
Feynn
Feynn is a web-based strategic intelligence platform whose AI agents research a company, its competitors, suppliers and industry, then tie every claim to a dated source, so strategists can hand a board-ready briefing to the people who decide.
What is Feynn?
Feynn is a strategic intelligence platform published by Feynn AS in Oslo, Norway, and delivered entirely in the browser. Its promise fits in three words on the home page: Understand. Simulate. Decide.
You start by naming a company, your own, a client's or a competitor's. AI agents disambiguate the entity and build a sourced fact map around it: strategic objectives, competitive landscape, supplier ecosystem, industry dynamics. That base then feeds eight modules. Strategic Posture scores the organisation across 6 strategic domains and 24 metrics, plotting internal capability against market opportunity. Deep Research runs multi-source investigations, either Fast, around 90 seconds and 7 credits per entity, or Deep, around five minutes and 20 credits. The Decision Engine puts a scenario to five specialist agents, covering systems thinking, competitive forces, financial and risk, innovation and technology, and the human and organisational angle, before a Strategic Director AI reconciles their views into a board-level architecture. Game Theory models competitive interactions, Nash equilibria and probable rival responses. Signals watches world news for competitor pivots, supplier disruption and regulatory change. Industry Lens scores a sector across six sourced dimensions, Innovation Horizon maps emerging technology by maturity stage, and Strategy Studio turns the whole thing into a recommendation with rationale, risks and KPIs.
What the product insists on is traceability. Every non-trivial fact carries a publisher, a date and an extract, plus a confidence score and a recency indicator, drawn from regulatory filings, regulator pages, company websites, reputable press and sector analyst content. The publisher claims zero hallucinations, says unsupported claims are flagged rather than laundered, and accepts the corollary: it tells you when its evidence is thin instead of inventing depth. Several model and search providers sit underneath, including OpenAI, Anthropic, Google Gemini, Tavily, Perplexity and NewsAPI.
Everything runs on credits, priced per operation: onboarding 5, each competitor or supplier 1, Fast Research 7, Deep Research 20, a Signals search 1, a Decision Engine agent 4, the Director 8, Game Theory 5, Strategic Posture 7. The published path is deliberately linear: enter a company, get a verified briefing, monitor signals, stress-test moves, leave with a board-ready recommendation.
What it does
- Research a company, its competitors, its suppliers and its industry with AI agents that attach a source to each fact
- Generate a Fast Research brief in around 90 seconds (7 credits) or a Deep Research report in around 5 minutes (20 credits) per entity
- Score strategic posture across 6 strategic domains and 24 metrics, capability against opportunity
- Stress-test a scenario with five specialist agents, then have a Strategic Director AI reconcile them into one recommendation
- Model competitive interactions and likely rival responses through game theory
- Monitor market signals continuously: competitor pivots, supplier disruption, regulatory change
- Map an industry across 6 scored dimensions and track emerging technologies by maturity horizon
When to use Feynn / When not to
A quick filter to help you decide if Feynn is the right fit.
When to use Feynn
- Strategy leaders and heads of corporate development who must defend a move with evidence rather than opinion
- Consultants and advisers building client-facing work where every claim has to trace back to a publisher and a date
- M&A and corporate development teams running early-stage target, competitor and market diligence
- Procurement, sourcing and supplier-risk managers watching vendor disruption and industry concentration
- Founders and lean teams with no dedicated research analyst, described by the publisher as “founder-grade strategy without a founder-grade research team”
When not to use Feynn
- Teams whose decision turns on broker research, expert-call transcripts or licensed financial datasets: Feynn licenses none of them and says so
- Anyone seeking professional advice, since the publisher states that AI Output is not financial, legal or strategic advice
- Developers wanting to embed strategic research in their own product, as no public API documentation exists
- Mobile-first users, because the product ships as a web application only, with no iOS or Android app
- Occasional users hoping to keep researching for free, since the free plan grants 30 credits once and for life, not every month
How to use Feynn
A typical end-to-end flow, from setup to results.
