A course built from citation cases
I am Dorian Vale, a Netherlands-based teacher of AI visibility for agencies that work across languages. My work sits where multilingual SEO, B2B content structure and answer-engine citation behavior meet. I built this course for Dutch agency teams that already understand audits and briefs, but need a clearer habit for reading cited AI answers.
Dorian Vale
I teach students to read the answer before they rewrite the page, because the citation usually shows the problem first.
In one teaching case, a French software page sat open on my screen with all the usual signs of competence: clear service language, decent headings, a recognizable company name, and no obvious crawl disaster. Yet Perplexity kept citing a thin directory instead. The directory had almost no depth, but it gave the system one tidy sentence about what the company did. The client’s own page made a better human argument and a worse machine quotation. That small irritation became a useful teacher.
I am from the Netherlands, and I came into this work through multilingual B2B search projects. For years I edited technical service pages, reviewed content architecture, and helped agency teams explain complex offers across borders. Dutch agencies are used to this problem in a practical way: the client is in one market, the evidence may be in another language, and the buying committee reads with different expectations than the search system. When answer engines entered the workflow, the question changed. A page had to rank as a destination and also survive being selected, quoted, compressed and compared.
That is why I began treating Perplexity SEO as evidence work. I look at what the system can extract, what it is willing to cite, and where it quietly bends the company description into something simpler. I opened this course for Dutch boutique SEO agencies working on French-market B2B clients because they already have the base craft: audits, briefs, structure, reporting. What they need is a stricter reading habit for AI answers. In these lessons I start with small observable cases, then name the principle after the evidence is visible. It is slower than slogan-driven advice and leaves fewer loose screws.
I begin with a visible problem: a client absent from an answer, a citation going to the wrong source, a French page summarized through English evidence, or a follow-up question that exposes a weak content boundary. From there we separate three things that often get blurred together: what Perplexity can extract, what it can cite, and what it may distort while sounding confident. Each lecture keeps that order. We inspect the answer, compare the sources, identify the missing evidence, then decide what an agency can recommend. The aim is a disciplined habit of reading. Students leave with a clearer way to explain AI visibility problems to clients without pretending that citations can be forced.
Learn the method from the cases upward.
The course starts with Perplexity’s answer-and-cite behavior, then moves into audit, rewriting and agency workflow.