Verifying the Invisible Subscriber

Telcos across emerging markets are hitting a ceiling with their traditional business. Voice and SMS revenue keeps eroding and mobile data plans have become a commodity that gets re-priced downward whenever a new entrant undercuts on tariffs. Average Revenue Per User (ARPU) is flat or falling in market after market.

Ultimately, as a telco operator, there are two ways to grow revenue: – add more subscribers or generate more revenue from your existing subscriber base.

In emerging markets, the greater challenge is attracting and onboarding new subscribers, most of whom have no credit bureau record. This article explores how telcos can reliably identify and onboard applicants quickly and cost-effectively without unknowingly admitting fraudsters or adding unnecessary delays and inconvenience to genuine customers.

Every device-financing or microlending program in an emerging market runs into the same wall: how do you verify the individual or assess the onboarding risk when oftentimes there’s no mortgage history, revolving credit line, or credit bureau record. Traditional underwriting assumes a paper trail that a huge share of the addressable market in emerging markets simply doesn’t have.

The instinct is to reach for alternative data. The problem is that microloans and device financing — often $30 to $300 — cannot absorb a heavy per-applicant onboarding bill. A $2 all-in IDV call on a $50 handset loan isn’t a rounding error; it can be the entire margin. So, the real design question is where to use cost-effective alternative data, and where in the workflow it should sit.

1datapipe is an identity intelligence partner recently added to the Provenir Data Marketplace, specializing in emerging markets. It builds on a deterministic identity graph across 30 emerging markets and describes its “Living Identity” approach as resolving fragmented data into persistent, verified identity entities maintained over time with full provenance, auditability and explainability, — as opposed to traditional providers who offer point-in-time verification as part of a broader, more expensive all-in IDV check typically including document authentication.

As well as asking “does this name, ID, and phone combination match the applicant’s submitted details right now?,” its ID Graph capability asks, “has this identity been stable and coherent over time, across multiple authoritative sources?” Coverage runs across markets in LATAM, Southeast Asia, MENA and Africa with over two billion verified profiles — precisely the geography where credit bureaus are patchy or absent.

Even if a credit bureau exists in a region, accessing a credit report presents a significant cost of acquisition; — often one that can wipe out much of the profit. Additionally, they can’t help with thin-file applicants, because there’s no credit history. So, the key questions are “is this person who they say they are, and have they existed consistently long enough to be a trustworthy counterparty?” Identity persistence becomes a proxy for stability when repayment history doesn’t exist.

Consequently, the more cost-effective approach treats identity verification as a waterfall. Placing deterministic identity resolution high in the waterfall, before credit bureau reports and/or extensive IDV checks, means you’re not paying for these checks — or an agent’s time — on an application that’s fraudulent, synthetic or duplicated.

1datapipe’s identity intelligence is a key factor in reducing fraudulent applications that erode the margins of telco and other microfinanciers, an issue growing in emerging markets where historically accurate and comprehensive identity data has been very challenging to access. Additionally, housing it within Provenir’s Data Marketplace allows lenders to leverage 1datapipe’s identity intelligence along with other data sources covering identity, fraud, credit risk, open banking and affordability without building and maintaining separate integrations. That “pick and mix” architecture is the difference between a six-month integration project per country and a configuration change in a decision flow.

Larger loans can absorb a $5–$15 all-in verification cost because the ticket size and interest income justify it. Microloans and device financing can’t absorb that cost, which is exactly why they’re the products most often declined outright to avoid onboarding fraudulent applicants. In the process, however, genuine, thin-file applicants are declined, and the associated subscription revenue is lost. Accessing cost-effective data sources first and only accessing a credit report or IDV check for the applicants on which additional checks are warranted — is what turns “we can’t afford to know this customer” into “we can afford to serve this customer at a sustainable margin.” In markets where the credit data is thin, absent, or simply too expensive for a $75 handset loan, that sequencing decision is arguably more important than the underwriting model itself.

If these challenges resonate with you, learn more about 1datapipe® and how it enhances the Provenir ecosystem at the link below: