Sarah Heffron Advisory & Consulting
Most execution problems aren't execution problems. They're operating model problems.

The right structure isn't just how you fix what's broken. It's how you capture what's possible.

Most organisations arrive at an operating model problem the same way: through growth that outpaced the infrastructure built to support it, or complexity that accumulated faster than the design could absorb it. By the time it surfaces as an execution problem, the structural causes have been there for months.

Sarah Heffron brings 20+ years of senior leadership experience, including redesigning product operations across 200+ countries at Visa and building the operating model for a new $800M+ business unit at Mercer from the ground up. That experience is in identifying where operating models fracture under pressure, and designing the infrastructure that holds.

The work starts with diagnosis. Everything else follows from that.

Operating model diagnosis, design, and transformation for organisations where growth, complexity, or AI has raised the stakes on execution

When complexity has outpaced execution, the instinct is to act. And when a major opportunity arrives, whether that's a scaling inflection point, an AI transformation, or a strategic pivot, the instinct is the same: reorganise, hire, invest, launch a programme.

These can be the right answer. But without diagnosis first, they're more likely to treat the symptoms than the cause.

Whether the driver is scaling, complexity, or AI, the starting point is a conversation.

Whatever you need: a structured diagnostic, a full operating model redesign, fractional transformation leadership, or a review of transformation already underway, get in touch.

Why this work is different

Most transformation approaches start with a solution. A methodology to apply, a framework to implement, a programme to run. The diagnosis, if it happens at all, is shaped by the answer that's already been chosen.

I start with the system. Before any design work begins, I need to understand which structural failures are present: where scaling or complexity has outpaces the design, how they connect, and what the operating model is actually producing versus what it was designed to produce.

The work is also continuous across phases. The person who does the diagnostic is the same person who does the design, and if you choose, the same person who leads the transformation. No handover between phases, no junior team interpreting a senior partner's framework.

The scaling inflection point

When growth reveals what the operating model can't absorb.

The complexity threshold

When complexity has made everything harder than it should be and no one can explain why.

The AI investment gap

The technology is only as effective as the operating model designed to support it.

Signals worth paying attention to

Seven consistent patterns signal that an operating model is, or soon will be, in trouble.

Strategy that looks clear but isn't translating into execution.
Decisions that loop and never land.
Work that is invisible to the people who need to prioritise it.
Cross-functional initiatives that consistently underdeliver.
Different business units, functions, or geographies pulling against each other rather than working together.
Incentives and metrics that produce the wrong behaviours even when teams are doing exactly what they're measured to do.
Growth that is revealing the gaps rather than validating the model.

These patterns appear in organisations that are struggling. They also appear in organisations that are scaling, or trying to capture a new opportunity and finding that their current infrastructure wasn't built for it.

Risks and opportunities with AI

Organisations are investing heavily in AI. Most are discovering that the returns depend less on the technology than on the operating infrastructure surrounding it. Unclear decision rights, fragmented processes, and weak governance don't disappear when AI is introduced. They become more visible, and more costly.

But the relationship runs in both directions. AI, deliberately applied to specific execution gaps, has the potential to address coordination failures and seam issues more effectively than traditional operating model redesign. The organisations getting the most from AI are deliberately designing their programmes around those gaps, not just managing the implementation risk.

Both conversations start in the same place: a clear-eyed diagnosis of where the operating model is actually breaking down, and what needs to change. I work with specialist AI implementation partners to ensure that the infrastructure design and the technology delivery are aligned from the outset.

The experience behind the approach

Two decades designing and running execution infrastructure in some of the most complex operational environments in fintech and professional services. At Mercer, across two tenures within a $4.5B enterprise, including the design and operational leadership of the multinational client group ($800M+ portfolio across 40+ countries) and a CFO-sponsored revenue model transformation. At Visa, leading the redesign of the global product operating model across 200+ countries: building the decision rights framework, cross-market coordination infrastructure, and delivery governance that connected global product strategy to regional execution at scale.

Links to Selected Case Studies

Frameworks and Assessments

Link to Perspectives & Articles