AI Visibility Is Becoming the New Front Door for Startups

By Julia Maslennikova, founder & CEO of 25/8 PR, an international agency for startups & tech companies. This is an Op-Ed on Operators' Stack, Menlo Times.

A startup can create an excellent product and successfully build its media presence, yet still go unnoticed when a potential customer asks an AI for information about it. 

I see this as a real business problem, since people are increasingly using ChatGPT and other LLMs to search companies, compare products, and decide who to work with. 

For startups competing against companies with much larger marketing budgets, being accurately represented in LLM responses is a crucial part of gaining a foothold in the market. 

You need to start asking yourself: What does AI say about us when a potential customer or investor asks a relevant question?

AI Search is Changing What Visibility Means.

We are already seeing this trend among our clients, and it is important to note that many companies haven’t yet realized just how significant AI visibility has become. A few months ago, several clients reported that traffic generated by LLMs accounted for only fractions of a percent or low single-digit percentages of their total traffic. Today, for some of them, those figures have reached tens of percent. The exact trends vary by company, but the overall trend is: people are Googling less and are increasingly turning to ChatGPT or other LLMs.

LLMs can gather information from a company website and LinkedIn, as well as from specialized publications, Reddit, Quora, reviews, interviews with founders, industry blogs, platforms like G2, and other third-party sources. Together, these sources form an information ecosystem around the brand. 

At the same time, AI goes beyond simple search, as agents are capable of conducting research, comparing options, and taking action on behalf of users. This creates a new layer of customer interaction, increasingly described as B2A, or business-to-agent. Therefore, companies now need to make their products understandable not only to people but also to AI systems.

What Actually Makes a Company Visible to AI?

What determines how visible a company is to LLM systems? Generally speaking, it depends on whether the AI can identify the company as a suitable solution to a specific customer problem.

A recent study by John Gale, Luca Cian, and Luc Wathieu shows how varied the results of such visibility can be. Of the 716 brands studied, only 8.4% appeared identically in ChatGPT, Claude, and Gemini. Among the brands that appeared on multiple platforms, 55% were interpreted differently across the various systems. A company may have high visibility in one LLM system but be absent or described differently in another.

For AI to recommend your company, it must understand what customer problem the product solves and why your company is the right fit to solve it. The product must also be described using specific and comparable characteristics - for example, its features, use cases, performance, and measurable results. General statements like “innovative” or “best-in-class” won’t be very helpful.

The query itself is also important - AI can recommend different brands within a company in response to a general question or a specific problem. That’s why it’s important to check the queries that your customers actually ask.

Our analysis of how clients are represented in LLM responses compared with competitors shows that AI systems do not rely exclusively on Tier1 publications with millions of monthly visitors. Smaller and quality resources can also play an important role - we have seen local markets publications with 100K monthly visits, and sometimes even fewer, appear among the sources referenced by AI systems. 

What Should Companies Measure?

As we have seen, AI visibility should be assessed based on several metrics.

AI Visibility Score and Average Position shows how often your company appears as an option and whether it is considered within a relevant group of competitors. Other useful metrics: What AI Reads that shows which publications, platforms, communities, and other sources are shaping how AI understands a company and its market. What AI Cites shows which sources AI actually references in its responses, helping you understand which sources have the strongest influence on how they are represented to potential customers and investors.

It is important to note that AI-based search may generate less direct traffic than traditional search, so mentions and citations should be considered alongside publications in traditional media and conventional PR efforts.

The Shift From Being Searchable to Being Understandable.

I am confident that companies do not need to wait for a universal AEO standard to emerge - they can start by identifying issues of commercial significance, examining how various LLMs currently represent the brand, and comparing these results with competitor data. 

Your next step is to make the company more understandable to the AI. Structured Q&As, specific testimonials, product comparisons, and clear use cases can help with this. At the same time, PR, the founders' ideas, interviews, reviews, blog discussions, and other independent sources can reinforce the company's expertise and positioning.

Finally, companies should track how the situation changes over time by repeatedly searching for the same relevant queries and comparing the results with those of competitors. You should pay attention not only to whether the brand appears in the results, but also to how it is described, which sources are cited, and whether the response reflects the positioning the company wants to establish. Our goal is to provide the AI with sufficiently clear information and reliable evidence so that it understands what the company does, recognizes its expertise, and recommends it when appropriate.