- Read the sample report published at /sample/novalabs-strategy-studio, which needs no account, to see what a finished deliverable looks like
- Create an account with an email address and password, or sign in with Google, LinkedIn, GitHub or Apple
- Start without a payment card: the free plan requires none
- Complete onboarding by entering the company to analyse, 5 credits, which the platform disambiguates and verifies
- Add the competitors and suppliers that matter, 1 credit each
- Launch a Fast Research run, 7 credits, or a Deep Research run, 20 credits, on each entity
- Check the credit cost, which is shown before execution wherever possible, then review the report in the browser
- Chain the advanced modules: Strategic Posture, Decision Engine, Game Theory, Signals, Industry Lens, Innovation Horizon, Strategy Studio
- Export the finished report to PDF or Markdown for a deck, a memo or a board pack
- Set up saved monitors and recurring briefings to keep watching the entities you track
Pros & Cons
Pros
- Traceability is the product, not a feature: each non-trivial fact carries a publisher, a date, an extract and a confidence score
- The publisher commits to flagging thin evidence instead of filling the gap with confident prose
- Costs are predictable: the credit price of every operation is published and shown before execution
- Unused credits roll over up to twice the monthly allowance on paid plans
- A free plan with no payment card, and a sample report readable without an account
- Legal documentation unusually complete for a publisher this size: terms, privacy, legal, security and cookies pages, all dated 30 November 2025, with eleven subprocessors named
- Publisher established in the EEA under Norwegian law, no analytics or advertising cookies, and a stated policy of not training its own models on customer content
Cons
- No security certification of its own: the SOC 2 Type II and ISO 27001 badges shown on the security page belong to Lovable, the application host, and the publisher says so explicitly
- No hosting country or region is named, only “regional hosting options” at Supabase
- No customer DPA is published or offered; the data processing agreements cited are those between Feynn and its own subprocessors
- No documented way for a customer to opt its data out of AI provider training
- The free plan is a one-off, lifetime allocation of 30 credits, not a monthly quota, so recurring use requires a subscription
- Credits are deducted when a job starts, and a refund after a system failure is granted at the publisher's discretion
- A very young, very small company, incorporated in late 2025 with a single officer on the register, and no mobile app, public API, third-party integration or social presence linked from the site
Pricing & Plans
A free plan is available: 0 USD for a one-off lifetime allocation of 30 credits, with access to all platform features and no payment card required. The lowest paid entry point is the Pro plan at 20.00 USD per month for 200 credits per month, with further tiers at 40, 80 and 200 USD per month for 400, 800 and 2,000 credits respectively. An Enterprise plan is offered at custom pricing on request. There are no setup fees, subscriptions may be cancelled at any time, and payment is handled by Stripe.
- 0 USD
- described as free forever
- 30 credits granted once for life
- all platform features
- no payment card required
- from 20 USD per month
- marked most popular
- sold in tiers of 200
- 400
- 800 or 2
- 000 credits per month at 20
- 40
- 80 or 200 USD per month
- custom pricing and flexible billing
- with custom plans
- dedicated support
- custom integrations and advanced security
- available through Contact Sales
Data, GDPR & hosting
A consolidated view of how Feynn handles your data.
GDPR overview
GDPR is addressed explicitly and in unusual detail. Feynn AS is established in Norway, inside the EEA, so the regulation applies directly and no Article 27 representative is required or named. The terms state that the company processes personal data in accordance with applicable data protection laws, including the EU/EEA GDPR. The privacy policy names its legal bases by article: contract 6(1)(b), legitimate interests 6(1)(f), consent 6(1)(a) and legal obligation 6(1)(c). It lists the rights of access, rectification, erasure, restriction, portability and objection, withdrawal of consent, and complaint to a local supervisory authority. Transfers outside the EEA rely on Standard Contractual Clauses, adequacy decisions and the EU-U.S. Data Privacy Framework. Requests go to privacy@feynn.ai. Only strictly necessary cookies are used, hence no consent banner. No DPO is named, and Feynn AS claims no certification of its own.
Who owns the data?
The terms are explicit: you retain ownership of your User Content. In return you grant Feynn AS a limited, worldwide, non-exclusive, royalty-free licence to process, store and use that content solely to operate, maintain and improve the service. Rights over the AI Output run the other way and are handed to you: you may use, reproduce, modify and commercially exploit what the platform generates, with a single carve-out, namely that the output may not be used to train competing foundation models. The controller for the personal data attached to your account is Feynn AS, organisation number 936 609 961, Carl Jeppesens Gate 22, 0481 Oslo, Norway.
Reuse rights
You do not need permission to reuse what the platform produces. The terms grant you the right to use, reproduce, modify and commercially exploit the AI Output, the only restriction being that it may not be used to train competing foundation models, and reports export to PDF or Markdown so they can move straight into a deck or a memo. On the input side, Feynn collects your email address and password, the company details you type in, your prompts and research scenarios, and your support messages. It also captures technical data automatically: IP address in short-lived logs, device and browser metadata, authentication events and error logs. Those data serve to run the account, generate the AI results, process payments, diagnose problems, send notifications and meet legal obligations. Text you submit is sent to external AI providers for the sole purpose of producing the output; the publisher states that it does not use customer content to train its own AI models and that, where provider controls exist, it configures them to reduce retention and prevent training on customer content. It also states that it does not sell personal data. Social sign-in through Google, LinkedIn, GitHub or Apple passes only your email, name and account identifier, never a password. The policy has applied since 30 November 2025.
