Basic
$199/moBelow DA/DR targetNot a fitExcluded from recommendation
AI SEARCH & BRAND DISCOVERABILITY
SITETRAIL CERTIFIED
Your learning path: 12 lessons.
Module 01 / Lesson 02 · Read, apply, check
Explain training, retrieval and answer generation without claiming access to a model’s hidden decisions.
Training changes a model through a development process using data; retrieval supplies information for a particular task or answer. An AI product may combine learned knowledge, search results, conversation context and other tools. The visible answer does not expose the contribution of every component. A newly published page appearing in a sourced response is evidence of that observed response, not proof that the foundation model was retrained. Explain this distinction to clients before selling a content programme. It prevents a common misunderstanding in which ordinary website work is described as directly rewriting what a commercial AI system permanently knows.
Different products and modes can use different retrieval processes and display different source information. Some questions trigger a search; others may be answered without visible web activity. The conversation history, language and location can also affect what is relevant. Record the actual product, mode and visible search indicators rather than assuming all AI answers come from one common index. A result from a developer API is not automatically a measurement of the consumer interface. When a platform does not reveal a setting or version, record it as unknown. A precise limitation is more useful than a guessed model label.
OpenAI distinguishes OAI-SearchBot, used for search, from GPTBot, associated with possible foundation-model training; their controls are independent. That distinction allows a site owner to discuss search access separately from training preferences. Google’s AI-search guidance also connects supporting-link eligibility with ordinary search accessibility and indexing requirements. Eligibility does not guarantee selection. Ask the responsible web team to review the relevant platform’s current documentation and the client’s intended policy. Do not change crawler controls casually or tell a client that allowing a bot compels a system to cite them. Access is a prerequisite question, not a purchased outcome.
Use a practical chain: information exists, a system may access or retrieve it, the answer may use it, and the interface may display a source. Each step can fail or remain unobservable. A server request can show access under the logging assumptions; it does not prove the content shaped a later answer. A citation can show a displayed source; it does not expose every retrieved page or the model’s internal reasoning. Build recommendations around the steps the agency can improve, such as clear evidence and accessible pages, while keeping the selection process outside the service guarantee.
Make useful, accurate information available.
Check relevant site and platform requirements.
Information may be selected for a particular task.
Record the visible answer and source, not hidden assumptions.
Fieldmark publishes a current comparison of planned and reactive maintenance, supported by an engineer’s explanation and a dated methodology note. A week later, an AI search answer cites the page. The client asks whether the agency has now trained the model. The account lead explains that the observed citation is consistent with the page being used as a source in that answer, but it does not establish a training update. The web team separately reviews access settings and finds that search access and training preferences were being discussed as if they were one switch. They document the client’s intended choices without changing unrelated controls. The agency records the product, mode, prompt and date, then continues observing relevant questions. It has a useful result to report and a clearer technical conversation, without making a claim about hidden learning processes.
A competing agency sells a service called permanent AI memory and promises that publishing ten articles will force every major assistant to remember the brand. It uses one favorable answer as proof and ignores different results in other modes. The claim combines several systems and mechanisms without evidence. If the answer changes next week, the agency has no credible explanation because its promise depended on control it never possessed rather than on a defined programme of content and observation.
Crawl access, indexing, retrieval, citation and recommendation are distinct conditions. Improving one can be useful without guaranteeing the next. Avoid both extremes: claiming complete control over AI answers or assuming nothing can be improved. Agencies can strengthen public information and measure visible responses while remaining honest about the parts they cannot inspect.
Platform documentation and access controls evolve. Verify current guidance with the responsible site owner before making changes. This lesson explains concepts, not instructions to alter a client’s crawler policy without authorization. Do not infer private training data, hidden ranking factors or permanent model changes from isolated outputs.
Do this today · 30 minutes
Explain Fieldmark’s new citation to a client who thinks it proves retraining.
Use a document or spreadsheet and the fictional case in this lesson. Work with public or synthetic information; no purchase or confidential client records are required.
Before: Make useful, accurate information available.
After: Record the visible answer and source, not hidden assumptions.
Notes stay in this browser. Download a copy to keep them.
Ask a colleague to challenge the interpretation. Record the evidence, uncertainty, next action, owner and review date before applying the plan to a client campaign.
Before you move on
A client’s new page appears as a source in one AI search answer. The client says this proves your campaign retrained the foundation model and wants the same outcome guaranteed everywhere. What explanation and next action are most defensible given the evidence you can actually observe?
Next lesson →OpenAI: search and training crawler distinctions
Google Search Central: AI features and website eligibility
Lesson edition 1.0 · References reviewed 16 October 2026 · Practical exercise reviewed 16 October 2026. Examples are illustrative teaching scenarios.
Sitetrail service · NewsPass
Put your SEO, PR and generative engine optimization strategy into practice. NewsPass supports search visibility, media presence, brand credibility and discoverability in AI answers.
Compare the delivery costs for your campaign using the same calculator as the Sitetrail homepage. Choose a target authority range, your typical placement cost and expected monthly volume.
Illustrative cost comparison using October 2026 plan prices; check current terms before purchase. Assumes one eligible placement per credit and comparable selected opportunities, not a prediction of rankings, revenue, coverage or AI citations. Actual suitability depends on audience fit, editorial requirements and the current plan terms. DA/DR are third-party metrics, not Google scores.
Choose publications available on your plan, prepare useful content for the audience and submit it for editorial review. Track accepted placements in your dashboard and evaluate them against your campaign goals.
Clear evidence. Useful context. Better decisions.
Use the worksheets to turn each lesson into a real decision. Build a credible campaign as you learn from buyer questions, original evidence and repeatable observations.
Course outline · Free assessment · Privacy
No account is required. Progress, first-attempt lesson answers, worksheets and assessment results are stored in this browser when storage is available. They do not sync across devices. Download copies to retain them; clearing browser data removes local records.
Your free trial starts automatically. No license key is needed yet.