How Payhawk Turned Seven-Year AI Vision Into Reality
Speaking to FinTech Magazine, Payhawk reveals how its long-held artificial intelligence ambitions are finally coming to fruition.
When Chief Financial Officer Konstantin Dzhengozov and his co-founders established Payhawk seven years ago, they harboured an ambitious vision that seemed ahead of its time.
"We started the company seven years ago, and one of the ideas that we had back then was exactly this, to have a chatbot where you can have a conversation of, 'Hey, where is my invoice? What amount do I need to pay to this supplier? Or, what's my expense and cash burden going to look like in the next month?'"
The Bulgarian expense management platform found itself in a familiar position for many fintech pioneers - conceptualising solutions that the underlying technology couldn't yet support. "Seven years ago, it was just an idea, and the technology was not there yet," Dzhengozov explains.
From experimentation to implementation
Today, the landscape has shifted dramatically. Payhawk has moved beyond the experimental phase that characterised much of the fintech sector's early engagement with artificial intelligence.
"Right now we are past the phase of just experimenting and trying different things out, and the technology is there so we can build on this," Dzhengozov says.
The company is now actively developing and bringing these AI capabilities to market, fulfilling a long-held ambition.
As Payhawk develops its AI capabilities, the company faces the complex challenge of regulatory compliance in the financial services sector.
Dzhengozov acknowledges the evolving nature of AI regulation, noting that "this is still the very early stage. It is going to be maturing more in the upcoming years as AI in general is still very early stage."
However, the company's approach to AI development has been shaped by its existing regulatory framework. "Because of how our product is structured and because we handle payments, we need to have specific audit trails, we need to have different permissions and different rules," Dzhengozov explains.
This has influenced Payhawk’s AI architecture fundamentally. "All of this is taken into account when we're building AI. So AI is a layer on top of all the controls and all the frameworks that already exist."
"One of the clear indicators early on that we are seeing is that companies can close their month much faster, usually two, three or four days quicker than before"
Measurable impact on financial operations
The early results from Payhawk's AI implementation are delivering tangible benefits for clients. The company has observed significant improvements in operational efficiency, particularly around month-end financial processes.
"One of the clear indicators early on that we are seeing is that companies can close their month much faster, usually two, three or four days quicker than before," Dzhengozov reports.
This efficiency gain creates a ripple effect throughout financial operations. "This gives them more time to spend on analysing performance or making decisions or stress testing their financials, or preparing different scenarios and different alternatives. So yes, we are seeing the benefits of our product for our clients already."