Why Your Competitors Show Up in ChatGPT and You Don't?

In simple terms, not having enough internet footprint (owned media and brand mention) is the reason why your brand is not appearing for the relevant queries in ChatGPT. 

Go to ChatGPT and ask: "Which agencies specialize in AI search optimization for B2B companies?" You will see a list of the top Organic growth agencies for SaaS in the answer, including companies like Omniscient Digital, SimpleTiger. SageMedia. Skale. TenSpeed and First Page Sage. Our brand, ScaleTheory, has not appeared once in the 200+ runs we made using our internal AI visibility tool. 

Let us explain the lack of internet footprint for the case of ScaleTheory.

• We do not have all the necessary pages on our website for ChatGPT to consider our pages as a reference for the queries relevant to us
• Our external brand mentions are just 31 (As of Aug 5, 2026)

This is also the problem that most startups face. This guide is specifically for companies that have started out with SEO in the past 6 months and yet to see an impact from the AI channels. 

Why Scale Theory doesn't show up — and why you're probably in the same position

Let us look under the hood of ChatGPT when a user asks a commercial SaaS query that triggers a web search. 

ChatGPT breaks down the query into multiple queries (query fanout), reads the live web including review platforms, comparison pages, directories, roundups, publications, etc. The companies that get the most weightage (influenced by the quality and quantity of mentions on the retrieved pages and the brand relevance to the topic) will be considered in the answers and citations.

All the companies that appeared in ChatGPT when I asked for ‘AI search optimization for B2B companies’ have already built a good presence on the web both on and off-site. When the model looks for an AI SEO agency, the evidence for each of them is abundant and consistent. Scale Theory has a website, a handful of blog posts, and a few directory listings. This is not enough for ChatGPT to recommend us with confidence.

AI visibility tracks directly with internet presence. We need to increase our brand presence across the internet - both on websites and on topics relevant to our ICP. This should be a super targeted strategy that we will be running as quarterly sprints covering 2 to 3 topics in a quarter. 

Step 1: Identify the prompts you actually need to win

This starts by identifying the topics that we need to focus on. Let me give you a couple of examples. For ScaleTheory, the first two topics we need to win will be ‘AI SEO agency for SaaS’, ‘SaaS AEO/GEO agency’

You can read our blog on how to build prompts that are relevant for your brand to get a detailed view on how we build relevant prompts to track. 

Build your list across three prompt types. Category prompts ("best [service] for [ICP]") are where most buyers start and where competition is highest. Alternatives prompts capture buyers already evaluating specific competitors by name. Problem prompts catch buyers earlier, before they've settled on a solution category — these tend to be less competitive and are often the fastest to win.

Step 2: Benchmark where you actually stand

We run the prompts in our internal tool ScaleX to benchmark the presence and citation number, understand the same metrics for competitors, and identify the opportunities. We run the prompts at least 5 times across the top 5 LLMs (around 25 runs for a single prompt), giving us enough data to plan the strategy with confidence.

For ScaleTheory, below is our presence data for the theme of ‘AI SEO agency for SaaS’ after 5 runs of 30 prompts under the tag of ‘Commercial: AI SEO agency for SaaS’. From a dataset of 600 answers.

You can do the same with the AI visibility tool you're already using. If you're not using one yet, it would make my day if you could check out our tool and share your honest feedback with me. The free plan lets you track up to 50 prompts across ChatGPT and Perplexity.

Here's how to set it up:

1. Log in here

2. Add your website, and all the other brand details will be auto-updated. Verify and proceed

3. You have two options: auto-generate the prompts or add prompts of your own. I highly recommend you add your own prompts, which will ensure that the tagging of prompts is exactly how we want them to be. 

4. Run the prompts on an alternative day basis every day (default setting, can be changed)

Step 3: Work out what's actually driving the gap

There are two major areas that we can work on from the insights of the tool: content and brand presence. 

Opportunity 1: The content opportunity

Group the cited pages based on the core keyword. This grouping helps you to analyze the citations based on keywords. 

Let us consider an example to simplify this. Consider the pages below that are cited for the prompts under the tag of ‘Commercial: AI SEO agency for SaaS’ along with the number of citations and core keyword.

Now we group this on the basis of keywords like in the table below.

Now I can see the topics that are getting cited the most. From this, we build our optimization and new content calendar by following the logic below

Opportunity 2: The outreach opportunity

Filter to view the cited pages where our brand is not mentioned. Targeted outbound should be run for a select few pages with a goal to get our brand mentioned in the target pages. 

Two types of pages to prioritize:

Bonus opportunity: The structural gap

Before content and outreach can compound, certain pages need to exist. For Scale Theory, and probably for you, the checklist looks like this:

• Homepage positioning: One clear sentence stating what you do, who you help, and what they get. A direct answer to "what is this company?" that the model can extract and use confidently.
• Solution and use-case pages: One page per specific problem, service, or customer type
• Proof pages: Case studies, client testimonials, and an FAQ
• Entity description: A single uniform sentence defining your brand across the internet. Homepage, G2 profile, LinkedIn bio, every directory listing. When the model sees the same description repeated across independent sources, it builds a much clearer picture of what you are

Step 4: Weekly progress tracking

Frequently asked questions

Can I pay to appear in ChatGPT recommendations?

No. There's no ad slot inside ChatGPT's answers — everything you see in AI recommendations is earned through content the model can find and cite, and through third-party sources that mention your brand. The upside of this is that a well-structured smaller brand can outperform a funded incumbent in AI answers if it builds coverage in the right places. The downside is there's no shortcut, and anyone telling you otherwise is overselling.

Why does ChatGPT recommend a competitor that's clearly not as good?

Because it isn't scoring on quality — it's naming brands it has the most evidence for. A competitor with more coverage, more review-site presence, and more structured pages gives the model more to work with, even if their product is weaker. The gap isn't in the product; it's in the coverage. And coverage is fixable.

How many prompts should we be tracking?

Start with 20-40 across category, alternatives, and problem types. VisibilityX handles up to 50 at no cost. Don't make the list too broad — you want prompts your actual buyers are running, not every possible permutation. 

Does improving Google SEO help with this?

Yes, but it's not the same thing, and the overlap varies by platform. Strong rankings help most with Gemini, which relies heavily on Google's index. A well-ranked page is also easier for ChatGPT's browsing mode to surface. But ChatGPT also weights third-party mentions and page structure independently of rankings — so SEO alone won't close the gap. Treat them as overlapping work rather than identical work.

How do we know if outreach placements are actually working?

Check the citation count for the specific pages where your brand was added. If those pages start driving more citations for your category prompts in VisibilityX, the placement is pulling through. If a high-volume page includes your brand but your citation count stays flat, something else is off — either the mention isn't structured in a way the model can extract, or the page hasn't been re-indexed yet. Give it 3-4 weeks before drawing conclusions.

What's the minimum starting point for a brand-new website?

Three things, in this order: homepage positioning, entity description, and at least two topic-specific solution pages. Without clear positioning, the model can't describe what you do. Without a consistent entity description, your signals don't reinforce each other across independent sources. Without topic pages, there's nothing to cite even if the model finds you. Get these three right before focusing on anything else.