI strategies
- Test your AI Share of Voice proportion
- For an excellent deeper breakdown, hit the AI Share of Voice tab
This reveals your model’s share of voice throughout the prompts most definitely to be seen by actual customers.
That’s as a result of Model Radar focuses on popularity-weighted visibility—not simply whether or not you seem, however whether or not you seem the place it counts.
Sidenote.
Each immediate in Model Radar is seeded from search demand information, which means each comes with related impressions to offer you a clearer concept of subject recognition. It’s also possible to consider these impressions as immediate “volumes”. We observe on this method that will help you prioritize the prompts which can be most definitely to matter.
You’ll be able to observe AI Share of Voice over time to see in case your content material technique, PR efforts, or brand-building actions are literally bettering your AI visibility.
Rising AI SoV means ChatGPT is recommending you extra typically. Declining AI SoV means you’re shedding floor.
Use this benchmark as your start line, then dig into the particular methods in the remainder of this information to know why your AI SoV is what it’s and how to enhance it.
You’ve seen which opponents are beating you (and which you’re beating)—now it’s time to trace the precise content material that’s incomes you visibility in AI search.
The goal is to work out what ChatGPT already “is aware of” about your model, and believes you to be an professional in.
Model Radar breaks down precisely which subjects you’re really useful for, which queries point out you most frequently, and which domains are chargeable for your model getting talked about and cited.
Right here’s how you can discover your high subjects:
- Navigate to the Matters tab
- Filter the AI platform by:
ChatGPT - Focus in your model subjects with the filter:
your model: talked about - View an inventory of the subjects most carefully related together with your model in ChatGPT
This report offers you a quick overview of the core subjects ChatGPT surfaces your model for, however for extra element, head to the AI responses report.
That is the place you’ll discover the precise prompts and solutions your model is included in.
Hitting “Present extra” ³ will allow you to view the precise response, so you may perceive the context of your mentions.
As an illustration, under, Ahrefs is talked about in reply to the question “Which instrument is finest for competitor evaluation?” due to a advice from the location One Little Net.
Responses present you precisely what ChatGPT says about your model, so you may spot outdated info, incorrect claims, or lacking options that want publicizing. Use this data to jot down new articles that fill in info gaps, or replace previous articles to appropriate something outdated.
And at last, you may audit your cited pages: the URLs which can be most definitely chargeable for your mentions in ChatGPT responses—in our case, that’s a Wiki about “Google Search Console”, a Tech Radar evaluation, and a Forbes instrument checklist.
A few of these pages will explicitly point out your model, and others will simply be cited within the AI responses alongside your model point out.
Deal with pages that don’t suggest you as an outreach goal, to strengthen your visibility in ChatGPT.
On this report, you can too see which of your web site pages get cited probably the most by ChatGPT, by including a area filter.
Because you management these pages, and also you already know they’re most well-liked sources, updating and optimizing them is the quickest method to fill gaps in what ChatGPT is aware of about your model.
Which brings me neatly on to the following step…
Probably the most actionable factor you are able to do when monitoring your ChatGPT visibility is locate subjects the place opponents get talked about however you don’t, then goal these gaps.
There’s a fast and simple method to do that in Model Radar straight from the dashboard.
Simply head to the ChatGPT “Mentions” bar chart and hover over your model, then click on “Others solely”.
This can immediately generate a number of filters within the AI responses report, exhibiting you the prompts, responses, and fan-out queries that exclude your model, however embody your opponents.
Use this subject route to create new content material, optimize current posts, and reclaim a few of that market share out of your opponents.
Right here’s an instance of what I imply.
Not too long ago, I ran a niche evaluation for Ahrefs throughout a small variety of bottom-funnel customized prompts that we actually care about proper now.
I found we had been getting talked about in 42 ChatGPT responses whereas a competitor appeared in 66.
From that small point out hole, I used to be capable of finding out precisely the place we had been shedding floor and which subjects we would have liked to focus on to have a greater likelihood of closing that visibility hole.
By analyzing prompts and fan-outs within the AI responses report, I realized that our subjects gaps vary from professional endorsements (highlighted in yellow under) to ROI justification (inexperienced), and platform protection (orange).
In different phrases, we have to create extra decision-stage content material in clear codecs.
For instance:
- Knowledgeable case research on how groups use Model Radar to behave on AI visibility (“How Hostinger good points visibility in AI and automates repetitive duties”).
- ROI and price range analysis guides (“What ROI really seems like for AI visibility monitoring”).
