TrustSphere Vendor Assessment: Recorded Future, Threat Intelligence That Now Has to Earn Its Seat on the Fraud Floor


The problem Recorded Future addresses is the one at the heart of today's companion theme post: the intelligence that would stop a fraud chain early almost always exists, but it lives in the cyber estate and never reaches the fraud decision. Compromised credentials circulate in stealer logs, cloned card data sits on dark web shops, and phishing infrastructure is registered days before it is used, all visible to a well-instrumented threat intelligence function and all invisible to most fraud engines. The category question is whether a threat intelligence platform can become an operational input to fraud prevention, rather than a briefing service for the security team.
TrustSphere's house view is that Recorded Future is the strongest collection and analysis platform in the category, and that fraud leaders should nonetheless buy it with their eyes open. It is a cyber-native product whose fraud offering, Payment Fraud Intelligence, is genuinely useful but card-centric, and the platform does not make fraud decisions: it informs them, and only if the institution builds the plumbing to consume it. Since Mastercard agreed to acquire the company from Insight Partners for 2.65 billion US dollars, announced in September 2024, the payments orientation has sharpened, which is good news for issuers and a strategic question mark for everyone else.
The scale claims are the company's own but are stated publicly and consistently: more than 1,900 client organisations across more than 75 countries, including numerous national governments, served from its Intelligence Cloud, which indexes open web, dark web and technical sources in real time. Fifteen years of collection history and graph infrastructure is not something a competitor replicates quickly, and it is the foundation on which every judgement below rests.
One framing point before the scores. We assess Recorded Future here as a fraud and financial crime buyer would, not as a CISO would. On a pure cyber scorecard it would rate higher; our index weights the distance between intelligence and a prevented loss, because that distance is where financial institutions actually live. Readers comparing this assessment with cyber analyst rankings should keep that difference of lens in mind.
RiskTech Index Score and Capability Profile
Recorded Future scores 6.9 on the TrustSphere RiskTech Index, against an index mean of 6.06. Capability scores: threat intelligence collection and analysis 9; identity and credential exposure intelligence 8; payment fraud intelligence 7; integration and automation tooling 7; real-time fraud decisioning contribution 4; AML and investigations support 3; mid-market accessibility and total cost 4.
The shape of that profile tells the story better than the headline number. This is a platform with world-class inputs and honest gaps at the point of consumption. The 9 and the two 8-adjacent scores reflect collection breadth, analyst tooling and the credential exposure data that fraud teams most need. The 4 for real-time decisioning reflects the fact that Recorded Future does not sit in the payment path and never will; the 3 for AML reflects a product that was simply not built for monitoring, screening or SAR workflows, whatever a sales deck implies. A 6.9 overall places it comfortably above the index mean, earned on intelligence quality rather than fraud-stack completeness.
What It Genuinely Does Well
Collection is the moat. The Intelligence Cloud's coverage of criminal forums, stealer-log marketplaces, card shops and phishing infrastructure is as broad as anything we have assessed, and the entity graph linking indicators to actors, tooling and campaigns turns raw mentions into usable context. For a bank, the practical output is specific: which of our customers' credentials are circulating, which of our BINs are being traded, which typosquatted domains were registered this week.
Payment Fraud Intelligence is the offering most relevant to this week's fusion theme, monitoring dark web carding channels to identify compromised cards before they are used, so issuers can reissue or step up authentication pre-emptively rather than eat the fraud and the chargeback. The direction of travel under Mastercard is visible in the companies' own announcement, which describes a jointly developed service alerting institutions when cards are likely compromised and claims, on their own figures, a doubling of the rate at which compromised cards were identified year on year. We treat that as a vendor claim rather than an independent finding, but the mechanism is sound and matches what our clients report.
The identity intelligence is, for fraud teams specifically, the quiet star of the platform. Stealer-log and credential exposure monitoring against a bank's own customer and employee domains gives an account takeover early warning signal that most fraud stacks simply do not have, and it maps directly onto the fusion telemetry layer described in today's theme post: exposure detected in the cyber estate, consumed as a step-up trigger in the fraud estate. Institutions we have reviewed that operationalised this feed moved their account takeover detection point from the payment back to the login, and in some cases back to before the login, which is where the economics of prevention are at their best.
Integration tooling is better than the category norm. A mature API, prebuilt connectors into SIEM and SOAR platforms, and analyst-facing AI summarisation mean the intelligence can reach workflows rather than inboxes. Institutions that treat Recorded Future as a data feed into their own risk engines, not as another portal, extract several times the value of those that do not.
