Transform the Core

The Word

Walk into any enterprise transformation and ask what is being transformed. The answers will be long and varied. The technology stack. The process. The operating rhythm. The customer journey. The data architecture. The change muscle. The portfolio. The leadership. Each answer may be true. None of them is the same answer. That is the first problem.

Transformation has become a catch-all for nearly any initiative that is large, urgent, expensive, cross-functional, technology-enabled, or executive-sponsored. A strategy refresh is called transformation. A platform implementation is called transformation. A cost-reduction program, a shared-services move, a process-improvement portfolio, an AI pilot, a target-operating-model design, and a change-management campaign all claim the same label. Some of that work matters. Much of it is necessary. Each can wear the label. None of it is automatically transformation.

Language shapes governance. When every major initiative carries the same word, leaders lose the ability to tell the difference between activity that improves the current enterprise and change that alters how the enterprise actually works.

Executives inspect milestone progress. Boards review initiative volume. Investors underwrite value-capture plans. Platform providers deploy capability. Transformation offices report status. The activity is real. The transformation is often not.

The market has been measuring this for years. McKinsey cites roughly 70 percent of transformations falling short.¹ Bain’s 2024 research puts that figure at 88 percent.² BCG, on digital transformations, reports 70 percent.³ The methodologies differ. The signal does not: enterprises have become better at launching transformation than landing it.

The sharper question is the one boardrooms rarely ask. Where is the working operating model itself being transformed — not described, not optimized around, not re-platformed, but changed? That question usually produces silence. The silence is not accidental.

When everything is transformation, nothing is.

The Category Error

The category error hides in plain sight. Every discipline that surrounds the operating model assumes the operating model can absorb what the discipline asks of it. Strategic planning sets ambition. Enterprise architecture draws the future state. Program offices track milestones. Change management supports adoption. Lean and Six Sigma remove waste. Enterprise platforms digitize workflows. AI now promises to accelerate them. Each works at a useful altitude. Each presumes the Core can absorb what it asks. That assumption is the category error.

When the operating model is assumed rather than examined, every test that should distinguish transformation from activity gets replaced by something easier to measure. Initiative volume substitutes for change. Milestone completion substitutes for conversion. Platform go-live substitutes for capability. Cost achieved substitutes for value. Board updates substitute for governance. Each metric tells leaders something true. None catches the Monday-morning reality: the performance succeeds where the conversion does not. The customer still waits the same days. The exception still finds the same VP. The forecast still misses for the same reasons.

This is why a strategy can be coherent and still fail because the operating model cannot convert it. A target operating model can be elegant and still fail because the transition into working state is underbuilt. A transformation office can run disciplined governance and still fail because the work entering the portfolio was never validated against real capacity, margin, capability, or evidence logic.

The category error is not harmless. It weakens executive oversight, because leaders review activity instead of conversion. It weakens board governance, because directors see initiative volume without seeing what changed. It weakens investor diligence, because value-capture narratives are accepted before the operating model has been tested. It weakens platform investment, because capability gets deployed before the enterprise has specified what the Core must absorb. It weakens AI investment, because use cases multiply before workflow, decision rights, evidence quality, and correction loops are ready.

The result is a familiar pattern. Transformation everywhere. Transformation nowhere.

See Exhibit 1

Exhibit 1 draws the boundary. The test is not whether the work matters. The test is whether it changes the Core.

Every method works around the Core. Few examine it.

The Missing Subject

The subject most transformation work surrounds, and most rarely changes, is the working operating model. This paper calls it the Core. The name matters because most transformation literature names the operating model in forms that stay abstract: as ambition, as future state, as a deck. The Core is not the org chart, the technology stack, the process repository, the target-operating-model deck, or the initiative portfolio. Each of those may describe, support, or surround the enterprise. None of them is the Core by itself.

The Core is the conversion system: the practical architecture through which demand enters, work is classified into governable units, accepted work becomes executable commitment, execution produces evidence, evidence becomes correction, and correction creates value. Quote-to-cash, hire-to-retire, procure-to-pay, contract lifecycle, demand-to-delivery, issue-to-resolution — these are the lifecycles through which the enterprise actually moves. Big-T Transformation changes how these lifecycles work, not what they are called on an architecture diagram.

This is why a coherent strategy fails when the Core cannot convert it. Cost returns when the Core that produced it remains intact. Margin leaks when the qualification, pricing, scope, staffing, delivery path, and proof logic are not governed as one system. Each failure looks operational. Each is structural.

The Core is where strategy becomes operationally honest. A growth strategy becomes honest when demand can be captured, qualified, routed, resourced, priced, delivered, proven, and improved. An AI strategy becomes honest when use cases connect to classified demand, embedded workflow, governed decisions, proof-quality evidence, and correction loops. Margin, digital, and integration strategies follow the same rule: each becomes honest only when the Core can carry it.

