Which IDV solutions work well for thin-file or unbanked customers?

Young adults, people who recently moved countries and those who've simply never used mainstream banking often have perfectly legitimate identities. They just don't have the decade of credit history older verification systems expect. 

Hold every real person to that standard, and you're not filtering out fraud so much as turning away customers who had nothing to hide.

GBG is an IDV solution that works well for thin-file or underbanked customers because it expands the data universe beyond traditional credit bureaus. In this article, we will look at how this works in practice.

Why traditional IDV can struggle with genuine customers

Many identity verification checks lean quite heavily on credit bureau data. And while that works for people with an established credit history, it offers little for anyone who hasn't built that specific kind of record. 

Thin-file groups often overlap with each other and include:

• Young adults opening their first bank account or credit card have little credit history simply because they haven't had time to build one.
• New-to-country consumers often have established identities elsewhere, but their data hasn't transferred to domestic credit bureaus yet.
• Consumers without traditional credit products regularly pay rent and utility bills without taking out mortgages, credit cards or auto loans.
• Underbanked consumers using alternative financial services generate real financial activity that doesn't show up in mainstream credit files.

For example, a 45-year-old with a mortgage, two credit cards and seven years at the same address gets verified in seconds because there's a deep financial file to check against. 

On the other hand, a 20-year-old applying for their first account has no credit history but has used the same mobile number and email address for five years. A credit-only check finds very little, not because their data is wrong but because there is not enough information from that specific data source.

What should an IDV solution do differently for thin-file customers?

Two changes make a big difference here: widening the data you check against and refusing to let one missing match end the process.

Use a broader range of data to establish identity

When a credit source comes up short, a multi-source data strategy gives you somewhere else to look for evidence.

Rather than firing off every data source at once, a waterfall approach works through them in order of relevance:

• Query the primary, most relevant source first.
• If it provides enough evidence, approve the user and continue onboarding.
• If the first source returns insufficient data, automatically query an alternative source before marking the check as failed.
• Move through additional sources only when required.

Don't let one “not found” result end the application

A missing bureau match means the source came up empty, not that the applicant failed. Before reaching for a higher-friction verification method, check whether another trusted source can fill the gap.

For a thin-file applicant, that might mean moving from conventional credit data to alternative sources like mobile or email intelligence. Only when the available data still can't provide sufficient evidence should the journey need to rely on another form of identity proofing.

Read more: Onboarding Gen Z and thin-file consumers: Improve pass rates

How GBG verifies thin-file or unbanked customers that traditional data may miss

GBG is an IDV solution that combines credit bureau data with alternative sources, like mobile, email and recurring-payment records, using the same waterfall logic described above across a wider set of sources than most platforms draw on. 

Someone with a thin credit file has usually still left a trail elsewhere, in a phone number they've kept for years or a rent payment made on time every month and that trail can confirm who they are just as well.

GBG can match 300 million consumers using mobile and email data, which opens a verification route for people underrepresented in credit files. It also gives you access to more than 100 million alternative financial-services users, extending verification to people who transact outside traditional banking.

Coverage spans hundreds of authoritative global and local sources across 50 countries, and using this multi-source strategy can improve identity pass rates by 9%.

Routing in the GBG Go platform adjusts the same way, by jurisdiction and risk score, and more than 110 modules covering identity checks, fraud signals and risk checks can be brought in whenever the data on hand is insufficient, or the risk calls for closer inspection.

Stefan Gajewski, Head of Product – Identity at GBG, sees the same failure mode again and again: a static journey that runs every applicant through the same data sets. The cost isn't just a bad experience. It's genuine customers failing checks they were never going to pass.

"A static journey runs everyone through the same data sets to verify them, even when the customer is, say, young with a thin credit file and the system is screening them against an expensive credit reference check they aren't going to match on." 

How to tell if your IDV process is excluding genuine customers

Your overall pass rate can be misleading: you might be sitting on a perfectly healthy aggregate approval number while shutting the door on entire groups of genuine applicants and never know it because the average smooths right over the problem.

Start looking at the breakdown. These are some of the metrics worth taking a look at:

• Pass rate by age group: If younger applicants are failing at a noticeably higher rate, that's usually a sign your checks are leaning too hard on long-term credit history they haven't had time to build.
• Pass rate by geography: Big changes by region point to uneven data coverage – your sources are probably stronger in some territories than others.
• Data-source match rate: Look at how each individual database is performing. The ones that keep coming back "not found" tell you where adding an alternative source would help.
• Manual-review rate: Check who's landing in the review queue. If it's disproportionately thin-file users, they're not being flagged for risk. They're being flagged for having a sparse credit file.
• False-positive rate: This tells you if your rules are catching genuine fraud or just being overly cautious with real customers.
• Drop-off by verification step: Wherever people abandon the flow is where the friction is. If it clusters around document requests, for example, that's your answer.
• Traditional versus alternative-data outcomes: Run the comparison. Before-and-after numbers on pass rates, manual reviews and fraud catches will tell you if alternative data is pulling its weight.

Choosing the right identity verification balance for thin-file populations

The thin-file customers who fail checks aren't necessarily riskier. They're just harder to see with the data most platforms rely on. 

Widen the data and build in a fallback, and you can approve them without loosening anything on the compliance side.

To learn more about how GBG can help you do this, get in touch with our team.

FAQs: Thin-file and underbanked identity verification

What's the difference between an unbanked and an underbanked customer?

An unbanked customer has no account with a traditional bank or credit union. An underbanked customer holds a traditional bank account but still relies on non-bank financial services.

Does using alternative data for identity verification increase fraud risk?

No. Mobile records, email intelligence and recurring payment histories are tied to a real person's actual, everyday activity. Paired with fraud prevention checks, they can confirm a genuine identity and catch fraud earlier in the process.

Why do traditional credit checks fail for thin-file applicants?

Traditional credit checks rely on years of reported credit activity, such as credit card balances, auto loans, and mortgages. Thin-file applicants haven't had the time or need to build those specific credit records, meaning credit bureaus return little or no information even when the applicant's name and address are correct.