Mular
Crypto-to-Naira payments designed to feel like a bank transfer.
People were moving real money through crypto rails, but most of them didn't care about crypto. They just wanted their family to receive Naira quickly. How do you design a crypto product for people who don't want to think about crypto?
A money movement product hiding crypto complexity.
The interface had to help people move value, understand rates, trust transaction states, and recover from uncertainty without making the underlying rails the centre of the experience.
The constraint that shaped everything.
Crypto-to-Naira conversion sits at the intersection of two user fears: "will I get scammed?" and "will my family actually receive the money?" Every design decision had to answer both at once.
The Naira rate moves. Gas fees are invisible to most users. Exchange platforms have collapsed mid-transaction in Nigeria before — publicly, with user funds locked. The people using Mular had heard these stories. Some had lived them.
That context made trust design the primary work, not the conversion flow.
Wallet management was supposed to build confidence. It built confusion instead.
The first version of the onboarding included a wallet management section. Users would see their crypto balance, manage addresses, and track multiple assets before initiating any transaction.
It made sense on paper. Users would feel in control of their funds. The product would feel transparent.
Every user who saw it in testing stalled. They expected it to work like the crypto apps they had seen before — complex, technical, requiring knowledge they did not have. The wallet section, designed to add confidence, was doing the opposite.
We removed it entirely. Wallet management moved to the backend. Users see a Naira balance. Nothing else.
That decision cost us the power user segment — people who want to see their on-chain positions and manage addresses themselves. We made that trade deliberately, because the people sending money home are not power users. They are people who need to know their family in Lagos got the money.
Users were not trying to use crypto.
They were trying to get Naira into an account, protect value in USD, avoid P2P uncertainty, and know what happened when money moved.
The user's reference point was not Coinbase or Revolut. It was bank transfer behaviour, WhatsApp reassurance, and Nigerian money movement habits.
Product breadth had to feel calm before it felt powerful.
The app surface brings wallets, sending, receiving, swapping, history, and account states into one restrained money-movement system.
The model was Nigerian mobile banking, not crypto apps.
The model for Mular was not Coinbase or Binance. It was Nigerian mobile banking — GTBank, Access, Opay. Products users already trust for moving money.
That meant no wallet jargon. No confirmation hashes. No network fee breakdowns. The four-step conversion flow used plain language throughout: how much you're sending, how much they receive, all fees shown before commitment, a status tracker in human language. "Confirming," not "awaiting blockchain confirmation." "Sent," not "transaction mined."
The WhatsApp support button was in the main navigation. Nigerians trust WhatsApp. Putting it there was a trust signal, not a customer service decision.
Fee transparency, rate countdowns, and plain-language status.
Fee transparency moved to the calculator screen. Early builds showed the total fee at checkout — the standard e-commerce pattern. Testing showed that users who saw fees at the final step dropped off at higher rates than those who saw them upfront. We moved the full breakdown to the conversion calculator. Before the user started the flow, they could see exactly what the recipient would receive. No surprises at checkout.
The rate display updated every 30 seconds with a visible countdown. Crypto rates are volatile. Users needed to know the rate they saw was the rate they'd get — or close enough that they understood the window. A countdown made the rate feel live without creating anxiety. Static rates created distrust; a live rate with a countdown created urgency and honesty simultaneously.
Transaction status used five plain-language states, not two. Most crypto apps show "pending" until a transaction confirms, then "complete." The wait between those two states is where users panic. We introduced intermediate states with human language: "We received your crypto," "Converting to Naira," "Sending to your recipient," "Your recipient's bank is processing," "Done." Each state arrived as a push notification. Users stopped reaching for WhatsApp support mid-transaction.
A referral programme we had not earned the right to launch.
Beyond wallet management, I argued against a referral programme at launch. The team wanted to drive user acquisition through referrals — a standard fintech growth lever. My position was that we had not yet earned the trust required for users to stake their reputation on recommending us. A referral is a personal endorsement. Users who had only completed one or two transactions were not ready to make that endorsement, and asking them to would feel premature.
We launched without referrals. The first growth came through organic word-of-mouth in the same WhatsApp groups the users were already in. That was more valuable than a structured referral mechanic, because it came with social proof attached.
Two users. Opposite directions. Same infrastructure.
Mular routes by goal before feature: the crypto-first spender and USD saver are opposite flow directions, not the same journey.
Intent routing board
Maps how two opposite user goals share the same financial infrastructure without forcing one generic journey.
Open board full size ↗- Separates the crypto-first spender from the USD saver.
- Shows why routing by goal mattered before routing by feature.
- Frames Mular as bank-like money movement, not crypto education.
Home is the command centre.
Fund, Swap, and Send are action verbs on Home. Bottom navigation supports Home, Wallets, History, and Settings without turning the product into a maze.
Information architecture board
Shows how Home becomes the command centre for funding, swapping, sending, history, wallets, and settings.
Open board full size ↗- Keeps action verbs on Home instead of burying them in navigation.
- Separates funding, sending, wallets, and history into understandable surfaces.
- Makes the app legible before a user understands the underlying rails.
From crypto balance to Naira in a bank account.
The primary flow is USDT to rate visibility to bank recipient to transaction state to receipt. State clarity is the design argument.
Primary flow board
Follows the path from USDT balance to bank recipient, transaction state, and receipt.
Open board full size ↗- Connects balance, rate visibility, bank recipient, and confirmation.
- Shows where transaction state clarity carries user trust.
- Documents the core crypto-to-Naira flow without private data.
Every unclear moment gets a visible state.
Rate changes, fees, balances, timelines, and network warnings are treated as product states the user can understand.
Trust mechanics board
Collects the visible states that reduce anxiety when rates, fees, balances, and network conditions change.
Open board full size ↗- Names rate expiry, fee visibility, and transaction progress as trust surfaces.
- Shows where uncertainty becomes explicit product state.
- Keeps risky blockchain mechanics out of the user-facing mental model.
- Rate countdown
- Rate expiry / Refresh Rate
- Fee transparency
- Transaction timeline
- Balance privacy
- Network warning
Where the design earns trust.
Insufficient balance, rate expiry, wrong-network funding, and bank delays are named instead of hidden.
Edge-case states board
Documents the moments where the product has to be honest about blocked actions, wrong networks, and delays.
Open board full size ↗- Covers insufficient balance, expired rate, wrong-network funding, and bank delay states.
- Shows how error handling becomes part of the trust experience.
- Captures product debt to revisit before public launch storytelling.
What I would revisit.
- Zero-balance wallet percentage display should show "-", not "-100.00%".
- "Send to Self / Tax who?" is culturally sharp but may need compliance-sensitive treatment as the product scales.
- Beneficiary ordering should likely become recency/frequency weighted for repeat transfers.
Measured traction, source-verified.
Source: Internal Mular data, confirmed June 2026.
Merchant onboarding was an afterthought, not a distinct surface.
The merchant onboarding was an afterthought. We designed consumer flows first and adapted them for business accounts, which meant merchants encountered consumer-oriented trust patterns that did not map to their actual needs. A business owner converting ₦500,000 at a time needs different assurances than a person sending ₦30,000 home. I would have run separate research tracks from the beginning and built merchant flows as a distinct surface rather than an adaptation.