- And sensible, platform-specific blogs (“Monitoring model visibility throughout AI Mode: Our strategy”).
This can assist us present up extra typically in high-purchase-intent queries in ChatGPT.
And it’s not nearly “Point out” gaps. It’s also possible to observe the place your opponents are getting cited, however you aren’t.
Simply hit the Citations tab on the dashboard, hover, and click on “Others solely”.
From there you’ll see ChatGPT responses the place rival domains are cited, however you aren’t.
To get a whole image of your missed visibility, add an additional filter so that you’re conversations the place you’re neither cited nor talked about.
Then seize concepts from opponents’ pages to counterpoint your individual content material—whether or not that’s including in unique information and statistics, bettering formatting and construction, lowering time-to-insight, optimizing meta information and snippets, or constructing out your protection throughout the subject on the complete.
Your AI visibility doesn’t exist in a vacuum. ChatGPT pulls from content material that’s already performing throughout the net, YouTube, Reddit, and different platforms.
Model Radar helps you to observe your model throughout all these channels in a single place, so you may see how your broader visibility feeds into AI mentions.
Right here’s how you can analyze every channel:
Monitor your YouTube visibility
YouTube is the second-largest search engine and the third most-cited supply by ChatGPT for model suggestions.
YouTube mentions additionally correlates most carefully with model mentions in ChatGPT, in keeping with our personal analysis of 11 totally different AI model visibility elements.
That means that when creators make movies about your product, ChatGPT could use that content material as supply materials for suggestions.
Click on into Video visibility > YouTube to see movies mentioning your model.
You’ll be able to cross-reference this information towards the Cited pages report back to see which movies containing mentions of your model are additionally informing ChatGPT solutions.
Search for patterns in what subjects are lined and which creators point out you—then it’s only a case of manufacturing and sponsoring extra of that content material.
Test your net visibility
In accordance with our personal analysis of 75,000 manufacturers, branded net mentions have one of many strongest correlations with visibility in ChatGPT.
So, having the ability to see tendencies in each your net mentions and AI visibility will help you’re employed out how your off-site efforts are impacting your ChatGPT efficiency.
Navigate to the Net visibility tab to view these tendencies, and see which precise pages point out your model throughout the net.
There’s one other straightforward method to spot whenever you’ve been talked about by a web page that’s cited in ChatGPT.
Simply head to the Cited pages report and take a look at the “Mentions” column.
Everytime you spot your model’s initials in inexperienced, it implies that the URL in query has talked about your model identify throughout the content material of the web page.
Monitoring metrics like model mentions will help show how impactful your PR and model consciousness efforts are in driving ChatGPT visibility.
Monitor your search demand
Rising search quantity on your model is commonly a number one indicator of ChatGPT progress.
For instance, right here’s Canva’s ChatGPT visibility and branded search quantity side-by-side.
Whereas search quantity is just not as strongly correlated as model mentions, there’s nonetheless a transparent optimistic relationship.
In case your model is looked for extra typically, it’s extra prone to seem in search outcomes—and that visibility can enhance the possibilities of it being grounded in ChatGPT solutions.
You’ll be able to observe your branded search demand tendencies and queries inside Model Radar within the Search demand tab.
Scan Reddit conversations
Reddit discussions are a goldmine for ChatGPT as a result of they include actual consumer opinions and proposals.
In reality, Reddit is ChatGPT’s #1 cited area throughout 13.5M prompts, in keeping with Model Radar.
When somebody asks “What’s one of the best website positioning instrument?” on Reddit and 15 folks suggest Ahrefs, ChatGPT absorbs that consensus.
Navigate to SERP visibility > Reddit to see Reddit threads mentioning your model.
Evaluation high-traffic subreddits that point out your model, gauge the sentiment of these discussions, and create a plan to handle questions and information the narrative with out coming throughout as overly promotional.
Monitor TikTok mentions
TikTok may appear unrelated to AI search, nevertheless it’s more and more influential.
Movies explaining how you can use your product or evaluating you to opponents assist set up your class positioning—which ChatGPT then displays.
Go to Video visibility > TikTok to see TikTok movies that includes your model.
Model Radar offers you on the spot visibility into hundreds of thousands of prompts throughout ChatGPT, however you’ll most likely need to additionally monitor the queries that matter most to your small business by yourself schedule and throughout the platforms you care about.
