casestudybook
We write your case study and build it so that Google finds it and ChatGPT cites it. You get it in two days.
✓ Refund if you're not convinced · 48-hour delivery · one revision included
An SEO and GEO case study is the account of real work, written so two different systems can find it: search engines like Google and generative engines like ChatGPT and Perplexity. It carries numbers, quotes with a name attached, results anyone can check. Those are exactly the elements an AI uses to decide who to name.
SEO, search engine optimisation, is the set of choices that make a page findable on a search engine: content, heading structure, speed, structured data, links pointing at you.
What you're after is a position: showing up in the list when someone searches for what you do, and taking the click.
GEO, generative engine optimisation, is the set of choices that make content citable by a generative engine: ChatGPT, Perplexity, Gemini, the AI Overviews inside Google.
There's no list to slip into here. Either you get named inside the answer, or as far as the reader is concerned you don't exist.
SEO gets you a click. GEO gets you a recommendation, which carries far more weight: the reader treats the AI as a third party, not as your sales page.
One thing you'll hear often that is simply wrong: GEO does not replace SEO. To build an answer, generative engines go and retrieve pages from the web, and a page that doesn't rank on Google frequently never enters that retrieval at all. Classic visibility is still the price of admission. A case study works on both fronts because it has the right structure for each.
The first is your website, where they find what you wrote. The second is a chat with an AI, where they ask who's worth talking to for a job like theirs.
On that second route, what surfaces is whatever the market has left on record about you. If there's nothing there, the shortlist closes before you know you were in the running.
Getting into that conversation doesn't mean publishing more. It means publishing the format AI systems prefer when they have to name someone.
They cite whoever brings proof: numbers, attributed quotes, results you can check. That's a case study, plain and simple. We build yours so those levers are there in the shape an AI recognises, and so Google ranks it.
Research on how generative engines behave tells us which elements raise the odds of a piece of content ending up inside an answer. They're the same elements a case study is built on.
Sources: Generative Engine Optimization research, Princeton, Georgia Tech and Allen Institute (KDD 2024) · Omniscient Digital analysis of 23,387 citations · NetStrategy analysis of 190 Italian websites and over 35,000 queries to generative engines.
If an AI reassembles most of what it says from sources you don't control, a case study locked inside your own site is working at half capacity. So we hand it over ready for LinkedIn too, and ready for the company you name in it to publish its own version.
None of them are marketing agencies. They're product companies selling to other businesses, in technical sectors. We linked straight to the case, not the homepage: open it and look at the structure, not the design.
In the Armal case you read a 40% cut in machinery energy costs and payback on the investment in four months. An AI pulls sentences like that out exactly as written. Quotes carry a full name and job title, never a vague "the client".
→ open the case study
Every case opens with results in a bullet list, above the prose: 230 million process instances a day, visibility across 800 processes, 250 decision tables. It's the most extractable format there is, because nothing needs interpreting.
→ open the case studies
Cases are split by application: transport, defence, industrial plants, healthcare, smart cities. That's how you catch vertical questions, the ones where a buyer describes their own world instead of searching for a technology name.
→ open the case studies
Use cases are indexed by application rather than dumped into a single container. Someone searching for their own problem lands on the case that solves it, not on a homepage where they get lost.
→ open the use casesFour different details, four different companies. The case study we hand you has all four: numbers inside the text, results up front, a vertical angle on your sector, a dedicated page per case.
Who the client was, what you did, what changed for them. We send the questions, you answer those.
For Google and for AI answers, with the SEO and GEO kit already built in.
Ready to publish, with a round of revision still available to you.
And if the result doesn't convince you, we refund you.
GEO stands for generative engine optimisation. It's the set of choices that make content citable inside answers from ChatGPT, Perplexity, Gemini and Google's AI Overviews. Unlike SEO, where the goal is a position in a list, here the goal is being named inside the answer itself.
It's still needed, and that isn't a matter of opinion. Generative engines build answers by retrieving pages from the web: a page that doesn't rank on Google very often never enters that retrieval. GEO leans on SEO, it doesn't replace it.
It's a case study written so a generative engine can extract and cite it: clear structure, verifiable numbers, quotes with a name and role, headings that make sense read out of context, and sections that answer the questions your buyers actually put to an AI.
48 hours from the moment we have the complete brief. One round of revision is included.
Who the client was, what you did, what results you delivered. If you have the numbers, better. If you don't, we reconstruct them together during the brief.
The method doesn't change from sector to sector. What changes are the keywords and the questions your buyers ask, and we calibrate those on your case.
It depends on your professional body, and in some cases the standard format naming the client isn't workable. There is an anonymised version, with no comparative language and no client references, that keeps the structure Google and generative engines need. Flag it at brief stage and we work on that.
You publish it, and from there it starts working on Google and inside AI answers. The next step is having the results confirmed by the other party involved, so the proof stops being just your own claim.
Give it something to cite.
Get your case study