A failed KYC check is rarely a fraudster. More often it is a real person your data sources could not match: someone new to the country, young, recently moved, or thin on credit history. To improve your KYC pass rate and recover failed onboarding, you widen the data you check against, take a second pass over the failures, and tune the checks to the actual risk rather than rejecting on a first miss.
The customers who fail are usually the ones you most want to keep. They have chosen you, started signing up, and then hit a wall they cannot see. Most do not come back. The levers below recover a meaningful share of them without weakening a single control.
Why do good customers fail KYC checks?
Good customers fail because credit-reference-only checks cannot find a record to match against. Thin-file and no-footprint people, plus name, address and date-of-birth mismatches or stale records, all read as a fail even when the person is genuine. The check is working as built; the data behind it is simply too narrow to find them.
Around 7.1 million UK adults are financially excluded, roughly one in seven, per gov.uk and FCA figures. Many sit outside mainstream credit data entirely. If your verification leans on a single credit reference agency, those customers fail at the first hurdle regardless of how real they are.
How do you improve a KYC pass rate?
You improve a KYC pass rate by giving the check more places to find a genuine match. Add independent data sources beyond credit reference agencies, take a second pass over records that failed first time, set the number of required matches to the risk of the customer, and reserve document or biometric steps for cases that genuinely need them.
Each lever is independent. You can apply one or all of them, and you can measure the lift from each on your own data before committing to anything.
Add independent data sources beyond credit reference agencies
A single source has one view of a person. Matching name, address and date of birth across banks, mobile networks, insurance policy and claims data, public sector data, finance applications and credit reference agencies gives many more chances to confirm a genuine identity. A customer invisible to one source is often well documented in another.
This is the largest single lever for thin-file recovery. The person who has never held a credit card may well have a mobile contract and a current account, and those sources confirm them just as reliably for the identity step.
Run a second wash on the failures
A second wash is a second pass over the records that failed first time, run against additional independent sources. The recognised concept is batch KYC screening or remediation. Run it in real time at onboarding, so a failed customer is rechecked in the same session, or in bulk over an existing dataset to recover customers already turned away.
The real-time version protects live conversion. The customer never sees the first miss. They complete sign-up in one sitting, and the recovery happens behind the screen in the seconds it takes the additional sources to respond.
Configure the match count to the risk
The industry “2+2” convention means matching at least two identity attributes against at least two independent, reliable sources. It is a convention for satisfying the Money Laundering Regulations 2017, not a phrase written into the law itself. Higher-risk customers and sectors can require more confirming sources; lower-risk ones can sit at the baseline.
Configuring the required match count to the risk means you stop applying a single blunt threshold to every customer. A returning low-value customer and a first-time high-value one are checked to the standard each warrants, which lifts pass rates on the low-risk majority without softening the controls where they matter.
Route by exception, not straight to document scans
When a check returns a partial or no match, the common reflex is to escalate everyone to a passport scan and a selfie. That adds friction for thousands of people to catch a handful. Route by exception instead: send only genuine outliers to a heavier step, and recover the rest through additional data sources first.
For the end user, this is the difference between a verification that finishes in seconds and one that asks them to find their passport, photograph it, and scan their face while a balance sits waiting. Most drop out at exactly that point. Reserving the heavy step for the few who need it keeps the many moving.
Improving pass rates without lowering the standard
Improving a pass rate is not the same as loosening due diligence, and it must never become that. Every lever above recovers customers who are genuinely verifiable against more or better data. None of them passes someone who cannot be confirmed. The identity-matching step gets stronger because it draws on a wider, more independent evidence base.
The identity match is one part of customer due diligence. Your firm keeps the rest. Risk assessment and ongoing monitoring stay with you, along with source of funds where it applies. A certified verification service is a reliable, independent source for the identity step, and your firm remains responsible for the wider checks and ultimately liable. Recovery stays inside that boundary.
How much onboarding can you actually recover?
Recovery is often larger than firms expect, because the failures are concentrated in customers who are real but invisible to a single source. A Tier-1 gaming operator was failing credit-reference-only checks on close to 30% of new customers. A second wash matched data on 55% of those failures with around 45% showing no footprint at all, which is the genuinely high-risk tail.
Around half of the previously failed customers were onboarded, recovering around 15% of the total new-customer pool. Those customers would otherwise have walked. The no-footprint remainder, possible synthetic or under-age, stayed out, which is the correct outcome.
Frequently asked questions
How do I recover customers who already failed KYC?
Run a batch second wash over the failed records against additional independent sources. Customers who match can be re-offered onboarding. The Tier-1 gaming example recovered around half of its previously failed customers this way, recovering around 15% of the total new-customer pool.
Does improving pass rates increase compliance risk?
No, when done correctly. Recovery matches genuine customers against wider, independent data; it does not pass anyone who cannot be confirmed. Your firm keeps the wider customer due diligence and remains liable. The identity-matching step becomes stronger because the evidence base is wider.
Why do thin-file customers fail credit-reference checks?
Credit reference agencies hold limited or no records for people new to the country, young, or outside mainstream credit. With around 7.1 million UK adults financially excluded, single-source checks miss a meaningful minority who are entirely genuine.
What is the fastest way to test the uplift on my own data?
Run a sample of your failed records through a second wash and compare the match counts against your current pass rate. You see the recoverable share at record level before changing any live process.
See the uplift on your own data
The numbers above are someone else’s customers. Yours will differ, and the only way to know by how much is to measure it on your own records.
KYC Match is part of OneID, a digital verification services provider certified under the UK’s Digital Verification Services Trust Framework. It matches identity attributes across banks, mobile networks, insurance data, public sector data, finance applications and credit reference agencies, going beyond credit reference data alone, and returns the match counts you can set against your current pass rate.
Run 1,000 of your records through KYC Match for free and compare the results against your existing provider. Contact OneID to set up your comparison and see how many failed customers you could recover.
