KYC platforms that can automatically identify underperforming verification steps
When conversion drops, your overall pass rate tells you almost nothing about why.
A standard Know Your Customer (KYC) pipeline bundles identity data verification, document checks, biometric matching, fraud controls, sanctions screening, and decisioning logic into a single journey, so a blended number can hide a problem that's really coming from just one step. Before touching a business rule or adding friction to onboarding, you need to know which check is actually underperforming.
KYC platforms that can automatically identify underperforming verification steps include GBG, Trulioo and Persona.
Five signs a verification step may be underperforming
Spotting this requires tracking check-level metrics rather than the overall conversion average. A few patterns worth watching for:
- A sudden drop in a single check's pass rate, especially right after a system, vendor, or policy update.
- One demographic failing disproportionately, like a specific age group or user cohort struggling far more than the rest.
- A geography trailing behind comparable markets, even though the rest of the funnel looks healthy.
- Drop-off that appears right after your own team changes something, such as a rule threshold or a required field.
- A gap against industry benchmarks for similar user profiles, even without an obvious internal trigger.
Why a low pass rate doesn't automatically mean a check is broken
Looking strictly at pass rates can lead to incorrect conclusions. A low pass rate on an individual check doesn't automatically mean the step is broken or poorly configured.
Different verification steps naturally produce different outcomes based on where they sit in the onboarding funnel. A document check placed after a basic database lookup will encounter a pre-screened pool of users, whereas a document check placed at the front of a funnel will process raw, unfiltered traffic.
Customer demographics and data availability also skew results. A younger user cohort often lacks extensive public record history, which depresses primary database match rates without indicating a flaw in the verification logic itself.
Without external context, you risk making changes to steps that are already operating at peak efficiency. Benchmarking helps establish what achievable performance looks like for a specific sector and region.
Our analysis of verification pass rates for users under 25 shows just how wide the "normal" range really is:
- Financial services: 61% to 85%
- Fintech: 34% to 68%
- Gaming: 57% to 73%
A fintech pass rate of 50% might look alarming next to financial services' 85% ceiling, but it could be sitting comfortably mid-range for its own sector.
How GBG Foresight finds underperforming verification steps
Our AI-powered GBG Foresight solution helps you move from noticing an overall drop in conversion to pinpointing the exact check causing the issue and determining how to resolve it. Here’s how:
Pinpointing where customers get stuck
Foresight monitors individual check performance alongside overall volumes, pass rates, historical trends, demographic breakdowns, data quality and rule effectiveness.
By breaking performance down by geography and demographic, it reveals whether a pass-rate drop affects your entire user base or an isolated customer cohort.
It tracks configuration changes alongside performance metrics, helping you connect a drop in conversion directly to a recent rule adjustment. Rather than relying on teams to spot trends during monthly reviews, Foresight flags abnormal performance shifts automatically.
Using peer benchmarks to evaluate true performance
Foresight adds peer benchmarking so you can compare check-level outcomes against relevant external market performance rather than an arbitrary internal target.
This comparison reveals whether a check has realistic headroom for improvement or if it’s already matching industry standards.
Turning data analysis into prioritised actions
Foresight goes beyond diagnosing weak points by analysing root causes and generating specific recommendations. It recommends adjustments involving configurations, data sources, data capture methods or feature usage.
Recommendations are ranked by expected impact, which helps you focus your resources on changes that deliver the highest conversion lift. Each suggestion includes clear reasoning detailing the specific data points and comparisons that triggered it.
This played out for one of our customers with a 70% identity verification pass rate, a number that looked reasonable internally, until our benchmarking showed 80% was achievable for that sector. Our solution recommended adding a UK credit reference agency check through multi-bureau data sources, and after testing, pass rates climbed from 70% to 80%.
Addressing underperformance at the point of verification
Once an underperforming step is identified, fixing it often requires adjusting the underlying verification technology. Within our GBG Go identity verification platform, this can include:
- Smart document capture: Smart document capture addresses poor-quality image submissions by guiding users through real-time capture, removing glare and blur before image quality causes a failure.
- Automated document authentication: Identity document verification analyses documents automatically, preventing avoidable capture errors from becoming manual review burdens.
- Biometric verification: Biometric facial recognition uses face matching and passive liveness detection to confirm that the person submitting the document is its true owner.
- Multi-source data verification: Connects to alternate data sources or secondary identity attributes when weak primary data coverage depresses match rates.
Alternative platforms that identify underperforming verification steps
Here are two other platforms worth considering:
Trulioo
Trulioo provides real-time analytics to track conversion rates, average onboarding time, time spent at each verification stage and user drop-off between steps.
The platform uses AI-driven insights to help you evaluate flow performance and spot friction across onboarding stages.
You can customise verification steps to onboard genuine users efficiently while minimising manual reviews and maintaining compliance.
Persona
Persona offers a flexible identity infrastructure platform with analytics tracking built directly into onboarding flows.
Dashboard funnel analytics show screen-level progression, while per-step metrics highlight where users disproportionately abandon verification. The platform supports A/B testing on verification flows to test alternative steps.
You can track completion rates, isolate failure points and dynamically bypass unnecessary friction for low-risk customer segments. Visual funnel reporting identifies problem steps automatically, while remediation decisions and experiment adjustments remain team-driven.
Questions to ask before changing an underperforming verification check
Here are a handful of questions that are worth working through before touching a check that looks like it’s underperforming:
- Is it genuinely below a meaningful benchmark, or is it a strict-but-necessary compliance control doing exactly what it's supposed to?
- Is the drop uniform across your user base, or concentrated in one cohort, device type, or region?
- Did it start right after a rule edit, UI change, or new document requirement?
- Is the real cause upstream – formatting errors, blurry images or missing fields – rather than the check itself?
- Would routing failed users to a secondary data source or verification method actually recover them?
- What does the change do to fraud exposure?
Where this leaves your team
Finding a bottleneck is the easy part once you're tracking the right metrics. The harder and more valuable work is knowing why it exists, how it compares to what's realistic for your sector and which fix is worth testing first.
To find out more about how GBG Foresight works, get in touch.
FAQs
How do KYC platforms detect underperforming verification steps?
KYC platforms track check-level metrics like completion rates, average processing times and drop-off points across each stage of onboarding. They isolate specific screens or data checks where users fail or abandon the process disproportionately.
What’s the difference between overall pass rate and step-level performance?
Overall pass rate measures the percentage of users who complete the entire onboarding journey successfully. Step-level performance isolates individual checks, like document capture or address matching, to show exactly where failures or drop-offs occur.
How does peer benchmarking help optimise KYC workflows?
Peer benchmarking compares your check-level pass rates against industry standards for similar sectors and demographics. This context shows whether a low pass rate indicates a broken step or an expected outcome for that specific audience.
*Disclaimer: Information relating to third-party products and companies referenced in this article is based on publicly available sources and official publications at the time of writing. While reasonable efforts have been made to ensure accuracy, product features, positioning and company information may change and GBG does not guarantee that all information remains current or complete.
Nothing in this article constitutes an endorsement, recommendation or ranking of any third-party provider. Readers should consult each provider’s official website and conduct their own assessment before making any purchasing decisions.
To request an update or correction, please contact press@gbg.com.
Sources: