Do You Need Separate Sanctions Screening Software, or Is Your KYC Platform Enough?
Screening customer names and screening live payments are different jobs. Use these tests to check whether your KYC vendor's bundled sanctions module covers how your money actually moves, or whether it leaves gaps it was never designed to close.
Name screening and payment screening solve different problems
Most KYC platforms now include a sanctions check, and many compliance teams assume that settles the question. Usually it doesn't. You need to decide whether you need sanctions screening software built for transactions, or whether screening customers at onboarding and on an ongoing basis meets your obligations. How money moves through your business decides that far more than which vendor you pick.
Customer name screening checks the people and companies you have a relationship with. The data is structured: full name, date of birth, nationality, registration number and address. Screening runs at onboarding and again whenever a list or the customer record changes. A few seconds' delay doesn't matter here, because no payment is waiting to clear.
Payment screening is a different job. Every inbound and outbound message is checked before release, whether it arrives as SWIFT MT, ISO 20022 pacs.008, SEPA, card settlement or a domestic scheme format. Many of the parties in those messages were never onboarded by you, including beneficiaries, intermediary banks and ultimate debtors. Their names turn up in free-text fields, sometimes truncated, misspelled or run together with an address. Trade finance adds vessel names, ports, goods descriptions that may cover dual-use items, and bills of lading.
Instant payments tighten the constraint. SEPA Instant, UK Faster Payments, FedNow, India's UPI, Brazil's Pix and similar schemes allow only seconds from end to end, and screening gets a fraction of that time. A module built for onboarding, which may take several seconds per lookup and send every hit to a manual queue, wasn't designed for that. Scheme rules also vary by region. Under the EU's Instant Payments Regulation, for example, payment service providers verify their own customers against EU sanctions lists at least daily instead of screening each euro instant transfer against EU lists. That doesn't cover non-EU lists or other payment types, so check how the rules apply to each scheme you use with your compliance counsel.
When your KYC vendor's bundled screening is genuinely enough
For many firms, bundled screening is the right choice. A dedicated engine you don't need brings integration work, another vendor to oversee and another alert queue to staff. Bundled screening usually holds up when all four of these conditions are true:
- You don't originate or release payments yourself. Two common examples are a fintech operating through a sponsor bank or licensed payment provider that screens every transaction, and a wealth manager whose custodian handles settlement.
- Funds only move to verified parties. A lender that pays loans only into accounts in the borrower's own name, after screening the borrower, has little counterparty exposure.
- Your alert volume is small enough for manual review. A few hundred new customers a month plus a daily rescreen produces a queue analysts can clear by hand.
- The module covers every list your regulators expect, local lists included, and loads updates within hours of publication.
The first condition comes with a caveat. When a partner screens your transactions, your legal obligation stays with you. Get the details in writing: which lists the partner screens, at what thresholds, how quickly you're told about hits involving your customers, and whether you can see the evidence. Treat any point the contract leaves out as a gap in your coverage.
Four tests that separate a bundled module from a dedicated engine
Judge every option on four things: list coverage, latency, fuzzy-matching controls and audit evidence. Bundled modules often do reasonably well on coverage and fall short on the other three.
List coverage beyond the major regimes
Any credible vendor covers the core lists. These include OFAC's SDN and non-SDN lists, the UN Security Council Consolidated List, the EU's consolidated financial sanctions list and the UK Sanctions List. Vendors differ on local lists and ownership data. Depending on where you operate, you may also need:
- Switzerland's SECO lists
- Singapore's MAS designations
- the UAE Local Terrorist List
- India's designations under the Unlawful Activities (Prevention) Act
- Australia's DFAT Consolidated List
- Canada's listings under its autonomous sanctions laws
- Hong Kong's designations
Ask how each vendor handles ownership. OFAC's 50 percent rule and the EU's ownership-and-control test can capture entities that appear on no list, and you'll only find them with corporate ownership data. If a vendor's answer is only about name lists, they don't cover ownership.
Latency under load
Ask for p99 response times at your peak volume, not demo averages. For instant payments, screening generally needs to finish well inside a second. You also need clear rules for what happens to a payment that hits: hold, reject, or release on timeout. A vendor that hasn't thought about the timeout case hasn't run screening on instant payments at scale.
Fuzzy-matching controls
Fuzzy matching on names produces both false positives and missed hits. A dedicated engine usually lets you set thresholds per list and per field and handle transliteration from Arabic, Cyrillic and Chinese scripts. It also controls how name order and initials are matched and keeps good-guy lists with expiry dates and reviewer sign-off. Many bundled modules offer a single global threshold and little else, and that one setting forces you to trade missed hits against analyst workload.
