Tingg
Enterprise-grade payments, for businesses that aren't enterprises — a multi-market redesign of Cellulant's B2B payments platform, rebuilt to serve SMB merchants across Africa without losing the enterprise capability larger customers already depended on.
Problem: Tingg was built for enterprises. As it expanded to small and mid-size merchants across 30+ markets, the interface generated the support tickets and drop-offs the business could least afford.
Core decision: Keep one unified product — not an SMB-versus-enterprise split — and manage complexity through progressive disclosure, task-based navigation, and guided onboarding.
Outcome: Support ticket volume dropped by roughly half across the top five categories. SMB retention lifted. The platform scaled without scaling the support team.
Power without pain.
Tingg is Cellulant's B2B payments platform — multi-currency, cross-border, compliance-ready — built for the large enterprises that anchored Cellulant's early business. As Tingg expanded into small and mid-size merchants, usability started working against adoption.
Onboarding a new business, reconciling a transaction, issuing a payout — each required steps that enterprise finance teams could handle but owner-operators could not. How do you simplify a platform powerful enough for enterprises, without breaking it for the enterprises that already run on it?
The brief.
Redesign the Tingg merchant experience so a new SMB merchant can complete their first successful cross-border transaction without contacting support — while keeping existing enterprise customers fully operational.
Compliance non-negotiables across 30+ markets. Live production platform with real enterprise volume daily. Lean design resource through most of the project.
Research & discovery.
I worked across three sources: support ticket patterns (categorised with customer success to find the flows generating most volume), merchant behaviour on the live platform (where they dropped off or cycled back), and conversations with internal operators patching the breakdowns every day.
Onboarding was a cliff, not a ramp. Most SMB merchants who signed up never completed a first transaction.
Support dependency was structural.Merchants weren't confused — the platform required support.
Trust signals were missing where they mattered. Large cross-border transactions were abandoned mid-flow. In fintech, silence reads as risk.
Three moves that carried.
Task-based navigation over feature-based navigation. Primary navigation restructured around tasks — Get paid, Send money, Reconcile, Report — not internal product areas like Collections and Payouts.
Progressive disclosure over SMB / enterprise tiering. One unified interface: base experience surfaces what an SMB needs on day one; advanced capabilities reveal contextually as merchants grow.
Onboarding as a guided path, not a form. Contextual steps with clear framing about why each field is needed, save progress, and resume — because a long form forces abandonment.
Flows & trust.
Onboarding from signup to first transaction — business type, KYC, banking, and configuration in their own steps. Dashboard leads with incoming payments, send money, and recent reconciliation.
Transaction flows rebuilt around visible trust signals — FX rate disclosure before commit, transparent fee breakdowns, explicit status at each stage.
Design system.
Working solo across a growing platform, I built the design system in parallel with the redesign — every component shipped to be reusable as the platform expanded across 30+ markets.
On a platform at Tingg's scale, design leverage is a system question. What a solo designer can ship is bounded by how reusable and self-documenting the system is.
Control is the product.
As digital commerce expanded across African markets, Tingg's merchants were buying more online — SaaS, ad spend, cross-border supplier payments — but their tools didn't match the risk. Physical corporate cards were slow to issue, hard to control, and frequently compromised.
How do you design a new payment instrument so the businesses using it feel more in control, not less?
Card creation as a scoped action. Name the purpose, set a limit, optionally an expiry — the card is issued. Virtual cards only work if merchants issue them per-purpose instead of reusing one default card.
Status as product surface.Active, paused, cancelled, expired — visible on every card, switchable in a tap. Not "Status: Paused" but "No new charges will go through."
Transaction moments as trust moments. Real-time notification with merchant, amount, card used, and remaining limit. Declines show the reason — limit reached, card paused, region mismatch — not a generic failure.
The redesign was measured against a single question — can a new SMB merchant reach their first transaction without us? — and tracked through Cellulant's internal product metrics across the two quarters after launch. I can't publish the exact figures, but the direction wasn't ambiguous.
Support load fell sharply. The top five ticket categories — the routine onboarding and reconciliation questions the redesign targeted — roughly halved. SMB merchants stayed. Businesses that had been churning inside their first month were now running recurring transactions. More merchants got in. As onboarding drop-off fell, new registrations rose with it. First-transaction completion rose — the one metric the whole project was built against.
On Virtual Cards: activation ran ahead of early projections, and a meaningful share of activated merchants issued a second card within 30 days — proof that people were using cards per-purpose. For a feature touching money and security, tickets per active user came in well below comparable fintech launches.
The strategic result outlasted any single number: simplifying for the smaller merchant turned support from a cost that scaled with growth into one the platform could grow past.
What worked: Anchoring everything on one metric — can a new SMB complete a first transaction without support? What I'd change: The design system should have been scoped in week one, not built in parallel. What I took forward: At scale, design leverage is a system question, not a craft one.
Powerful financial infrastructure only creates value when businesses can actually use it.