Agentic business banking has moved from conference panels to board agendas. In meetings with banks across Europe over the past year, the question has changed. It is no longer whether AI agents will matter to commercial and SMB banking. It is how much time a bank has left to act on its own terms.
Three shifts have converged. Each one on its own would justify movement. Together they close the window for cautious waiting. Here is what is happening, and the choice it leaves banks that serve small and mid-sized companies.
Shift 1: hyper-competition, quite suddenly
For decades, banks competed but could still rely on dominant products and the size of their balance sheets. That posture no longer holds. Direct-to-company fintechs, big-tech entrants, brokers, and white-label banking providers are creating real churn at most incumbent banks. The small and mid-sized business segment is the most exposed.
The threat is workflow-shaped. It rarely arrives as a wholesale switch of bank accounts. It arrives one workflow at a time. A company starts sending international payments through a non-bank provider because it is cheaper and faster. It runs FX conversion through a multi-currency wallet because the bank’s spreads are wide and the experience is dated. It pulls cash flow visibility from its accounting software because the bank’s portal does not show the same picture. Each of those flows used to live with the bank. The relationship erodes one piece at a time.
Independent research puts numbers on it. In Capgemini’s World Payments Report 2026, satisfaction with bank merchant services sat at 15 percent among small merchants and 22 percent among mid-sized merchants, and around 40 percent said they were considering a shift to specialist payment providers. The opening is closing, but it has not closed yet, and most merchants still prefer their bank for financial services overall.
Shift 2: the primary channel is moving into AI
For most of the digital era, the bank’s app and portal were the channel. That is changing. When a business owner asks a general-purpose AI assistant a question about company finances, that assistant becomes the channel through which the daily relationship runs. The app, the relationship manager, and the call centre all move one step further from the owner’s attention. The relationship has not been switched. It has been mediated.
The numbers are already moving. Surveys put the willingness to use a third-party AI financial agent at around 57 percent of customers if their own bank does not offer one. At that point the bank is a back-end product provider rather than a relationship owner. In the medium term, an external assistant could move from suggesting actions to initiating them on the company’s behalf.
Shift 3: regulatory clarity is arriving
The third shift is the one banks have been waiting for. The EU AI Act, DORA, and updated model risk guidance give banks a defined path for putting AI into regulated workflows. Supervisors now expect documentation, human oversight, audit trails, and explainability as standard parts of the build.
The question has shifted from “is this allowed” to “what does the supervisor expect.” For a regulated bank, that is the more useful question. It can be answered, and it can be built against.
Wait, adapt, or compete
Put the three shifts together and a bank has three options. Wait, and see how the AI channel develops. Adapt, and settle into life as a product provider behind someone else’s AI interface. Or compete, and own the agentic layer for business clients directly.
Banks choosing to compete need agentic capability inside their own perimeter. We set out what that involves, month by month, in the Agentic Business Banking whitepaper.
Agentic business banking inside the bank’s own perimeter
Competing does not mean replacing the core or buying another platform. It means composing agentic capability on top of the channels and engines a bank already runs, under the bank’s own brand. This is what agentic business banking looks like in practice: the agents handle the workflows business clients touch most, including FX hedging, cash flow forecasting, liquidity, cross-border payments, and collections.
The commitment that frames every agent is approval-gated execution. Agents do the data work, the analysis, and the recommendation. The material decision stays with a person. No autonomous trade execution, no autonomous limit changes, no autonomous credit decisions. The bank owns the agent: its branding, its data, the model choice, and where it runs.
That is the version of agentic business banking a Head of Transaction Banking can take to a risk committee. To see the full 12-month roadmap, download the Agentic Business Banking whitepaper, and subscribe to TreasurUpdate for the rest of the series.
Curious where the agent should sit at your bank?
Talk to our specialists