AI Readiness
for a Scale-Up

Situation

A fast-growing SaaS organization of over two hundred employees wanted to use AI in the product and internally. A proposal for a multi-year AI platform commitment was on the table. The CTO wanted to sign, the CFO had doubts, and nobody could prove who was right.

Challenge

The question on the table was "which platform?" The question underneath it was "are we actually ready?" That one had never been asked. At a scale-up moving this fast, that is understandable: the data foundation grows with the product roadmap, not with an architecture plan.

Approach

Three weeks, five dimensions. Twelve interviews from business to engineering, a data scan of the key systems, and an assessment of fourteen use cases already floating around. The data scan was decisive: customer data sat spread across four systems with three different definitions of a customer.

Decision

The advice was to delay the platform decision by two quarters and first unify the customer definition and the data flows. Two of the fourteen use cases could proceed immediately, because they ran on a single source and needed no integration.

Result

The organization got a readiness score per dimension, three named blockers, and a 90-day plan with owners. The two immediate use cases went live within a month. The platform decision was made later, on a foundation that could carry it, and with a sharper scope that made the license cheaper.