136 Million NINs and the Dream of a Single KYC API

If you’ve ever had to write code that verifies a user's identity in Nigeria, you know the exact moment your hair started turning grey.

One day you are integrating a BVN lookup. The next day, a client asks why they can't use their driver's license. Then you're dealing with voter's cards that look like they were printed on a desktop inkjet in 2011, and your OCR library is crying.

So when I saw that the National Identity Management Commission (NIMC) crossed 136 million enrolments and is pushing the National Identification Number (NIN) as Nigeria’s sole official identity under a new 2026 Act, my developer brain immediately went to one place: Does this mean I can finally stop writing wrapper classes for five different buggy KYC APIs?

The API Nightmare We’ve Been Living Through

Building product in Nigeria means spending half your engineering budget just trying to prove your users exist.

If you are setting up a fintech app or a gig-matching platform, onboarding is where dreams go to die. You ask a user for their BVN, they get suspicious. You ask for a NIN, the third-party verification service goes down. You try to fall back on a phone number, but the telco's database is having a quiet crisis of its own.

Having a "one person, one identity" policy written into law is a massive step forward on paper. It targets the root problem: database duplication.

Right now, the Federal Road Safety Corps (FRSC) has their data, the banks have yours, the telcos have another version, and INEC has a completely different spelling of your last name. Getting rid of this fragmentation is the dream. If we can query one single, unified database that actually works, building digital services gets incredibly cheaper and faster.

The Real-World Execution Gap

But let's be real for a second. We’ve been burnt before.

As a builder, my worry isn't the policy. Abisoye Coker-Odusote and her team have done some serious heavy lifting to push enrolments to 136 million. That’s more than half the country, which is no small feat given the logistics of hitting rural areas.

My worry is the infrastructure.

Imagine you’re sitting in a shared office space in Gbagada, trying to debug a payment flow. Your app is ready to launch. But every time your backend hits the identity verification endpoint, it takes eight seconds to resolve, or worse, returns a 504 Gateway Timeout.

If the NIN is going to be the only official identity, NIMC's servers are about to handle traffic levels they have never seen before. Every bank account opening, every SIM registration, every passport application, and every micro-loan request in the country will hit their infrastructure. If the API isn't built to scale—if it doesn’t have high throughput, low latency, and 99.9% uptime—the entire digital economy will grind to a halt.

The "Sapa" and Edge-Case Test

We also have to think about the people who aren't online.

It is easy to design apps for the tech crowd in Lagos, but what about the trader in Onitsha or the student scraping by in Akure? If they lose their physical NIN slip, or if their biometric data doesn't match because the fingerprint scanner at the local local government office was dirty back in 2019, they get locked out of the financial system.

With over 100 million Nigerians still left to enroll—mostly children and rural populations—we need offline-first verification methods. We need a system that doesn't require a high-end smartphone and a 4G connection to verify who someone is.

Skeptical, But Rooting for the Win

Despite the skepticism that comes with being a developer in this environment, I really want this to succeed.

If NIMC can pull off being the root certificate authority and give us a clean, stable, and affordable way to verify identities, it changes the game. It means we can spend less time worrying about KYC compliance and more time actually building features that solve real problems for real people.

Until then, I’ll keep my error-handling code robust, my timeout limits high, and my fingers crossed. We don't gree for anybody, not even stubborn identity databases.