See Exhibit 2

Every transformation adjacency depends on the Core. Strategy, change, platforms, AI, cost programs, and governance all require proof of how value actually moves through this system. Exhibit 2 names the playing field.

Strategy is a claim. The Core is the proof.

The Hidden Intelligence

Leaders are surrounded by intelligence that describes everything around the Core and almost nothing about its working state.

Most executive teams already receive market intelligence on demand and positioning, financial intelligence on performance and capital, customer intelligence on satisfaction and churn, competitive intelligence on strategic threats, workforce intelligence on talent and capacity, and increasingly AI-derived intelligence to summarize, predict, recommend, or detect patterns. Each source improves a different decision. Together, they can make the enterprise look well instrumented.

But most of that intelligence describes the world around the Core: markets, capital, customers, competitors, talent, capacity, patterns in the data. It does not show the working state of the Core itself.

The evidence was already there: in the delays, workarounds, handoff failures, hidden labor, weak proof, exception paths, and corrections that never closed. Those signals show how the enterprise actually converts intent into work, commitment, evidence, correction, and value.

The answers come from the Core itself, not from the people interpreting it. The principle has scholarly lineage: W. Edwards Deming made systems thinking foundational to modern management — a system produces what it is designed to produce, and the discipline that does not study its own system keeps generating outcomes it cannot explain.⁵ Forty years on, the principle remains underused at the strategy layer. Most enterprises still diagnose their Core through interview, opinion, or framework. The Core is visible in the work. Most governance is looking somewhere else.

The Core has been speaking. The enterprise measured everything else.

Voice of the System needs structure, or the signal becomes anecdote. GSDPI provides it. Get reveals how demand enters. Sort reveals how demand becomes a governable unit of work. Do reveals how execution actually happens — workflow, systems, decision rights, exceptions, handoffs. Prove reveals whether completion can be validated by evidence rather than narrative. Improve reveals whether evidence becomes correction, or the same pain recycles through the next initiative. GSDPI is not a methodology to install. It is the universally applicable grammar of how work moves.

See Exhibit 3

It makes three planes legible at once — horizontal (how actors move, where handoffs fail), vertical (each stage as a diagnostic surface), and systemic (the pattern across the whole). Exhibit 3 renders all three.

A local pain may be a process issue. A pattern of pain across actors, stages, and seams reveals an operating-model challenge no single function can see from inside its lane. That pattern surfaces hidden Enablement requirements: authority, decision rights, workstreams, proof obligations, and ownership altitude to bind transformation at the seams.

Cross-functional decision rights cannot be reallocated by a function head. No CFO rewrites Legal’s thresholds; no COO resets Sales’ qualification rules. That happens at an executive layer with authority across the lanes. Bain’s research corroborates it: enterprises that succeed often install a dedicated transformation officer for the work no function can own.²

Local pain is process. Pattern pain is the operating model.

The Undeclared Work

The strategy deck named the ambition. The portfolio deck named the initiatives. The platform deck named the capability. The change deck named adoption. The transformation office named the milestones. None of them named the work that holds those decks together when execution begins — the undeclared work transformation depends on, met as friction nobody planned for.

It is not soft work, not optional, not “support.” It is authority that must be assigned, decision rights that must be reassigned, capacity that must be released, and proof standards that must be installed. Each is concrete, executable, structural. Most appear nowhere on a transformation Gantt chart until the moment they block one.

Modern leadership scholarship completes the picture. Amy Edmondson’s research on psychological safety shows that organizations carry intelligence about their own operating reality that stays unspoken unless the conditions to surface it are engineered.⁶ Voice of the System gives the model a way to speak; psychological safety lets people corroborate what it reveals.

See Exhibit 4

Exhibit 4 separates the two requirements this section names: the Transformation Requirement (what must change in the Core) from the Enablement Requirement (what must be owned, funded, sequenced, governed, and installed so the change can land) — and where transformation actually fails.

Transformation lands on the work no function will name.

The Reckoning

The urgency is not abstract. Five forcing functions are now exposing operating-model debt from outside the enterprise. None of them is kind to decorative transformation.

• 01Artificial Intelligence

  Gartner finds 80 percent of CEOs expect AI to force medium-to-high change in operational capability.⁷ Yet only 39 percent report enterprise EBIT impact, even as 88 percent use AI — the value has not arrived.^8, 9 AI scales whatever operating model it enters: orphaned approvals, unclear decision rights, weak data, and proof gaps scale with it.

• 02Platforms

  Gartner finds 75 percent of ERP strategies are poorly aligned to business strategy, and 70 percent of recent ERP initiatives will miss their business case by 2027.¹⁰ Target Canada is the cautionary case: SAP did not fail — it made the product data right only about 30 percent of the time, contributing to a roughly $5.4B pre-tax loss.¹¹ At go-live, the actual state becomes impossible to hide.

