LLM Ads & Content Optimization (LACO)

How to increase LLM visibility and LACO business opportunities.

TL;DR

• LLM Advertising Content Optimisation (LACO) is optimizing content to rank higher in responses from large language models (LLMs) like ChatGPT, both organically and through future ad placements.
• LLMs are already influencing consumer decisions (“ChatGPT recommended this…”).
• Ads within LLMs will happen.

In ChatGPT, type your full name. Below is mine. It thinks that I am a professor from an elite university (unfortunately, I’m not).

“What’s the best noodle restaurant in Brooklyn, New York, for my highly energetic twin sons?”

How about queries like “What’s the best CRM [or insert your product here]?” How could you influence the result in the way you want?

Tactics to Influence LLMs

This is a fast-evolving field. What works now can suddenly break.

Compared to SEO (search engine optimization), LACO is easier to game (as of July 2025), but also more volatile and complex. LLMs like clear, declarative, factual, structured content that is more ‘academic’ than traditional search engine preferences.

• Good SEO (Search Engine Optimization).
• Prompt Engineering: Anticipate user queries (e.g., “Best CRM for remote teams?”) and tailor content.
• Structure content clearly: TL;DR, bullet points, clear conclusions to help AI parsing.
• Publish on commonly scraped sources (e.g., Reddit, GitHub, public blogs).
• Optimize for accuracy, authority, and readability (LLMs love clean input).

Technical considerations

How to influence the crawling/training process (offline pretraining)

LLMs are not constantly using the internet. They are trained on snapshots of publicly available web data (Common Crawl, GitHub, Wikipedia, books, Reddit) and licensed datasets. This means you can’t control what gets trained directly unless you’re in the dataset. But you can increase the likelihood that your content gets in and is learned well.

How to influence AI’s tool-based agentic behavior (online querying / Retrieval-Augmented Generation)

Many newer LLMs search the web in real-time or use RAG (Retrieval-Augmented Generation) to provide up-to-date responses. The tactics in the previous section help you to get more visibility in this case.

Emerging Business Models

• LACO Agencies: Help brands rank and get mentioned in LLMs.
• Making boring content fun: There has been an overemphasis on how products can save/make more money and time. How about killing time?
• LACO Analytics & Tracking Tools: Measure and attribute visibility/referrals.
• Prompt Engineering Tools for reverse engineering user queries: Suggest what users might be asking LLMs.
• AI Scraping Workarounds: legally bypassing LLM crawling restrictions (e.g. LinkedIn blocks). Cloudflare enabled content owners to charge AI crawlers.
• Multi-platform LACO Services: Optimize across OpenAI, Anthropic, Gemini. They all have different versions of their models that give different outputs (GPT-4, o4-mini…).
• Query Popularity Proxies: Predict top prompts, like keyword planners for LLMs.
• Affiliate marketing + LLM Tools: Monetize via links embedded in LLM outputs.
• Non-English language LACO, e.g. Chinese and Arabic.
• Paid ads: businesses to create when people can buy paid LLM ads.
• ChatGPT store: Currently, you can’t directly monetize custom GPTs in ChatGPT, but you can drive affiliate traffic or brand visibility for monetization.
• Multi-modal push notification ads from ChatGPT. E.g. the specific hairdryer a user wants is at a 20% discount. The user can consume the ad in text or voice form about this sale.

How do people feel about this?

Big tech

Google’s share of global search traffic fell to 89.71% in March 2025. Brands such as Vercel and Finder.com.au are already getting marketing referrals from ChatGPT. Some people are not spending as much time in search engines and social apps as they used to. Will Google Gemini introduce ads? Will it eat into or blur the lines of search revenue?

SEO SaaS incumbents

SEO incumbents include Semrush, Ahrefs, and Similarweb. They haven’t grown as large as players in adjacent categories like CRM and Marketing Automation. Will the LACO market be fragmented too? Will the large LLMs take the lion’s share of the advertizing profits?

Agencies and brands

There’s increasing interest in “my client just got a business referral from ChatGPT, no idea what queries lead to it”, “could you help me look into how I can make my brand more visible for AI?”

Narrative Control and Ethical Risks

Ads within LLMs can be seen as an easy way to prey on people. It’s also a game of different interest groups trying to influence people’s views, as it always has been.

References

Friends: Yunfei O, Jon T, Steve Y, Cat H, Adrian L, Ray Y, and Arun P.

https://growfusely.com/blog/llm-content-marketing/

https://a16z.com/geo-over-seo/

https://gs.statcounter.com/search-engine-market-share#monthly-202206-202503

https://blog.hubspot.com/marketing/generative-engine-optimization