Stablecoin on- and off-ramps for developers in Africa

An on-ramp turns local currency into stablecoins. An off-ramp turns stablecoins back into local currency and pays it into a bank account. If you're building anything that moves value between Nigeria and the rest of the world, one of these is the piece you're integrating.

What does an on-ramp actually do?

It does four things, and a provider doing three of them well still isn't a working on-ramp.

1. Collect local currency. Usually a virtual account: a unique NGN account number issued to your user, so a normal bank transfer credits their balance.
2. Price the conversion. A quoted rate, ideally locked for a defined window, so the amount your user is shown is the amount they get.
3. Execute and settle. Move the value and deliver it, on-chain to an address, into an internal balance, or as a payout to a bank account.
4. Tell you what happened. Webhooks. Payment systems are asynchronous, and without reliable event delivery you're building a polling loop and a reconciliation problem.

An off-ramp is the same in reverse: receive a stablecoin deposit, price the conversion, pay out to an NGN bank account.

What should you evaluate in a provider?

Rate and uptime are the obvious ones. These are the ones that surface three months in.

Should you build this yourself?

Usually not, and the reason isn't difficulty. It's that the hard parts aren't the code.

You'd need banking relationships to collect and pay out naira. Liquidity on both sides, held in the right proportions as flow moves. KYC and AML infrastructure. Treasury operations to manage the float and the FX exposure. And under the current framework, a place in a supervised programme.

Integrating a provider is measured in days. Building the equivalent is measured in quarters, and running it is a team.

What does Nigerian regulation require of you?

This is a build consideration now, not a legal afterthought, and it changed twice in 2026.

Crypto is legal. The February 2021 measure restricted banks, not individuals, and was lifted by CBN circular in December 2023.

Two regulators, with a line between them. An Executive Order in mid-July 2026 created a Virtual Asset Council chaired by the CBN Governor, with the SEC Director-General and the Nigeria Revenue Service Chairman as vice-chairs. The CBN supervises virtual assets used for payments, settlement, custody and wallets. The SEC regulates virtual assets behaving like securities.

Supervision runs through two programmes. The SEC's ARIP had 14 participants as of 15 August 2026. The CBN's regulatory sandbox admitted virtual asset providers for the first time in Cohort 2, 12 to 31 August 2026, across a VASP track and a data-enabled financial services track.

Approval-in-Principle isn't a licence. Busha and Quidax received AIP from the SEC in August 2024, and the SEC has said it confirms a firm met the requirements to proceed rather than granting an operating licence. No Nigerian platform currently holds a final licence.

Read the August 2026 draft rules now, not later. The SEC proposed mandatory registration including for offshore firms serving Nigerians, minimum capital of ₦2 billion for exchanges and custodians, ₦500 million for platform operators and ₦200 million for VASPs, API-based regulator access to transaction data, client asset segregation, and explicit coverage of staking, lending, yield and P2P. These are drafts under consultation. If the API access requirement lands as written, that's an architectural decision rather than a compliance checkbox.

Tax obligations sit on the platform. The Nigeria Revenue Service issued guidelines on 31 July 2026. Virtual asset providers pay 30% companies income tax and carry transaction reporting obligations, with penalties of ₦10 million for a first month of default and ₦1 million per subsequent month. Gains are computed on dollar value at acquisition and disposal, with only the dollar gain converted to naira. A 1% withholding applies to gross disposal proceeds and 1.5% stamp duty to token and fiat transactions, remitted twice monthly.

What it means for you: if you're holding customer assets or operating as a virtual asset provider, supervision applies to you directly. If you're integrating a supervised provider, most of it sits with them. Take legal advice on your specific model rather than inferring from this page.

Where do integrations usually break?

There are five failure modes worth designing for before you hit them.

1. The user sends on a chain you don't support. Decide upfront whether you detect and recover, or whether you make it impossible in the UI.
2. The quote expires mid-flow. A user opens the confirm screen, gets distracted, and comes back to a stale rate. Decide whether you re-quote or fail.
3. The webhook doesn't arrive. Build reconciliation from day one rather than after the first stuck balance. Assume at-least-once delivery and make your handlers idempotent.
4. A payout fails at the bank. Wrong account number, closed account, name mismatch. Know the retry and refund path before a user is missing money.
5. The user's deposit is short. Someone sends ₦49,850 for a ₦50,000 quote because their bank took a fee. Decide whether you part-fill, refund, or hold.

How do you test it?

• Use a sandbox that reproduces failure, not just success. Ask the provider how to trigger a failed payout, an expired quote and a wrong-chain deposit.
• Test webhook replay. Deliberately drop an event and confirm you recover.
• Test the reconciliation path with a deliberately mismatched amount.
• Test with the amounts your users actually send, not $1. Fee and rounding behaviour differs at the edges.

Where the Daya API fits

The Daya API moves money between naira and stablecoins through one integration: NGN virtual accounts for collecting local currency, crypto deposit addresses for USDC and USDT, guaranteed rate locks, bank transfer execution in NGN and USD by ACH, wire, RTP and SWIFT, multiple settlement options, real-time webhooks, and a sandbox with production parity.

Full reference is at docs.daya.co.

If another provider fits your use case better, our comparison of the options in this market says so.