McKinsey's 2025 research finds that 80% of organisations approach AI with efficiency and cost reduction as their primary objective. These programmes generate savings. They produce faster versions of existing workflows. They rarely generate the compounding capability that constitutes a lasting competitive advantage.
The organisations generating the most value from AI share a different starting point. Their primary frame is growth or innovation: what AI enables the organisation to do that it could not do before. This is a strategic question, not an operational one. And because it is the first question asked, it governs everything that follows: which workflows are redesigned, which capabilities are built, which investments are made and how success is measured.
The framing decision is made implicitly in most organisations. A programme is approved, tools are selected and the first use cases identified are the ones that reduce visible costs. The efficiency frame is the default. The growth frame requires a deliberate choice.
The efficiency frame identifies existing processes and asks where AI can reduce cost, reduce headcount or reduce time. The output is faster versions of the same workflows. This creates value. It does not create strategic differentiation, because any competitor with access to the same tools can replicate the same efficiency gains in the same timeframe.
The growth frame asks a different question: given what AI enables, what can this organisation do that it previously could not? Which customer segments become accessible? Which propositions become deliverable? Which operations become possible at a cost point that was previously unviable? These questions require workflow redesign rather than workflow automation, and they produce capabilities that cannot be immediately replicated by competitors operating from the same technology.
An efficiency programme tends to produce a collection of tools with documented savings. A growth programme tends to produce a redesigned operating model with new competitive positions. The technology investment is often similar. The strategic outcome is substantially different.
Before the programme design begins, make the primary objective explicit at leadership level. The question to answer together: what does this organisation look like in three years if AI works as intended?
If the answer is primarily operational: costs are lower, processes are faster, headcount is reduced. The efficiency frame is in place. If the answer is strategic: the organisation serves different customers, delivers propositions it could not previously sustain, or operates in markets that were previously inaccessible. The growth frame is in place.
The answer governs the programme design. It determines which workflows are prioritised for redesign, which capabilities are built first and how value is measured throughout. Organisations that make this choice explicitly, at the outset, consistently design more ambitious programmes and generate more durable returns than organisations that let the efficiency frame take hold by default.