Ravelin
AI-native fraud, abuse and payments risk prevention for high-volume online merchants.
By Ravelin · HQ London, UK · 4.0/5 Value-Position score (estimate)
Positioning guardrails
Best for
- High-volume online merchants and marketplaces processing millions of digital transactions
- Ecommerce and travel platforms that want to cut chargebacks without blocking good customers
- Fraud teams that want machine learning models tuned to their own KPIs
- Businesses expanding into new markets that need 3D Secure and payment optimisation in one place
Ideal size: 1–10 fraud/risk specialists people · Scale-up or enterprise with high digital volume and an integration team
Not for
- Small merchants with low transaction volumes
- Buyers who need published self-serve pricing and instant sign-up
- Teams wanting a no-integration, plug-in tool for non-technical staff
Value metrics scorecard
Time-to-Value
6–8 weeks
~45 days to first production value
Total Cost of Ownership
On request
Enterprise, quote-based; pricing is not published and engagements start with a demo call.
Implementation Friction
3/5
Engineering + admin effort required
Value-Position score
out of 5 · model estimate
Full cost breakdown
Mandatory implementation fee
None
Seat tiers
Not listed
Add-on costs
- None
Company & support
Who is behind Ravelin, and how your team gets help once it is live.
Company
- Founded
- Not recorded
- Headquarters
- London, UK
How you get support
- PhoneNot listed
- EmailNot listed
- Live chatNot listed
- Support portal / ticketsNot listed
- Community forumNot listed
- Help centre / docsNot listed
- Dedicated account managerPaid plans
- In person / on-siteNot listed
- Hours
- Not recorded
- Response time
- Not stated
Site states merchants get a dedicated account manager plus Integrations, Product Support and Data Science teams within baseline pricing. No support phone line, email address, ticket portal or stated SLA is published.
“Not listed” means the vendor’s public pages don’t mention that channel, not that it is unavailable. Ask about it during evaluation.
Market position
Where Ravelin sits against its closest alternatives. Pick any two of cost, speed, friction and buyer score, and up to 9 companies to compare.
Quadrant view
Typical annual cost × Time-to-value
The lines cross at the median of the solutions shown, so about half sit on each side of each line. A dashed ring marks an outlier pinned to the edge; hover for its value.
Companies on the chart 6 / 10
- Ravelin
- Plan A
- Deepki
- Convera
- Cypago
- Elliptic
Add or change companies
Up to 10 companies including Ravelin. Listed closest first.
Stack fit signal
Compatibility with standard B2B ecosystems.
No supported MCP path today, so it cannot be driven from an AI client.
AI & MCP readiness
What Ravelin ships in AI, and what it asks of your ecosystem.
AI features shipped
Sources describe custom machine learning models that score fraud risk, multi-model decisioning, a consortium database of 9+ billion identity elements and graph link analysis. A third-party AI provider section exists but names no provider and no copilot or assistant feature.
Your data & models
- Trains on your data
- Trains on your data
- Runs on
- Not recorded
- AI pricing
- Not recorded
In your ecosystem
- AI connection
- Not supported
- Model key
- Not recorded
- AI usage audit
- Not recorded
Compliance attestations
* IAPP AIGP certifies individuals, not products. It means named staff hold the credential — not that the platform does.
Bottom line
Ravelin is an AI-native fraud and abuse prevention platform for high-volume online merchants. It combines custom machine learning, a consortium database of 9+ billion identity elements and graph link analysis to score payments, block account takeover, refund, promo and marketplace fraud, and run PSP-agnostic 3D Secure. Named customers include Flix, Trainline, Glovo, Just Eat and River Island. Pricing is quote-based and starts with a demo.
Frequently asked questions
How is Ravelin priced?
Ravelin does not publish pricing. The site routes every commercial enquiry to a demo or call, so contracts are scoped per merchant, likely on transaction volume and product modules. Budget for a sales cycle and for internal integration work rather than a self-serve plan, and ask for a written quote covering 3D Secure, scoring and any add-on modules.
How long does implementation take?
No timeline is published. Ravelin says its Integrations team works with your developers to ensure secure, accurate data transfer from the start of the partnership, and its data scientists then tune models to your traffic. A 6–8 week window is a reasonable planning assumption for a first production deployment at a mid-size merchant, but confirm it in scoping.
Which fraud problems does Ravelin cover?
Online payment fraud and chargebacks, refund and returns abuse, account takeover and multi-accounting, promo, policy and voucher abuse, and supplier, partner or marketplace fraud. It also supplies a PSP-agnostic 3D Secure server and SDKs plus transaction optimisation to raise frictionless approval rates.
What support is included?
Ravelin states that merchants get a dedicated account manager plus three teams covering Integrations, Product Support and Data Science within baseline pricing, including direct access to technical specialists and data scientists. No support phone line, published email address, ticket portal or formal response-time SLA is listed on its site, so ask for those in contracting.
What certifications does Ravelin hold?
Ravelin states PCI 3D Secure compliance and PCI DSS compliance, says it was assessed by the British Assessment Bureau against ISO 27001:2022, and is listed in the Visa Global Registry of Service Providers. It describes its consortium identity database as fully anonymised and GDPR-compliant. These claims are not published on a dedicated security or trust page, so request audit evidence directly.
Does Ravelin use customer data to train models?
Ravelin's privacy policy states it collects and processes end-user personal data and metadata to operate and improve its machine learning algorithms, on a legitimate-interests basis. It also says that where third-party AI providers are used, data is kept in a Ravelin-designated environment and is not used to train the third party's model. Buyers should confirm opt-out terms in the DPA.