LLMO & AI marketing
training for your team
Four modules teaching your own people to make the company visible to AI systems — executives, developers, writers and PR, each taught the part they actually control. Delivered on site or remotely, built around your website and your competitors rather than generic examples.
Who this is for
Companies with a marketing team already in place, where the constraint is not budget but knowledge. If you have people writing content, managing the website and handling PR, they are the ones who will either make you visible to AI systems or not. Training them is faster and cheaper than hiring an agency to work around them.
It is also for firms who have decided this matters and do not want to be permanently dependent on an outside consultant to maintain it.
The curriculum
Core fundamentals: LLMs versus Google
Executives and marketing leadership
- How language models actually retrieve — embeddings, vector search and retrieval-augmented generation, explained without the maths
- How ChatGPT, Google AI Overviews, Gemini, Claude and Perplexity each choose their sources, and why they disagree
- Audit methodology: reverse-engineering a prompt to see why a competitor is being cited and you are not
Technical execution: schema and crawler policy
Developers and web team
- Entity schema in practice — writing JSON-LD that defines your company as one entity rather than several similar ones
- llms.txt: what the proposed standard is, who currently reads it, and whether it is worth your time
- Crawler policy: deciding which AI bots to allow for citation and which to block, and configuring robots.txt to match
Content engineering
Writers and content team
- Writing for chunking — structuring pages so each section stands alone as a retrievable passage
- The direct-answer pattern: putting the answer in the opening sentences, in the words people actually use
- Proprietary data: original research, case studies and figures that exist nowhere else, so a model has to cite you to use them
Off-page grounding and sentiment
PR, social and communications
- The sources models actually read — review platforms, directories, Reddit, YouTube, industry press — and which matter in your sector
- Digital PR aimed at citation rather than backlinks, so your name attaches to the terms buyers search
- Measuring reputation as a model sees it, not as a dashboard reports it
What it costs
Full programme
All four modules, each built around your site, your competitors and your category. Includes the baseline measurement and every artifact your team leaves with.
Roughly two days of delivery, run consecutively or spread across a month.
Single module
One module, one team. Useful when only one part of the organisation needs it — most often Module 2 for a development team already committed to the work.
Four taken separately come to $16,000, so the programme saves $3,500.
Delivered remotely or on site in Los Angeles at no additional cost. Travel beyond that is billed at cost, never marked up.
If your budget is below this, do not buy training. There is a list of five things worth doing yourself first, and a team that has done them will get considerably more out of a session later.
How it runs
Each module is roughly half a day. They can run consecutively over two days or spread across a month, and only the relevant team needs to attend each one — your developers do not need Module 4, and your PR team does not need Module 2.
Every module uses your own site, your own competitors and your own category. Nobody learns anything from a generic example about a fictional coffee shop. By the end of Module 1 your team will have seen exactly what AI assistants currently say about you, and which sources produced it.
What your team leaves with
A documented baseline of where you stand across each assistant. A written entity definition for your company. A schema implementation on your own site rather than a slide about schema. A content pattern your writers can apply without supervision. And a measurement routine, so the work continues after I leave.
Start with the measurement. Before any training is agreed, the free audit shows what AI systems currently say about your business and which sources they used. If the answer is already good, I will tell you that.
Run the free auditBooking a programme
Training runs on a fixed date, so it needs a conversation first — which teams are attending, what you want them able to do afterwards, and whether the work between modules needs supporting.
Talk it through. A short call to scope the programme and agree dates. No proposal deck, no discovery process — you will know by the end of the call whether this is right for your team.
Book a call 323-657-7752What this does not include
No guarantees of placement in AI answers. Nobody controls what ChatGPT or Gemini recommends, and any trainer telling you otherwise is selling something they cannot deliver. What is teachable is the evidence those systems read, and how to keep it consistent, corroborated and structured.
It is also not a certification. There is no accrediting body for this discipline yet, and a certificate from one consultant would be worth exactly what it sounds like.
Common questions
How much does LLMO training cost?
The full four-module programme is $12,500 and a single module is $4,000. Four modules taken separately come to $16,000, so the programme saves $3,500. Delivery is remote or on site in Los Angeles at no additional cost, with travel beyond that billed at cost. Businesses with a budget below this range are better served doing the foundational work themselves first.
What does LLMO training cover?
Four modules. Fundamentals for executives: how language models retrieve information, how each AI assistant selects sources, and how to audit why a competitor is cited instead of you. Technical execution for developers: entity schema, the llms.txt proposal, and crawler policy. Content engineering for writers: structuring pages for retrieval, the direct-answer pattern, and building proprietary data worth citing. Off-page grounding for PR: the third-party sources models actually read, and digital PR aimed at citation rather than backlinks.
Who should attend LLMO training?
Each module is aimed at a different team, and only the relevant people need to attend. Module 1 is for executives and marketing leadership. Module 2 is for developers and the web team. Module 3 is for writers and content staff. Module 4 is for PR, social and communications.
How long does the training take?
Each module is roughly half a day. They can run consecutively over two days or be spread across a month depending on how your teams are structured and how much implementation you want between sessions.
Is llms.txt required for AI visibility?
No. llms.txt is a proposed standard, not a requirement — no major AI company mandates it and adoption is partial. It is covered honestly in Module 2, including whether it is worth implementing for your particular site. Anyone describing it as mandatory is overstating its current status.
Can you guarantee our brand will appear in AI answers after training?
No, and nobody can. No one controls what an AI assistant recommends. What can be taught and influenced is the evidence those systems read: consistency, corroboration, structure and relevance. Any trainer guaranteeing placement is selling something they cannot deliver.