Data retention & training
Hosting summary
The database, authentication, storage and edge functions run on Supabase, on Postgres, with encryption at rest, managed backups and replication capability; the publisher mentions “regional hosting options” without naming a region. The application and front end are hosted by Lovable, a platform certified SOC 2 Type II and ISO 27001:2022, and the publisher states clearly that these certifications belong to Lovable, not to Feynn AS. Encryption in transit uses HTTPS with TLS 1.2 or above, and AES-256 at rest through the underlying infrastructure. Postgres Row Level Security is applied to key tables, service roles are separated, and JSON web tokens are short-lived with refresh tokens. What is not said matters as much: no hosting country and no region are named anywhere. The privacy policy notes only that some subprocessors may process data outside the EEA, framed by Standard Contractual Clauses, adequacy decisions and the EU-U.S. Data Privacy Framework, which is a transfer safeguard, not a hosting location.
Things to keep in mind
Risks and trade-offs to weigh before adopting Feynn.
- The SOC 2 Type II and ISO 27001 certifications shown on the security page are Lovable's, the application host, not Feynn AS's own; the publisher states this itself, but a hurried reader will credit them to Feynn
- The 30 free credits are a one-off lifetime allocation, not a monthly quota: once they are spent, use stops until you subscribe
- Credits are deducted when a job starts, and a refund after a system failure is at the publisher's discretion, so a failed run can still cost you
- The documents disagree on age: the terms require 18, while the privacy policy says the service is not intended for users under 16
- No data hosting country or region is named anywhere, even though the publisher is established in the EEA, so a data residency requirement cannot be verified
- No customer DPA is published, so organisations subject to the GDPR will have to request one before any regulated use
- AI Output is not professional advice and the publisher declines responsibility for decisions taken on it: sourced does not mean correct, and outsourcing judgement to a well-cited briefing is the real cognitive risk here
Setup & Integrations
Technical difficulty
Very low. Nothing is installed: Feynn is a web application. You sign up with an email address and password or through Google, LinkedIn, GitHub or Apple, with no payment card required, and guided onboarding asks only for a company name, which the platform disambiguates for you. There is no technical configuration, no API key and no integration to wire up. The real learning curve is conceptual rather than technical: getting a feel for the credit economy and for the game theory framework. The publisher addresses that directly, offering to walk newcomers through every concept.
Deployment
Behind Feynn
Resources
All the official URLs gathered for verification and reference.
Frequently asked questions
What is a credit on Feynn?
Do the credits on the free plan renew every month?
Should I choose Fast Research or Deep Research?
Is my content used to train AI models?
Can I try Feynn without a payment card?
Is there a mobile app or a public API?
Can I export my reports?
What is the minimum age to use Feynn?
Who publishes Feynn, and which law applies?
Which sources does Feynn draw on?
Should you pick Feynn?
Feynn is built for people who have to defend a decision, not for people who want a quick answer. Its distinguishing discipline is citation: every material claim comes back with a publisher, a date, an extract and a confidence score, and the publisher openly commits to flagging thin evidence rather than papering over it. For a strategy lead, a consultant, a procurement manager or an analyst who will be asked where a number came from, that is worth more than raw speed.
The reservations are real and mostly structural. Feynn AS is a very young Norwegian company, incorporated in late 2025, with a single officer on the register, so longevity deserves thought before anything critical depends on it. It holds no security certification of its own: the SOC 2 Type II and ISO 27001 badges on the security page belong to Lovable, its application host, and the publisher says so plainly. No hosting country or region is named. No customer DPA is published, and no customer-side opt-out from provider training is documented, even though Feynn states it does not train its own models on customer content. There is no mobile app, no public API and no announced third-party integration.
Pricing, by contrast, is transparent and unusually granular: credits are priced operation by operation, shown before execution, and roll over up to twice the monthly allowance on paid plans, with Pro starting at 20 USD per month for 200 credits.
Trying it costs nothing. The free plan grants 30 credits once, for life, with no card required, which is enough for roughly six Fast Research runs or two Deep Research reports, and that is the right way to judge whether the sourcing discipline holds up on a market you already know well.
- Choosing a selection results in a full page refresh.
- Opens in a new window.