The Limitations, in Detail
First, Recorded Future is not a fraud platform and should not be scored as one. There is no transaction decisioning, no customer risk profiling, no case management for fraud operations, no device or behavioural biometric telemetry. An institution without a capable fraud decision engine will find Recorded Future tells it, eloquently, about losses it still cannot prevent.
Second, the payment fraud offering is card-centric. Compromised card monitoring does little for authorised push payment scams, mule networks or account funding fraud on instant payment rails, which is where many of our clients' losses now concentrate. The credential exposure data helps with account takeover across all channels, but the packaged fraud product is strongest for issuers and acquirers specifically.
Third, cost and consumption maturity. Pricing sits at the premium end of the category, modules are licensed separately, and the platform assumes an intelligence function exists to consume it. A mid-tier institution without dedicated threat intelligence analysts risks paying tier 1 prices for a capability it can only partially operate, which is exactly the duplicated-spend trap described in today's theme post.
Fourth, the ownership question. Mastercard ownership accelerates the payments roadmap, but buyers should think through the second-order effects: roadmap gravity toward card ecosystems, the perception question of feeding a scheme-owned platform with your fraud telemetry, and concentration risk for institutions that already depend on Mastercard scheme services. None of these is disqualifying; all of them belong in the due diligence file rather than discovered at renewal.
Fifth, intelligence volume is a tax as well as an asset. Without disciplined tuning of watchlists and risk rules, the platform generates more context than a lean team can action, and we have seen deployments where the weekly intelligence digest became the deliverable. Intelligence that does not change a decision is a cost.
A final observation on deployment rather than product. The institutions that extract full value from Recorded Future share an operating pattern: a small intelligence cell with explicit service obligations to the fraud function, agreed signal-to-action routes written down before go-live, and a quarterly review that retires any feed or watchlist that has not changed a decision. Where we have seen the platform disappoint, the cause was almost never collection quality; it was that nobody had decided, in advance, which team would act on which class of finding within what time. That is cheap to fix on paper and expensive to discover in production, and it should be settled during the proof of value, not after contract signature.
Who It Is Right For, and Wrong For
Right for: tier 1 and upper mid-tier banks with an established SOC and a fraud function willing to consume external signals; card issuers and acquirers, for whom Payment Fraud Intelligence has the most direct loss linkage; institutions building the fraud and cyber fusion telemetry layer, where it can serve both consumers on one contract; and firms with nation-state or ransomware exposure that need the broader intelligence estate anyway, making the fraud use case incremental.
Wrong for: institutions whose dominant loss is APP scams on instant rails, who should prioritise behavioural and payee intelligence first; mid-market and smaller firms without analysts to operate it, who will get better marginal return from a managed detection provider or their existing fraud vendor's intelligence add-on; and any buyer expecting a decision engine, case manager or AML capability, because none of those is on offer.
Five Demo Questions a Slide Cannot Answer
Take our top ten BINs and show, live, what your platform surfaced on them in the last seven days, and how that would have reached our fraud engine without an analyst copying indicators.
Show the full path from a stealer-log credential match on one of our customers to a step-up authentication decision in our stack: latency, format, and who maintains the integration.
What proportion of your payment fraud intelligence relates to non-card typologies such as APP scams and mule accounts, and can you show examples rather than categories?
Walk through how pricing scales across modules over three years for an institution of our size, including the analyst headcount you assume we will dedicate.
Since the Mastercard acquisition, what concrete roadmap items serve non-issuer financial crime teams, and which were delivered rather than announced?
Verdict
Recorded Future is the best intelligence collection platform a financial institution can buy, and a partial answer to the fraud problem it is increasingly marketed against. At 6.9 against an index mean of 6.06, it earns its premium where the buyer is an issuer, has a fusion-minded operating model, and treats the platform as a feed into its own decisioning rather than a destination. Buy it for the telemetry and the card intelligence; do not buy it instead of a fraud engine, an APP scam capability or an AML programme, because it is none of those things and, to its credit, has never seriously pretended otherwise.
Suggested Next Steps
Map your current threat intelligence spend across fraud, cyber and financial crime teams before any procurement, and quantify the duplication a single platform would remove.
Run a four-week proof of value on your own BINs and customer credential exposure, measured on fraud interceptions and reissuance decisions, not on volume of intelligence delivered.
Pressure-test the integration path into your fraud decision engine with your own engineers in the room, including latency and maintenance ownership.
Score the vendor against your actual loss mix; if APP scams dominate, sequence behavioural and payee-side controls ahead of card intelligence.
Sources: Mastercard and Recorded Future acquisition announcement; FATF; NCSC; TrustSphere RiskTech Index proprietary scoring; TrustSphere Risk Index, April 2026.
TrustSphere helps financial institutions design and deploy intelligent fraud and financial crime detection solutions. Visit www.trustsphere.ai



Comments