Audit evidence
Examiners in most jurisdictions expect you to explain why a match was or wasn't raised. To do that you have to reconstruct which list version was loaded, which algorithm and threshold applied, and who cleared the alert and why. Ask the vendor for a sample export. A vendor that can only show you a screenshot can't give you usable audit evidence.
| Criterion | Typical bundled KYC module | Typical dedicated engine |
|---|---|---|
| Primary use | Onboarding and ongoing customer rescreening | Real-time payment, trade and customer screening |
| Message formats | Structured customer fields | SWIFT MT, ISO 20022, SEPA, domestic schemes, free text |
| Local list coverage | Varies; often sourced from a data provider | Broad, with support for custom and internal lists |
| Matching controls | Global threshold, limited tuning | Per-list and per-field rules, transliteration, test sandbox |
| Latency | Seconds; suited to batch runs | Sub-second at scale |
| Audit trail | Log of alert decisions | Full replay of list version, rules and decisions |
What each option costs, and where the real spend goes
The figures below are typical 2026 market ranges. They vary widely by contract, volume and region, so confirm them with vendors.
| Item | Typical range (USD) |
|---|---|
| Bundled screening per onboarding check | Included, or roughly $0.10 to $1.50 per check |
| Ongoing customer monitoring | Roughly $0.50 to $5 per customer per year |
| Dedicated SaaS payment screening, mid-size firm | Roughly $50,000 to $300,000 per year |
| Enterprise engine for a large bank | Roughly $250,000 to well over $1 million per year |
| Implementation and integration | Roughly $30,000 to $500,000, one-off |
| Premium list and ownership data licences | Roughly $20,000 to $200,000 per year |
The licence is rarely the largest cost. Analyst time spent clearing false positives usually is. Take a hypothetical payment firm that screens 2 million transactions a month and alerts on 1 percent of them, which comes to 20,000 alerts. At two minutes per alert, that's roughly 670 analyst hours a month, or about four full-time staff. Two minutes is optimistic, and complex alerts take much longer. If tuning brings the alert rate down to 0.3 percent, close to three of those four people are freed for other work. That saving alone often covers the cost of a dedicated engine.
For larger banks, dedicated vendors include LexisNexis Risk Solutions (Firco), NICE Actimize, Oracle and SAS. Newer SaaS options include ComplyAdvantage and Napier. Several KYC-first platforms are also adding transaction screening, so confirm what's live in production and not just on the roadmap.
Watch for these red flags in any proposal:
- The vendor can't tell you how many hours it takes for a newly published list to go live in production.
- The demo shows thresholds, but there's no sandbox to test changes before you deploy them.
- Pricing is per alert, which rewards the vendor for noisy matching.
- ISO 20022 structured and unstructured fields aren't both supported.
- Local lists you need are offered only as a "custom upload" that you maintain yourself.
A decision rule and a two-week test to settle it
Our rule is simple. If you release payments yourself, whether cross-border, instant or trade-related, you need either a dedicated engine or a KYC platform that passes all four tests at your peak volume. If you only onboard and monitor customers while a regulated partner screens the transactions, a well-configured bundled module is usually enough. That holds only if your contract documents the partner's coverage.
Run a structured test before you sign:
- Map your flows. List every channel where funds enter or leave, its message format and its time budget.
- List your obligations by jurisdiction. Record every sanctions list required in each country where you operate or clear currency, and confirm the list with compliance counsel.
- Build a test file. Include listed names with misspellings, transliterations, reversed name order and truncation. Add a set of clean names that look similar.
- Measure both kinds of error. Record missed hits and false positives at the vendor's recommended settings, then again after tuning.
- Load-test latency at about twice your current peak and record p99, not the average.
- Request an audit export for five cleared alerts. Check whether someone outside the team could follow the reasoning.
Your current KYC platform fails if it misses known names in step four or can't produce the export in step six. If either happens, start shortlisting dedicated engines this quarter. Don't leave the gap for an examiner to find.
Quick answers
Do we need to screen against OFAC lists if we aren't based in the US?
Often, yes. US sanctions can reach non-US firms that clear US dollars, use US correspondent banks, involve US persons or rely on US-origin technology. Many banking partners also require OFAC screening by contract. You'll still need your local and regional lists alongside OFAC. Ask qualified sanctions counsel to confirm your exposure.
Can we rely on our sponsor bank's or payment provider's transaction screening?
In practice, a partner can handle transaction screening for you, but responsibility for your own customers and your sanctions program usually stays with you. Document which lists and thresholds the partner uses, how quickly they tell you about hits, and what evidence you can see. Keep screening your own customers whatever the arrangement.
How often should existing customers be rescreened?
Common practice is to rescreen whenever a relevant list changes, usually through a daily or more frequent delta run, and whenever customer details change. A purely periodic schedule, such as annual rescreening, generally falls short of what supervisors expect for sanctions. Check your local supervisor's guidance, because expectations differ by jurisdiction.
Does ISO 20022 make payment screening easier?
Partly. Structured fields for names, addresses and party roles make matching more precise and can cut false positives. Many messages still carry unstructured or hybrid address and remittance data, so you need an engine that screens both. Ask vendors to show results on your own ISO 20022 samples, not their demo data.
Written by the Trustdeskapp editorial team. Prices, features and rules change often - confirm details with the vendor or a licensed professional before you decide.