Until recently, ranking on Google’s first page was enough. Today half the answers come straight from an artificial intelligence, and if your company isn’t written in a way the AI can understand and cite, you simply don’t exist. We do both jobs.
Google and Bing send a person to your site. You win by being technically clean, fast, topically coherent and by answering the question better than anyone else.
ChatGPT, Gemini, Perplexity and AI Overviews answer instead of you. You win if your content is easy to extract, if you state clearly who you are and what you do, and if what you say is verifiable enough for the AI to feel safe citing it.
We comb through the site: real speed on mobile networks, crawl errors, orphan pages, duplicate content, redirect chains, heavy images, missing or wrong structured data. You end up with a list of fixes ranked by impact.
Not just classic keywords: also the whole questions people type into ChatGPT, which are longer and more specific. They are two different lists and they need different content.
Pages built around a real question, with the answer in the opening paragraphs, concrete data and no padding. It is what Google rewards and exactly what a language model can cite.
Organization, LocalBusiness, Service, FAQ, Breadcrumb, ItemList: complete, consistent markup so engines and AI know without ambiguity who you are, where you are and what you sell.
Descriptive file names, alt text written for a human, correct dimensions, modern formats, a dedicated image sitemap and captions that explain the context. Images bring traffic if you treat them as content.
One URL per language, reciprocal hreflang, x-default, clean canonicals, genuinely translated content. This is where almost every multilingual site gets it wrong and loses half its foreign traffic.
We tell you where the visits come from, which pages convert and which don’t, and what is worth doing next month. No forty-page reports that nobody reads.
The site you are reading is static, uses no profiling cookies, loads in under a second, has five complete language versions with reciprocal hreflang, structured data on every page, an llms.txt file for language models, separate sitemaps for pages and images, and every image named and described. If you found us here, it worked.
Traditional SEO aims to get your link into a list of results and get it clicked. AI optimisation — some people call it GEO, Generative Engine Optimization — aims to make a language model use your content to build its answer and cite your company as the source. The content format, the data markup and the trust signals required are all different.
Technical and speed work is felt immediately. Ranking on competitive terms usually takes three to six months of consistent work. On AI answers the timescale is often shorter, because models refresh their sources frequently and reward clear, well-marked-up content.
You need content that answers your customers’ real questions. If that takes the shape of a blog, fine; if it takes the shape of well-written service pages, equally fine. Publishing to fill space achieves nothing and for some years now has actively hurt.
Yes, it is a large part of what we do. We start with an audit, show you what is wrong and what fixing it is worth, then either do the work ourselves or guide whoever maintains the site.
Yes, and they are one of the most neglected areas. An image with a descriptive file name, well-written alt text, the right format and a place in a dedicated sitemap brings visits from image search and helps AI models understand what the page is about.
Free initial analysis, no strings attached. We reply within one working day, in English, Italian, German, French or Spanish.