Fortunately, customized immediate monitoring in Model Radar helps you to do precisely that. You’ll be able to observe the precise questions your clients are asking, whether or not that’s “finest CRM for building firms” or “is [your product] price it in 2026.”
The hot button is grouping associated queries collectively and analyzing combination patterns. This provides you a dependable directional view of your model’s efficiency.
For instance, you may cluster your customized prompts by intent, funnel stage, or product class.
Listed below are some extra concepts:
- Branded queries: To truth verify and spot misinformation patterns. (e.g. “Is Ahrefs information correct?”)
- Backside of funnel queries: To grasp how the AI recommends your model throughout buy selections. (e.g. “Greatest website positioning instrument for a small company?”)
- Mid funnel queries: To grasp your skill to coach AI audiences. (e.g. “What’s key phrase issue and the way is it calculated?”)
- Characteristic primarily based queries: To zero-in on conversations surrounding your focus merchandise, and perceive how higher to place and promote them. (e.g. “What are you able to do with Ahrefs Content material Hole instrument?”)
- Marketing campaign associated queries: To evaluate the affect of your PR and advertising efforts on ChatGPT visibility (“What’s Ahrefs’ AI model visibility correlation research about?”).
As one other instance, I’ve arrange a gaggle of 100+ BOFU customized prompts pertaining to the subject of “AI visibility instruments”, since that’s actively the place we’re attempting to develop our model consciousness proper now.
I did this by increasing a whole lot of trending AI visibility key phrases (found in Ahrefs Key phrases Explorer) into fuller “prompt-like” queries.
To watch your individual customized ChatGPT prompts, simply choose a Undertaking and select your monitoring schedule—each day for high-priority phrases, weekly for mid-priority queries, or month-to-month for model sentiment.
Then you may choose which platforms, places, and opponents to observe.
And at last, observe your progress.
Suppose: which manufacturers are talked about most continuously on common, which of your individual pages are persistently cited, and the way does your visibility development over time?
This provides you actionable insights with out getting misplaced in response volatility.
When ChatGPT can’t confidently reply a query utilizing its coaching information alone, it retrieves info from exterior sources equivalent to Google, in a course of generally known as retrieval-augmented technology (RAG).
As a part of this course of, it generates “fan-out queries,” breaking a single query into a number of associated sub-queries to discover totally different angles and supply up a extra full reply.
For instance, if a consumer searches “What’s the finest web site internet hosting firm?” ChatGPT generates fan-out queries to raised discover two angles: latest product opinions, and have comparisons.
Many SEOs consider that constructing and optimizing content material clusters round fan-out queries is vital to AI visibility.
The problem is, fan-out queries are simply as changeable as prompts.
That’s why it’s worthwhile to research plenty of them in combination, if you wish to establish recurring themes/subjects/angles to construct into your subject clusters.
Fortunately you may observe hundreds of thousands of ChatGPT fan-out queries in Model Radar.
Simply head to the AI responses report, and look out for the Fan-out queries column.
Professional tip:
Replicate top-cited content material—then improve it with fan-out optimization
Head to the Cited pages report, and discover the most-cited URL in your class.
Copy it.
Then head again to the AI responses report and paste that URL into the filter field.
You’ll discover the entire prompts, responses, and fan-out queries wanted to both create or replace your individual model of that content material—and you may even spin out a number of cluster items from the fan-out subtopics you uncover.
This can stand you in higher stead to be cited by ChatGPT.
Right here’s your motion plan:
- Begin together with your benchmark. Test your AI Share of Voice versus opponents throughout all queries in Model Radar—that is your baseline for measuring your general visibility in ChatGPT.
- Arrange complete monitoring. Create customized immediate monitoring on your most necessary queries (branded searches, bottom-funnel phrases, and new product options) and schedule them to run each day, weekly, or month-to-month.
- Monitor in context of different channels: Periodically monitor the channels that feed into AI visibility: YouTube mentions, Reddit conversations, net citations, and branded search demand.
- Evaluation and act on gaps. Each month, run a aggressive hole evaluation to search out subjects the place opponents seem however you don’t. Examine the fan-out queries and cited pages in these responses, then create or replace content material clusters that handle each the principle prompts and their associated subtopics.
- Monitor the affect. Watch how your AI SoV tendencies upward—each on the class and database degree—as you publish optimized content material, earn citations throughout high-authority pages, and construct presence elsewhere on-line—proving that your cross-channel efforts are translating into ChatGPT visibility.
In the event you’re curious the place you stand, get began with Model Radar.
