Why Strategy Fails in Execution

Every bank in the region has approved an ambitious digital transformation or efficiency strategy. And every bank can point to one that failed to deliver the promised results.

The question almost never asked is not why execution failed. It is a more uncomfortable one: does the board know what the organization is executing in its name?

The conventional explanation blames a lack of discipline and assumes that a clear mandate was abandoned. That is not what happened. The problem is not cancellation. It is visibility.

It is not a mandate that remains alive.

It is sediment.

Layers from the previous strategy, the one before it, and initiatives that never went through a formal decision: exceptions, operational workarounds, or criteria adjusted years ago by a manager who left no trace. All of them are executed with the same discipline as any declared priority.

That sediment takes three forms, residual, fossil, and ghost, which this analysis distinguishes below.

McKinsey documented the symptom in December 2021: fewer than one-third of organizational transformations improve performance sustainably, and 55% of the value loss occurs after implementation.

Strategy almost never fails for lack of ambition.

It fails because senior leadership unknowingly governs only the most recent layer.

The Consensus That Mistakes Technology for Transformation

The committee approving the technology budget assumes something that is almost never true: that it has complete visibility into every process the investment is meant to replace. It does not.

The conventional answer has been repeated for years in innovation committees: the gap between strategy and results can be closed by investing more in technology. Yet the organization may be executing rules that predate any strategy the current board has ever seen, inherited from previous administrations, mergers, or committees that no longer exist.

Modernizing the technology stack is not the same as modernizing the business model. Adding an application or a digital channel on top of a fragmented process is not redesign. It is a new surface laid over rules, controls, and accountabilities whose age no board member can establish with certainty.

KPMG, in its July 2025 report, "Banking Transformation: The New Agenda," based on responses from 228 banking leaders worldwide, confirmed the pattern: only 18% reported a high degree of transformation success, just 24% achieved their cost-reduction targets, and only 40% considered themselves well positioned to lead change.

BCG, in its December 2021 survey, found that only 35% of companies achieved their digital transformation objectives, barely above the 30% recorded in 2020.

It is worth noting that these surveys are market studies conducted by consulting firms, not independent institutions, although their findings converge with the IDB and FELABAN data cited below.

A new interface does not conceal an old process.

It conceals processes whose origins no one around the boardroom table can reconstruct.

The Gap Management Fails to See in Time

The damage accumulates before it becomes visible. It occurs when a priority is translated into frontline behavior, something board-level dashboards almost never measure.

Donald Sull, Rebecca Homkes, and Charles Sull documented in Harvard Business Review in March 2015, drawing on a survey of nearly 8,000 managers across more than 250 companies, that between 66% and 75% of large organizations struggle to execute their strategy, and that only half of middle managers can name even one of their company's five top priorities.

Middle managers are not resisting change. Many do not even realize they are executing a deviation: entrenched routines and unwritten rules that appear in no document.

The board did not cancel these initiatives. It never saw them.

• The residual initiative is a survivor of the previous strategy. It can be identified if someone looks for it, and its origin can still be traced to a set of minutes or a policy.
• The fossil initiative was inherited from previous administrations or mergers. Almost no one in the organization can reconstruct its origin. It survived because no executive had either the visibility or the authority to challenge it.
• The ghost initiative was never formally approved, yet it exists. Resources are assigned to it, it has de facto owners, and it produces outcomes that no one at the top measures or attributes to any declared priority.

The three are invisible for different reasons, but all three generate real execution that few people around the boardroom table can capture in a quarterly strategy progress report.

The hidden risk is not resistance to change.

It is the execution of accumulated mandates that even the board itself cannot fully enumerate.

The Hidden Cost of Digitizing Without Redesigning

When technology is added to the sediment instead of removing it, the cost ceases to be conceptual. What the IDB measured in Latin American development banking is precisely this gap between what institutions declare and what they actually operate.

The IDB, in Technical Note IDB-TN-3238 by Fernández and Heredia, published in 2026, evaluated digital pilots conducted in Colombia, Ecuador, and Paraguay between 2022 and 2023. It found initial maturity levels of only 36% to 49%, with persistent gaps in processes, strategy, culture, and data, precisely where fossil mandates accumulate.

FELABAN's Eleventh Report, published in 2025, confirms the other half of the problem: access to bank accounts in Latin America rose from 39.1% in 2011 to 69.7% in 2024. The channel changed. No one ever fully mapped what was operating behind it.

The real cost is not the expenditure.

It is governing without a complete inventory of what the institution is actually executing.

Go-live does not solve that problem. It only changes what the customer sees.

The FCA and PRA fined TSB Bank £48.65 million in 2022 over its 2018 IT migration, which left 5.2 million customers unable to access banking services. The regulators attributed the failure to inadequate governance, not to the technical migration itself. In 2020, the OCC imposed a US$400 million penalty on Citibank for deficiencies in data governance and risk management.

Wells Fargo is the series' clearest example: the systemic endpoint of a ghost mandate operating under the scrutiny of a regulator that ultimately pursued it.

In its US$3 billion settlement in 2020, the Department of Justice did not uncover a mandate that management had failed to cancel. It uncovered a sales culture that, between 2002 and 2016, developed rules of its own, including targets that drove employees to open accounts without customer consent, despite no strategic document ever approving such behavior as policy.

It lived in organizational behavior and reached the board only after the damage had become irreversible.

That is not the execution of a previous strategy.

It is the execution of a strategy the board never knew existed.

No board should accept a project being declared successful before it passes that test.

There is a line between manageable deterioration and structural deterioration. It is not identified when indicators fail. It is crossed when the organization has committed enough capital, structure, and narrative that reversing the path requires more than it can sustain.

That is the line a responsible leadership cannot cross.

You Cannot Cancel What Was Never Recorded as Alive

If the organization is executing layers of sediment, the first question is not what should be cancelled. It is what exists, so that it can be cancelled.

You cannot sign the death certificate of something that was never recorded as alive.

The first act of governance is not cancellation. It is making the invisible visible through three steps no board can afford to skip:

1. Map (inventory what has never been declared). Before assigning an owner with fiduciary accountability, and before aligning incentives with the economic value created, the divergent decision requires an inventory of the organization's actual mandates: what is being executed today, with which resources, under which incentives, since when, and with what traceable origin, or with no traceable origin at all. This is not a compliance audit. It is an audit of the undeclared.
2. Classify (each type must be cancelled differently). A residual initiative can be ended through a simple written decision. A fossil initiative requires forensic work before deciding whether it deserves a new mandate or should disappear. A ghost initiative is the most dangerous to touch without analysis. It may be creating value that one part of the organization has already learned to defend, and cancelling it blindly can be as damaging as allowing it to continue without governance.
3. Cancel and verify (with a date, an accountable owner, and evidence of adoption). Only at this point does the formal act occur, supported by an inventory, an effective date, and a named accountable owner. Sustained adoption at 90, 180, and 360 days is no longer the objective. It becomes the evidence that all three steps actually took place.

Strategy does not fail because the board forgets to cancel what is old. It fails because the board never knew how much of the old remained alive. Governing execution begins with knowing what is actually being executed, and only then signing off on what must be cancelled.