Plenty of websites don't have a content problem — they have a problem Google can't read. And Google alone isn't enough any more: answers now come from ChatGPT, Perplexity and AI Overviews, which only cite what they find on clean, well-structured pages. We fix the technical SEO that both depend on, and run Meta and Pinterest campaigns to cover demand while organic grows.
Relying on ads alone means renting your visibility: the day the budget stops, so does the traffic. SEO builds something that stays — a well-ranked page keeps bringing visits long after it was written.
We run technical SEO, content strategy and paid campaigns as one plan, on an honest timeline: 12 to 18 months for organic to carry a steady share of demand. Campaigns cover that demand in the meantime, and as organic grows, the budget shifts. And never a technique that puts your site at risk of a penalty.
Core Web Vitals, crawling, hreflang, structured data, dynamic sitemap, server log analysis. The part Google measures and that depends only on us.
Analysis by search intent, competitor comparison, content architecture around topic clusters and pillar pages.
A brief for every page that matters, on-page optimisation, deliberate internal linking and a review of what's already published.
Being cited by generated answers — ChatGPT, Perplexity, Google's summaries. Substantial content served as HTML readable without JavaScript, structured data that says who you are, and fully composed pages ready for the models' crawlers.
Campaigns on Feed, Stories and Reels. Retargeting by level of interest, lookalike audiences and systematic creative testing.
The channel where people search with images before they know which brand they want. Sponsored pins and shopping, with a content lifespan far longer than any feed's.
A live dashboard with rankings, traffic, conversions and return on spend. Every month, an analysis with the actions in priority order.
Technical site analysis, crawl report, study of the organic competitors, quick wins and long-term opportunities. The output is a roadmap in priority order.
A map of searches for every stage of the buying journey, content architecture, competitor comparison. We decide what to create, what to revise and what to remove.
Fixes on the site, structured data, loading times, crawl errors. In parallel the first campaigns start and cover demand in the meantime.
Optimised content, digital PR for authoritative citations, editorial relationships. Only techniques that survive an algorithm update.
Data analysis, landing page tests, refreshing content that's declining, opening new topic clusters once the first ones are consolidated.
The most honest way to show technical SEO is to let you inspect it. This site serves crawlers HTML that reads fine without JavaScript — service pages and articles arrive complete, so ChatGPT, Perplexity and Bing read the same content you see. It declares who we are through a structured-data graph, publishes an llms.txt file for language models, and handles Italian and English with correct hreflang.
To check, request a page the way a crawler does — for example with curl and GPTBot's user agent — and read what comes back.
A free analysis of where you rank today against your competitors. In half an hour you'll know where you stand, what's missing and what to fix first.
For the first two months little moves: the technical brakes are coming off, and the graph is flat by construction. Between month three and four the first positions arrive on specific, longer queries; between month five and eight the topic clusters start pushing each other and traffic becomes a curve rather than a series of jumps. Anyone promising page one in thirty days is talking about a term nobody searches.
With the technical side, always — not because it matters more, but because without it content does not matter at all. A site that answers in five seconds, refuses to be indexed or duplicates itself across three addresses can publish excellent articles that nobody sees. Once the base is fixed the ratio inverts and the work becomes almost entirely editorial.
It is worth it differently. Generated answers reduce clicks on simple questions — the ones a single sentence settles — and leave them nearly intact where there is a decision to make or something to buy. On top of that, models cite sources, and being cited needs everything it always did plus two things: substantial content in HTML readable without JavaScript, and structured data saying who you are. It is why this site serves AI crawlers a fully composed page.
Because they cover two different moments. Search — Google, and increasingly AI-generated answers — catches people already looking for what you sell: demand that exists and only needs holding. Meta and Pinterest work earlier, on people not looking for you yet, and create the demand search later collects. Held together, the creatives that work on social tell you which messages deserve an organic page, and the queries that convert organically tell you which audiences are worth building. Run by two different suppliers, that information never crosses between them.