Start with the metric.  
Then build the
team to own it.

Almost every large transformation that disappoints, shares the same post-mortem. The technology was sound, the delivery was on time and close to budget. And the business metric the program was meant to move, barely shifted.

The reason is usually structural and rarely technical. Transformation programs are routinely organized around technology delivery rather than business outcomes, and no platform, however well implemented, can compensate for being aimed at the wrong target. McKinsey’s finding that 70% of transformation programs fail to achieve their stated objectives is often read as an execution problem. It is not. The most commonly cited root cause is the failure to tie technology delivery to clear ownership of business outcomes.

Putting it straight: technology-led transformation is a category error. Technology is an enabler. The transformation is in how work gets organized, how decisions get made, and which outcomes the organization is genuinely accountable for.

Start with the number, not the platform

Every transformation should begin with a single, honest question: which number are we trying to move, and by how much? Growth rate, cash conversion, customer retention, revenue per employee, time-to-decision. Whatever the answer, it should define the scope, the team, the success criteria and the way progress is measured.

In practice, the order is usually reversed. “What should we implement?” arrives first, and “what are we trying to achieve?” is answered afterwards, written to justify a decision already made. That inversion is where most transformation value is lost. Gartner reports that organizations structuring programs around business outcomes rather than technology deliverables are 2.5× more likely to report measurable ROI within 18 months. BCG Henderson Institute finds the same multiplier from a different angle: enterprises that redesign work around outcomes before selecting tools achieve 2.5× more value than those that drop technology into unchanged processes.

Fusion teams are an accountability model

If the outcome is the starting point, the next question is who should own it. The answer is the fusion team: a group built around a specific business metric, combining domain experts who understand how the business actually works with technical specialists who know how to change it.

Domain expertise is the most undervalued asset in transformation. The people who know where the process breaks, where decisions stall and why the workarounds exist are the people closest to the solution. Technology specialists know what is possible and how to change systems. Neither group, working alone, can redesign work well. Harvard Business Review notes that companies involving domain experts directly in designing AI-powered workflows, rather than translating requirements through IT, see 40% higher adoption and faster time-to-value. AI can augment that knowledge. It cannot replace it.

Fusion teams are modelled around accountability and not staffing. The team owns the metric, not the delivery plan, and that changes everything about how it operates. Decisions move closer to the work, get made faster and carry better context. Deloitte Insights finds that cross-functional teams with genuine decision rights and shared accountability for business metrics deliver outcomes 35% faster than siloed program structures.

What it takes in practice

A fusion team is not a project team with mixed membership. Putting a process owner on a steering committee and a business analyst in a delivery squad is coordination, not fusion. The real thing needs three conditions: a genuinely shared metric, real decision rights delegated to the team, and the organizational space to reimagine how work is done rather than simply automate it.

That last condition is the hardest. Automation preserves the current structure and makes it faster. Reimagination questions the structure itself: 

Fusion teams, with the right mandate, can do the harder thing.

It also means technology choices follow the business design, never precede it. AI is a powerful enabler here, but only once the team is clear on what needs to change and why. Deployed earlier, it tends to accelerate the very processes that should have been redesigned. MIT Sloan Management Review reports that 85% of executives see organizational alignment, not technical capability, as the primary barrier to realizing transformation value, with fragmented ownership of outcomes cited most often. The bottleneck is not the tool, it is the structure around it.

So, before your next transformation program launches, ask two questions. Does it start with the business outcome, or with the technology? And are the people who understand the business and the people who can change the technology genuinely working as one team, or merely coordinating across silos?

Be honest about the answers. They will determine whether the metric moves.