AI · OPTIMISATION

Before Agentforce: the data foundation banks actually need

Every bank wants the AI agent. Few want to hear that the agent is the last 10% of the work. Here's the unglamorous 90% that decides whether your AI tells the truth.

Why AI projects stall in banking

An AI agent is only as honest as the data it can see. In most banks, the customer exists five times — once in core banking, once in cards, once in the CRM, once in a complaints tool, once in a spreadsheet someone built in 2019. Point an agent at that landscape and you get confident answers assembled from contradictory sources. In a regulated industry, that's not a productivity tool; it's a liability generator.

The foundation, in order

The banks winning with AI didn't buy better agents. They built a foundation where an honest agent was possible.

The good news about sequencing

None of this delays value until "someday." Each foundation layer pays for itself before any agent arrives: resolved identity improves every service conversation; connected data kills swivel-chair work; consent clarity de-risks marketing. The agent, when it comes, is a switch you flip — not a leap you take.

A practical first step

Audit one customer journey and count the systems a human touches to answer "what's this customer's situation?" That number is your AI-readiness score. Above three, you have foundation work — and that's normal. It's also exactly the work we do.

This article relates to our Optimisation & AI work. If it raised a question about your own programme, the conversation is free: book a consultation.
Start the conversation

Let's talk.

One conversation with the architect — and a clear view of what your bank could ship next quarter. If we're not the right fit, we'll tell you in that call.

Book Free Consultation  →