Founders, investors and developers now ask ChatGPT and Perplexity "which project does X?" before they open Google. We make sure your name is in that answer, with the press to back it up.
AI SEO for a Web3 or AI startup means getting the project named when someone asks an AI assistant about its category: "best restaking protocols", "decentralized inference networks", "which wallet SDK supports MPC". It works when two things line up: tier-1 coverage in outlets the models trust (CoinDesk, Cointelegraph, The Block, Decrypt, Forbes) and answer-first pages on your own site that repeat the same facts in a form models can quote. Shilika Jain runs the first. Rahil Jain builds the second.
A pattern we see often in crypto and AI. The team has landed real press, but when you ask ChatGPT about their category, a competitor with worse coverage gets named instead.
Docs and landing pages built as single-page apps. GPTBot and PerplexityBot fetch the HTML, find almost nothing, and move on.
TVL on the site says one thing, the CoinDesk piece another, CoinMarketCap a third. Models skip brands whose numbers don't match across sources.
There's a homepage, docs and a blog of announcements. There's no page titled the way buyers actually ask: "X vs Y", "how does Z work", "best tools for W".
Comparison pages, explainers, "best of" lists and glossary entries for the prompts your buyers type. 44% of ChatGPT answers cite list-style pages, so we write a lot of them.
CoinDesk, Cointelegraph, The Block, Decrypt, Blockworks, Forbes and AI Magazine. Each placement is pitched with the same three facts your answer pages use.
Server-rendered landing pages, Organization and SoftwareApplication schema, FAQ blocks, llms.txt and a clean path for AI crawlers through your docs.
Op-eds, podcasts and consistent profiles on LinkedIn, X, Crunchbase and the byline outlets, so the founder resolves as one person that models can attribute quotes to.
Reddit, YouTube, X Spaces and the aggregators (CoinGecko, CoinMarketCap, DefiLlama, Product Hunt, Hugging Face) all carrying the same story and numbers.
A fixed set of buyer and investor prompts, run across ChatGPT, Perplexity, Gemini, Claude and Google AI Mode. You see who gets named, who gets cited, and how that moves month to month.
The press side of the proof: Gaia in Forbes, Decrypt and Benzinga in five months. MANTRA's $11M raise in CoinDesk. RARI Chain's mainnet in The Block. Each of those articles is now a source an AI engine can cite. Read all seven case files.
Already working with me on Web3 PR, token launch PR or AI startup PR? AI SEO plugs straight into the same calendar, so every placement gets a matching answer page the week it lands.
Sometimes, but slowly. Branded mentions are the strongest predictor of showing up in Google's AI Overviews (0.66 correlation in Ahrefs' 2026 study of about 75,000 brands). Mentions without matching pages on your own site tend to get your competitors cited instead, because the model finds a cleaner answer elsewhere. PR plus answer pages is faster than either alone.
Only if the HTML has the content in it. Many docs and landing pages render client-side, and most AI crawlers don't execute JavaScript. We check what GPTBot, PerplexityBot and Google-Extended actually receive, then fix the render path or publish server-rendered versions of the pages that matter.
We write the same way we write press: no price talk, no return projections, no investment framing. Answer pages explain what the protocol does, how it compares and who uses it. That's also what AI engines prefer to quote.
The same three plans as every AI SEO client: $1,000 one-time for the audit and strategy, $3,000 per month for Strategic Execution (100 AI-feed pages and 25 branded placements), or $4,000 per month for Deep Execution (200 pages and 40 placements). A full PR retainer is scoped separately. Plan details.
A 30-minute teardown: we run your category prompts live, show who gets cited, and map the first 20 pages and placements we'd ship.