• 03Cost

  BCG puts the odds of success at 20 percent for blunt reductions versus 80 percent when cost work is operating-model-led.¹² A task-layer productivity gain is not structural value until roles, capacity, sourcing, service levels, workflow, and proof are governed as one system. Otherwise the cost returns.

• 04Integration

  McKinsey finds that post-merger implementation of the combined operating model determines whether synergy targets are met.¹³ Two reporting structures can be combined on a Friday. Two operating models cannot. The enterprise runs as one conversion system, or as two wearing one logo.

• 05Boards

  NACD’s 2026 Governance Outlook reports that 60 percent of directors rank execution oversight as their number-one priority for 2026.¹⁴ Boards no longer ask for richer transformation dashboards. The question now is what management can defend on the next results call: did the company change, or did only the deck change?

The technology landed. The value did not.

The Test

Leaders need a standard the board, the executive team, the investor, the operating partner, and the transformation officer can all apply to any transformation claim. Not a framework. A test.

See Exhibit 5

Exhibit 5 names that test in five categories, each addressed to a different governance question, each answerable in plain language by the people who fund and oversee the work.

• 01Foundation

  Change the model, or redistribute value within it? The first cut.

• 02Traceability

  Does every funded initiative trace to evidence, a requirement, or a constraint?

• 03Evidence Chain

  Can each unit show whether the intake forecast held, and where value leaked?

• 04Value Assurance

  Did value actually move — leakage, variance, benefits, impact, ROI exposure?

• 05Defensibility

  Can leadership prove it to a board, investor, auditor, or regulator? The last gate.

None of these questions is new. Boards have always wanted to know whether the company changed. Investors have always wanted to know whether value got captured. What is new is the discipline that connects them: actual-state evidence at intake, requirements that separate transformation from enablement, an executive layer with authority to bind change at the seams, and a proof loop that closes correction back into the operating model.

The standard does not make transformation easier. It makes transformation harder to claim and easier to prove.

Either the Core changed, or it didn’t. There is no third answer.

The Charge

Our work across F500 enterprises, private-equity portfolios, post-merger integrations, and digital-first launches keeps surfacing the same fault line: capability deployed, conversion absent. Cost programs funded, operating models intact. AI piloted, decisions ungoverned. Integrations announced, two operating models still going to market under one nameplate. The activity changes. The Core does not.

This is a structural argument, not a rhetorical one. The Core is the thing that does not change when everything else does. It is also the only thing whose change makes the rest of the work mean anything.

So stop calling platform go-lives transformation. Stop calling AI pilots transformation. Stop calling process-improvement portfolios transformation. When the work changes how strategy becomes value at the conversion seams — that is transformation.

The test is in your hands now.

Transform the Core.

Sources

1. McKinsey & Company, Losing from day one: Why even successful transformations fall short, McKinsey Global Survey, Dec 7, 2021. Survey of 1,034 participants.
2. Bain & Company, 88% of Business Transformations Fail to Achieve Their Original Ambitions, press release, Apr 15, 2024. Transformation & Change Survey of 400+ executives; Chief Transformation Officer impact analysis.
3. Forth, Reichert, de Laubier & Chakraborty, Flipping the Odds of Digital Transformation Success, BCG, Oct 29, 2020. 70 transformations; 825 executives.
4. Krivkovich et al., A new operating model for a new world, McKinsey Quarterly, Jun 18, 2025. Cites the 2021 survey for the 30% strategy-to-performance gap.
5. W. Edwards Deming, Out of the Crisis, MIT Center for Advanced Engineering Study, 1986.
6. Amy C. Edmondson, The Fearless Organization, John Wiley & Sons, 2018.
7. Gartner, 80% of CEOs Say AI Will Force Operational Capability Overhauls, press release, Apr 23, 2026. Survey of 469 CEOs and senior executives.
8. McKinsey & Company, The state of AI in 2025, QuantumBlack, Nov 5, 2025. Global survey of 1,993 respondents across 105 countries.
9. Heric, Doddapaneni & Debarre, Your AI Budget Is Growing. Your Returns Aren’t, Bain & Company, Jun 1, 2026. Automation and AI Pathfinder Survey 2026.
10. Gartner, Latest Enterprise Resource Planning (ERP) Insights, accessed Jun 18, 2026.
11. Joe Castaldo, The Last Days of Target Canada, Canadian Business, Jan 1, 2016; Target Corporation press release, Jan 15, 2015.
12. Kelley, Meyer, Jahn & Baxter, Quadrupling Your Odds in a Cost Transformation, BCG, Dec 9, 2025. Survey of 2,080 respondents across six countries.
13. Kordestani, Kaetzler & Bernauer, Unlocking merger value through operating model design, McKinsey & Company, Feb 13, 2026.
14. van der Oord & Sikora, Boards Shift Their Focus to Execution, NACD 2026 Governance Outlook Report, Dec 10, 2025.