Bolt-on is a bandaid on a broken leg. Systematic AI is the surgery.

I wrote that this week — intertwined with both the AI hype train and the enterprise software value crash. But diagnosis without proof is just consulting. Many are now likely to knee-jerk into ‘bolt-on’ AI. Don’t do it.

Get the right leaders in and do the job properly — here’s a little proof-point about getting it right:

Everest Group is the enterprise technology industry’s independent examiner. They grade on architecture, capability, and demonstrable market impact. One question: does this thing actually work the way the vendor says it does?

I took an enterprise underwriting platform from unrated — invisible to the analyst community, absent from every shortlist — to Everest Group Leader in twelve months.

Not by buying better technology. Not by bolting AI onto the surface of an unchanged system. By changing what the product fundamentally was.

We had a configured workbench. Capable. Functional. The kind of platform the 95% operate — technology that works, architecture that doesn’t compound. I redesigned it as an intelligence platform. Wove the thinking into how the product operated, not just what it displayed. Systematic AI before I had a name for it.

That required holding the full arc simultaneously. Commercial positioning. Technical architecture. Carrier relationships. Analyst engagement. Competitive strategy. Not delegated across five teams and a steering committee. Held — because architectural decisions don’t survive delegation. They require someone who understands the system well enough to make design choices, not just investment choices.

That’s what it looks like when one person holds the commercial, technical, and competitive dimensions rather than distributing them across an organisation and hoping they converge. The general expert doesn’t observe the architecture. They are the architecture.

Everest Group saw the result. Leader. Twelve months.

That recognition set in motion a chain of acquisitive activity that culminated in Advent International’s $2.5 billion transaction. The architecture compounded because the leadership compounded.

The 5% producing real returns from AI didn’t use different tools. They had different leadership — someone in the room who treated transformation as a practice, not a procurement exercise. Someone who could hold architecture, commercial model, and market positioning in the same head at the same time.

So, as we navigate $285 billion of lost value and refactor our thinking, is it time to consider leadership and execution? The architecture question doesn’t end at insurance.

I’m easy to find.

#AI #SystematicAI #ProductLeadership #CPO #EnterpriseTransformation #InsurTech

First Comment

151 patent claims across 15 inventions. Three domains — enterprise insurance, autonomous fintech, relationship intelligence — validated by the same methodology. The architecture transfers because it was designed to.

That’s the part the exam doesn’t capture. Everest Group validated one platform in one domain. The methodology underneath has been proven across three. Same structural thinking. Different surfaces. Each one compounding on the last.

The conversation always starts with architecture. But it never ends there.