Evolve: The Operating Model AI Demands makes a structural argument: the Plan-Build-Run technical operating model is the binding constraint preventing organisations from capturing AI value at scale, and the product operating model is the governance form that enables continuous evolution.
The urgency is empirical. Boston Consulting Group’s survey of 1,250 senior executives across nine industries identifies three distinct performance tiers: approximately 5% of organisations are achieving AI value at scale, 35% are generating partial returns, and 60% report minimal gains (Boston Consulting Group, 2025). The critical insight is that the 5% are not using more AI tools than the 60%—they have redesigned their processes around AI capability. As Brynjolfsson and McAfee (2014) identify, realising technology’s productivity potential requires restructuring work around what the technology makes possible, not computerising existing tasks.
The book diagnoses the structural reason most organisations cannot make that transition. Plan-Build-Run separates strategy, delivery, and operations into sequential phases governed by handovers, approvals, and project-scoped funding. That design cannot adapt at operating speed. Every AI cycle produces learning that the operating model has no mechanism to act on.
Enabling you to evolve from Plan-Build-Run to a modern, AI-ready operating model.
Operating Model
Portfolio Management
Governance
Portfolio Management
Product Management
Technology Strategy
Portfolio Management
Matthew Hodgson has spent over thirty years inside complex, high-compliance, resource-constrained organisations — government and corporate — helping executives through the same types of problems: why investment in technology change so rarely produces what it was designed to achieve. He co-founded Zen Ex Machina after establishing, across enough engagements to be certain, that the cause is structural — that the organisational design itself determines the outcome, regardless of how well the people within it execute. Most practitioners identify what is not working and move on. Matthew stays until the replacement model is in place, and behaviours are self-sustaining, because the conditions that produced the original failure will reassert themselves if the replacement is not deeply embedded. He has found, over and over again, that few practitioners are willing to do this and fewer still are capable of it.
Evolve is the diagnosis he has been developing for thirty years.
Matthew holds a Master of Knowledge Management from the University of Canberra and a Bachelor of Science in Psychology from the University of Newcastle. He is one of only a few people in the Asia-Pacific region who holds both a SAFe SPC and Scrum.org PST certifications. He lives in Canberra, Australia, with his partner Mia and their dog, Chewie.