AI-Readiness & Generative Engine Optimization

What we do

We make your website legible, citable, and quotable to AI engines. That means a focused engagement on five concrete fronts:

Render and crawl. We make sure your most important pages — catalogues, articles, location pages — are visible to AI crawlers in the server-rendered HTML, not hidden behind client-side JavaScript that crawlers do not execute.

Structured data. We add the machine-readable JSON-LD that AI engines lean on when they decide whether to cite your business: LocalBusiness, Organization, Product, Offer, AggregateRating, FAQPage, BreadcrumbList, Article. The right schema, on the right pages, with verified values.

AI-discovery surfaces. We publish llms.txt and llms-full.txt — the emerging conventions for telling LLMs which of your URLs you most want them to read, and providing the authoritative full text for those URLs.

Content amplification. We audit your existing knowledge base, blog, and product pages against what AI engines actually quote — and recommend (or write) the specific, opinionated, story-driven pieces that get cited at rates an order of magnitude above generic SEO copy.

Measurement. We baseline how often your business is mentioned today across ChatGPT, Claude, Perplexity, Gemini, and Google's AI Overviews, against a representative set of queries your customers might ask. We re-baseline at thirty, sixty, and ninety days.

What you get

A typical engagement produces:

• A written audit report, prioritized by impact and effort, suitable for your team or any other agency you might bring in.

• All production code changes: schema, rendering, alt text, contact surfaces, AI-discovery files. We ship them in your stack — Next.js, Astro, Nuxt, plain HTML, whatever you have.

• A baseline measurement report and a measurement playbook your team can run going forward without us.

• A one-time pass over images, meta descriptions, and FAQ content using vision and language models, with your team in the editing loop.

• An optional one-page case study you can use, if you want one.

What we don't do

We do not write SEO blog spam, build link farms, run paid campaigns, or sell dashboards with confident percentage-change graphs that nobody can verify. AI-engine measurement is still a deliberate, partly manual practice in 2026. Anyone promising you otherwise is selling a story.

We also do not take on engagements where the underlying content is genuinely weak. AI engines will not cite a generic business no matter how clean its schema is. If your voice and your story aren't where they need to be, we will say so, and we may be able to help with that as a separate piece of work — but we will not paper over it with technique.

How we work

Week 1 — Audit. We pull your site apart, by hand, with the technical and editorial detail it deserves. You receive a written report at the end of the week.

Weeks 2–4 — Implementation. We ship the changes in your stack. You see daily progress. Your team is welcome alongside if you have one, in which case we mentor as we go.

Week 5 — Baseline and measurement plan. We run the query baseline, hand over the measurement playbook, and document everything.

Day 30, 60, 90 — Re-baseline. We check what changed. If something didn't move, we explain why and what to try next. This is included.

Who this is for

• Premium boutique businesses with strong story and weak technical surface.

• B2B SaaS companies whose product is increasingly being recommended (or not) inside AI answers to comparison queries.

• Knowledge-driven businesses — law firms, consultancies, specialist e-commerce — whose published thinking is excellent but invisible to AI engines.

This is not for businesses without a real voice yet, or for very large enterprises with their own in-house SEO teams who need a vendor to take orders. We are small, senior, and selective.

Pricing

Engagements start at €6,000 for a four-week audit-and-implementation sprint, with an optional monthly retainer for ongoing measurement, content amplification, and iteration starting at €1,200 per month.

We are taking on a limited number of engagements through the second half of 2026 as we re-engage publicly after a deliberate pause to retool around AI-native delivery. If you would like to talk, write to paul@gorazdo.studio.

Frequently asked

Is GEO the same as SEO?

No. The two overlap by roughly sixty percent — both reward fast pages, clean structure, and authoritative content. The remaining forty percent of GEO is its own discipline: server-rendering for AI crawlers, structured data weighted for AI grounding, llms.txt, opinionated content built to be quoted rather than ranked, and a measurement practice that does not yet have a dashboard. We respect SEO agencies and work alongside them. We do not replace them.

Will I rank #1 in ChatGPT?

Nobody can promise that, and anyone who does is misleading you. What we can promise is the inputs (schema coverage, render fixes, AI-discovery files, FAQ structures, descriptive alt text), a measurement plan that tracks the right signals, and honest reporting on what moved.

Do you work in our stack?

We are Next.js and React specialists by background, but the principles transfer. We have shipped this kind of work in Astro, Nuxt, Eleventy, WordPress, Shopify, and plain server-rendered HTML. If you are on something exotic, write to us and we will tell you honestly whether we are the right people.

Can you help with multi-language sites?

Yes. AI engines treat languages as separate citation surfaces — meaning a German-language LocalBusiness block and a German-language FAQ section are not redundant copies of the English versions but distinct opportunities. Our work on internationalization (hreflang, locale-aware schema, translated knowledge bases) is part of any engagement where it applies.

What about accessibility?

Accessibility and AI-readability are the same project in different clothes. The semantic HTML, the heading hierarchy, the form labels, the alt text, the ARIA landmarks — every one of these helps a screen reader, and every one of them helps an LLM parse your page. We use Playwright with axe-core to test continuously in CI, and accessibility passes are included as part of the standard engagement.