Frequently Asked Questions
Common questions about Fodda, knowledge graphs, MCP integration, pricing, and expert AI context.
How is Fodda different from Mintel, GWI, Gartner or WGSN?
Mintel, GWI, Gartner and WGSN are designed for humans to read. Fodda is designed for AI agents to query. Most teams that try Fodda keep their existing research subscription and use Fodda to ground the AI work they're doing alongside it. The two solve different parts of the same problem.
Where does the data come from?
Three sources. First, PSFK's editorial pipeline, which has been publishing trend research since 2004 and refreshes monthly. Second, 80+ institutional sources across 30+ countries, including Census, FRED, BLS, BEA, PubMed, Pew Research, Google Trends, World Bank, WTO, OECD, Semantic Scholar, plus country-specific central bank and statistical office data from UK ONS, Eurostat, ECB, Bundesbank, INSEE, CBS Netherlands, Statistics Canada, Bank of Japan, and more. Third, partner graphs built by named industry analysts. Every node in the graph carries an evidence list pointing to the source material it was built from.
How often is the data updated?
PSFK graphs refresh monthly. Institutional data sources refresh on the schedule each source publishes to. Earnings intelligence updates live as filings publish. Many nodes carry a last-updated timestamp so you can verify freshness.
Will my AI hallucinate when querying Fodda?
Fodda researches your question against the sources you've connected and cites each item it used. Hallucination happens when a language model invents content. Fodda removes the invention by grounding every answer in retrieved evidence. The same discipline applies to Human Agents: if a question is outside the expert's lane, it says so and can refer you to a colleague's Human Agent.
Is Fodda just PSFK with extra steps?
No. PSFK is the founding knowledge partner and supplies a large chunk of the editorial content. Fodda also includes 80+ institutional data sources across 30 countries, partner graphs from named analysts outside PSFK, synthetic analyst personas, and the option to bring your own internal research. PSFK is the anchor, not the whole catalog.
What industries are covered?
Live today: retail innovation, beauty and culture, sports and media, consumer electronics design, and broader innovation and culture. We add verticals based on demand. If you need a vertical we don't cover yet, tell us and we'll let you know where it sits on the roadmap.
What does it cost?
Free to start, no credit card. Paid plans scale by usage.
How do I get started? Do I need a credit card?
No credit card. You enter an email, get an API key, and can immediately connect to Claude or the other supported tools.
Do I have to leave Claude or Copilot or Gemini to use Fodda?
No. Fodda lives inside the AI tools you already use through the Model Context Protocol. You install it once as a connector and it shows up alongside your other tools.
Is my prompt data private? Where is it processed?
Your prompts are routed to Fodda's connector, which runs the research across the expert sources you've connected and returns cited results. We log query metadata for usage tracking and don't share prompts with third parties or use them to train models.
Can I add my own research into Fodda?
Custom dataset ingestion is available on a consultative basis and on our enterprise roadmap. You can work with us to turn your private research reports, decks, and internal trend analysis into a secure, private expert collection. Your private collection uses the same MCP connection as the public ones.
What's the relationship between Fodda and PSFK?
PSFK was founded by Piers Fawkes in 2004 and has been publishing trend research since then. Fodda is built by Piers and uses PSFK's archive as its founding expert collection. PSFK is one supplier among several, but a substantial one.
Why is it called Fodda?
Fodder. As in, what you feed your AI. Pronounced FOD-uh.
What is a Knowledge Graph Marketplace?
A Knowledge Graph Marketplace is an open, queryable ecosystem where named industry experts, research organizations, and institutional data providers publish machine-readable graph schemas. AI agents can discover, price, and query these knowledge graphs on demand via standard protocols like MCP and REST API.
How do AI agents pay per query using Shared Payment Tokens (SPT) and HTTP 402?
Fodda implements autonomous pay-per-query via Shared Payment Tokens (SPT) and HTTP 402 Payment Required responses. When an unauthenticated AI agent makes a request, the server emits a 402 status with token pricing (/bin/zsh.50 via SPT). The agent attaches a signed payment token in headers, paying only for the exact query without requiring a user subscription or account setup.
How do I connect Fodda to Claude, Copilot, or my IDE (Cursor, Windsurf, Claude Code)?
Connecting takes 60 seconds via Model Context Protocol (MCP). In Claude.ai or Claude Desktop, go to Settings → Connectors → Add Custom Connector and paste https://mcp.fodda.ai/c/:token (leaving OAuth blank). In Cursor or Windsurf, add the MCP server URL to your workspace mcp.json config.
What's the difference between a Human Agent, a Synthetic Analyst, and a Synthetic Executive?
A Human Agent (Digital Twin) is modeled on a real human expert and their specific archive. A Synthetic Analyst is an AI consultant built on data across multiple sources, not modeled on a specific person. A Synthetic Executive (Synthetic C-Suite) is built on earnings calls, real speeches, and company data to represent a brand's leadership perspective.
How is this different from just using ChatGPT's web search or Claude's web browsing?
Web search is broad and unfiltered. Fodda is narrow and editorially curated. When you ask Claude about checkout innovation, web search pulls whatever's been published recently across the internet. Fodda plans which sources to check (expert collections, institutional data, earnings), runs the research against material retail researchers have spent years building, and shows you the plan and citations. You get fewer answers, more relevant ones, and you can audit where each one came from.