If you’ve been running a business on Paystack for a while, you’ve probably noticed how the dashboard grew over the years. What started as a clean, simple interface gradually became more layered, more features, more tabs, more places to click before you could answer a basic question like “Why did revenue dip last Tuesday?”
READ ALSO: What Employees Gained from Africa’s Top 5 Acquisitions
That complexity wasn’t a mistake. It was the natural result of Paystack doing its job: adding payment options, analytics, settlement tools, and integrations as African merchants’ needs evolved. But somewhere along the way, the dashboard stopped serving its most important purpose giving business owners fast, clear answers.
On May 21, 2026, Paystack changed that. The company launched the first full rebuild of its Dashboard in ten years, introducing a redesigned interface, simplified navigation, mobile parity, and an AI-powered Command Centre that lets merchants ask questions about their business activity in plain language.
Before getting into the features, it’s worth understanding why this rebuild matters so much in the African context.
Across Africa, the vast majority of businesses often more than 80% in some countries are small and informal operations whose owners rely heavily on smartphones to run day-to-day operations, from customer communication to inventory management.
These aren’t venture-backed startups with dedicated finance teams. They’re market traders, boutique owners, service providers, and freelancers who need tools that work as fast as they think. The old dashboard, despite all its power, wasn’t built for the way most of them work.
More than 300,000 businesses now use Paystack to process trillions of naira monthly across five African countries, including Nigeria, Ghana, and South Africa. Serving this audience well means making complexity invisible, and that’s exactly what AI enables.
The headline feature of the new dashboard is what Paystack internally calls Canvas an AI-native Command Centre baked directly into the interface.
The redesigned product introduces an AI-native Command Centre that allows businesses to ask questions in plain language and receive answers grounded in their own Paystack data.
Think about what that actually means in practice. Instead of navigating to the transactions page, filtering by date, exporting a CSV, and cross-referencing your records, you can just ask: “What happened with this transaction?” or “Why is revenue down this week?” and get a direct answer.
As Senior Product Designer Dara Assim-Ita, who led the rebuild, put it: “Businesses don’t come to their dashboard because they want to click through pages. They come because they have questions. Over the last decade, we have seen firsthand how much time merchants lose navigating tools that were built to display data rather than deliver answers. With this rebuild, we have changed that.”
That’s a quietly radical shift. Most business software ,whether it’s accounting tools, CRMs, or payment dashboards ,is still fundamentally built around displaying data. Paystack is betting that the future is built around answering questions.
One legitimate concern with AI in financial tools is accuracy. When an AI answers your question about revenue or transactions, you need to know it’s telling you the truth, not hallucinating numbers or making assumptions.
Paystack addressed this head-on. To power the Command Centre safely, Paystack developed Project Canvas API, a proprietary service that manages LLM conversations, connects to model providers, and interfaces with existing Paystack systems.
The safety architecture has three layers:
Grounded Data Retrieval ensures answers are strictly bound to the merchant’s real, verified data to prevent AI hallucinations. Compliance Screening means all outputs pass through strict safety and compliance filters. And Rigorous Testing, Paystack’s Data Protection and Privacy team subjected the system to extensive adversarial testing and a comprehensive Data Protection Impact Assessment.
In other words, the AI isn’t guessing. It’s pulling from your actual data and surfacing it in a more human, conversational way. That distinction matters enormously for businesses making real decisions based on these answers.
The Command Centre gets the most attention, but the rebuild goes much deeper.
The dashboard has been restructured to reduce cognitive load ,fewer clicks to get to the things you use most, less hunting through menus. The rebuild also introduces full mobile parity, which is significant given how smartphone-first African merchants are; the full power of the dashboard is now accessible from a phone without compromise. The Command Centre itself isn’t just a chatbot pinned to the side of the screen ,it is woven directly into the dashboard UI, meaning answers come with charts and context where relevant, not just text. And this is explicitly a foundation: the company noted that the current release focuses on core payment workflows, with more of Paystack’s products expected to migrate to the new architecture over time.
This dashboard launch doesn’t exist in a vacuum. It’s part of a much larger strategic shift at Paystack.
Five months ago, Paystack launched The Stack Group (TSG), a new holding company, as it continues its expansion beyond payments into banking, consumer finance, and AI-led product development. TSG’s founding shareholders include Stripe, Paystack founder and CEO Shola Akinlade, and existing Paystack employees.
TSG Labs, the company’s venture studio, will focus on building new products using emerging technologies, including AI, and will operate independently from the group’s regulated entities ,allowing Paystack to experiment with AI-led products without exposing its core financial businesses to regulatory risk.
COO Amandine Lobelle was clear about the company’s direction: “We’re not moving away from financial technology. We’re expanding our scope. Emerging technologies could include AI or stablecoins. The mandate is to solve both fintech and non-fintech problems that are critical to Africa’s digital future and development.”
The AI dashboard, then, is less of an isolated product decision and more of a statement of intent, proof that Paystack’s AI ambitions aren’t theoretical.
