The $200,000 Problem Nobody's Talking About

Table of Contents

Ask ChatGPT "who's the best Webflow developer in Manchester?" and you'll get a specific answer. Not a list of links. Not a set of ads. A recommendation — with reasoning. Now ask yourself: when your ideal clients ask that question about your industry, your category, your specialty — does your name come up? Or does a competitor's?

This is the question that's reshaping how businesses think about visibility. And the answer, for most small businesses, is uncomfortable: they're not in the conversation at all.

The Question That Changed Everything

Every day, potential clients are asking AI for recommendations. Not Google. Not LinkedIn. AI. And AI is answering with specific names, specific businesses, specific justifications for why it chose them.

The businesses getting recommended share a common trait: they showed up in the AI's training data and retrieval systems with clear, structured, authoritative content. Everyone else? Invisible. Not ranked lower — absent entirely.

This isn't a gradual shift. It's a binary one. Either AI can find and understand your content well enough to recommend you, or it can't. There's no middle ground.

The Numbers

OpenAI has launched advertising within ChatGPT. The pricing is revealing:

• Minimum spend: $200,000
• CPM (cost per thousand impressions): approximately $60
• Context: That's roughly 3x Meta's advertising costs and nearly 2x Google Search ads

The early adopters are brands with substantial advertising budgets. They can afford to pay for placement inside AI conversations, and they are buying visibility in a channel that is still growing.

For context, Conductor research reported that 32% of digital marketing leaders were prioritising Generative Engine Optimisation in 2026, with 12% of digital budgets allocated to it. A third figure often quoted alongside those two, that 97% of those investing in GEO report positive results, I have not been able to trace to the study, so I am not going to lean on it.

The market opportunity is real. The price of entry through advertising is not accessible to most businesses.

Who Can Afford to Play

A $200,000 minimum spend creates a hard barrier. This isn't a matter of stretching a budget or starting small and scaling. It's a price point that excludes the vast majority of businesses:

• Solopreneurs and freelancers
• Small and medium agencies
• Early-stage SaaS companies
• Local businesses
• Bootstrapped startups

Even businesses that could technically afford it face a questionable ROI. At $60 CPM, you need massive volume to justify the investment. For niche businesses serving specific markets, the numbers rarely work.

This creates a two-tier system: enterprises that can buy AI visibility, and everyone else who needs to earn it. The question for smaller businesses isn't "should we advertise on ChatGPT?" — it's "how do we get recommended without advertising?"

Why AI Isn't Like Google Ads

There's an important distinction between AI recommendations and traditional advertising. Google Ads are transactional: you pay per click, you get traffic, you stop paying, the traffic stops. It's a tap you can turn on and off.

AI recommendations work differently. When ChatGPT recommends your business, it's not because you paid — it's because the AI found your content, understood your expertise, and judged you relevant to the user's question. That recommendation carries implicit trust. The AI isn't saying "here's a sponsored result." It's saying "based on what I know, this is who you should talk to."

Research indicates that AI-driven referral traffic converts at significantly higher rates than traditional search traffic. Users who arrive at your site because an AI recommended you are pre-qualified — they've already been told you're relevant to their specific need.

The value of an organic AI recommendation is qualitatively different from a paid impression. And unlike paid advertising, organic recommendations continue working after you stop spending.

The Alternative to Paying

If you can't buy AI visibility, you need to build it. The good news is that this is achievable. The not-so-good news is that most businesses haven't started.

AI systems don't recommend randomly. They follow patterns — selecting sources that are structured for machine readability, demonstrate clear expertise, and provide explicit signals about what the site offers. The businesses that earn organic AI recommendations have typically done three things:

1. Made their content machine-readable — Structured data, clean HTML, llms.txt files, and clear content hierarchy. AI systems must be able to parse your content before they can cite it.
2. Demonstrated expertise explicitly — Named authorship, credentials, specific data, and original insights. AI systems need to trust a source before recommending it.
3. Told AI what to focus on — An llms.txt file is the equivalent of a curated menu for AI systems. Without one, you're relying on AI to guess which of your pages matter — and it usually guesses wrong.

None of this costs $200,000. It costs time and attention — but the financial barrier is orders of magnitude lower than paid AI advertising.

Getting Started

The first step is diagnosis. Before you optimise anything, you need to know what AI can see when it looks at your site. Most businesses are surprised by the answer, and not pleasantly.

Score your own pages against the MASTERY-AI framework. Check whether you have llms.txt deployed. Search for your core services in ChatGPT, Claude and Perplexity and see whether you appear. That baseline takes less than an hour and tells you where you stand.

The $200,000 problem is real — but it's only a problem if you think paying is the only path. For a business willing to earn its AI visibility, the methodology is published and the window for early advantage is still open.