There is a difference between a CEO who sponsors a transformation and one who leads it. Most don’t notice the difference until it’s too late to fix.

A sponsor approves the plan, names a capable leader, funds it, and asks for a monthly update. It’s a reasonable way to run most initiatives. It’s also the most common way transformations fail at the top, because a transformation is the one effort where the CEO’s personal involvement isn’t a nice-to-have. It’s the mechanism.

Almost everything in a transformation can be delegated: diagnostics, planning, tracking, daily decisions. Four things cannot.

Make the reason personal, and repeat it. The reason for change can’t be communicated once in a town hall and considered handled. It has to come from the CEO, in their own words, repeatedly, with enough conviction that people believe they mean it. When the message comes from the transformation office, people hear a program. When it comes from the CEO, consistently, they hear a direction the company is actually taking.

Make the choices that create enemies. Every real transformation requires a few decisions that make someone powerful unhappy: a division closed, a senior executive asked to leave, a favored project defunded. A delegated leader can’t make these — they don’t have the standing to absorb the cost. When the CEO makes them visibly, the organization learns the transformation isn’t negotiable. When the CEO avoids them, it learns the opposite.

Lend authority, visibly and repeatedly. A transformation office has no inherent power. Its authority is borrowed from the CEO, and the loan has to be renewed in public. When a CTO calls a meeting and a senior executive doesn’t show, what happens next decides everything. The first time the CEO fails to back the transformation leader against a powerful executive, the authority is gone.

Change their own behavior first. The organization watches the CEO more closely than the CEO imagines. If the transformation demands faster decisions and harder accountability and the CEO doesn’t visibly adopt them, the message is: this applies to you, not me. The transformations that hold are the ones where the CEO models the new behavior before asking anyone else to.

I worked with a CEO who did everything the textbook recommends — strong leader, ambitious plan, proper funding, monthly reviews. The program stalled in five months, because every hard decision was routed to the transformation leader, who couldn’t make it. Executives learned they could wait out any demand by appealing quietly to the CEO. The turn came when the CEO removed a senior executive who’d been blocking the program. Nothing in the plan changed. What changed was the organization’s read on whether the transformation was real.

So if you want to know whether a CEO is leading or just sponsoring, watch what happens in the weeks they’re absent. A sponsored transformation slows the moment their attention moves. A led one keeps moving, because the hard decisions are already made and the direction is already real.

A CEO can delegate almost everything. The four things they cannot are the four that decide whether it works.

Luciano de Castro Carvalho — Chief Transformation Officer. São Paulo. 20+ years leading large-scale transformation and turnaround programs across 30+ countries. Former McKinsey & Company and Alvarez & Marsal.

Related articles

• Why Corporate Transformation Is No Longer Optional
• The Hidden Cost of Delayed Transformation

Contact

· Website · Linkedin · Newsletter

Time. Specifically, the eighteen to thirty months between when the operating problems were first detectable and when the transformation finally started.

That window is the most expensive line item in the entire investment thesis. It never appears in the deal memo.

The math of delay. A transformation initiated in month 1 of detection is preventive. Cash is intact, lenders are constructive, key talent is stable. The same transformation in month 18 is reactive. Cash reserves have been significantly depleted, covenants have tightened, and the strongest operators have started to leave. The operational moves are not fundamentally different. The same portfolio rationalization, the same operating model redesign, the same working capital discipline. But executing them eighteen months later typically costs three to five times more. Not because the work costs more. Because the optionality around the work has collapsed.

What collapses during the delay. Four things. Capital optionality: a transformation funded by unrestricted cash becomes one that depends on lenders who have lost confidence. Talent optionality: the leaders willing to drive change in month 1 are gone by month 18, and the bench has been drained of exactly the people the program needs. Customer optionality: strategic customers detect distress before boards do. They reprice, diversify, and quietly hollow out the franchise. Strategic optionality: in month 1 the sponsor can choose between turnaround, refinancing, partial divestment, or accelerated exit. In month 18, half of those options have disappeared, and the rest are priced at distress.

Why the delay happens. Three forces converge, and the pattern is predictable. Management optimism: portfolio CEOs are selected for confidence, and there is always a credible-sounding macro explanation for missing the plan. Sponsor reluctance to write down: acknowledging the problem forces a markdown conversation with LPs, so each quarter of deferral feels cheaper than it is. Asymmetric risk perception: acting early and being wrong looks like wasted resources, while waiting and being wrong looks like giving management a chance. The second narrative is socially safer. The first is financially safer. Most boards optimize for the social one.

What right looks like. The sponsors who avoid this pattern install operating partners with real intervention authority before the company needs intervention. They review portfolio companies on leading indicators, not just EBITDA, because by the time EBITDA reflects the problem the optionality is already eroding. And they treat transformation as positive-NPV insurance: a pre-emptive program that turns out unnecessary costs 100, while a reactive one that arrives late costs several times that. The expected value math overwhelmingly favors acting early.

The eighteen months nobody billed for is the line item that decides whether a fund returns 1.8x or 2.5x.

It is rarely too early to start a transformation. It is almost always too late by the time everyone in the room agrees.

Related articles

• Why Corporate Transformation is No Longer Optional
• Operating Model as Competitive Advantage

Contact

· Website · LinkedIn · Newsletter

A transformation is usually declared a failure about eighteen months in, in a meeting where someone says the strategy was right and the problem was execution.

It is a comfortable thing to believe. It protects the people who designed the plan and blames the people who delivered it. After two decades doing this work across more than thirty countries, I have found it is almost always false.

Most transformations are decided in the first ninety days, in four design decisions that get avoided or deferred. By the time execution begins, the outcome is set. Execution doesn’t cause the failure. It reveals it.

The reason for change was never made non-negotiable. “We need to be more competitive” is not a reason to ask thousands of people to change how they work. A real reason names a number, a threat, and a deadline. When it stays vague, the organization reads it as optional, and abandons it the first time the program meets resistance.

Scope was negotiated, not designed. Scope should follow the value. In practice it follows political capital. I once watched the most underperforming function in a company get kept out of scope because no one wanted to confront the executive who ran it. It turned out to be the single largest source of the problem the program existed to fix. When a leader wins the fight to stay out of scope, that is usually the clearest sign their function should have been first in.

The leadership team was never actually aligned. They attended the offsite and signed the memo. Then alignment turns out to mean four different things: the CEO thought it meant support, the CFO the targets, the COO the timeline, the head of people the change approach. The first hard trade-off dissolves it, because it was never tested against a real decision.

The incentives were left untouched. If the bonus plan keeps rewarding the behavior that created the problem, the program is finished before it starts. The strategy lives in the deck; the behavior lives in the compensation plan, and the compensation plan wins. My rule: if a team won’t redesign incentives in the first ninety days, I assume the program delivers under half its business case.

A ninety-day test. Can the CEO state the reason for change in three sentences with one quantified threat? Is any unit exempt from scope without a value-based reason? Have the top leaders pre-committed to three painful decisions? Have incentives been redesigned, with the board behind them? Two or more unresolved, and the outcome is visible before the first initiative launches.

Most CEOs who bring in transformation help think they are buying execution capability. They are usually buying the resolve to make four decisions they could not make alone.

So before launching, don’t look first at the plan. Look at the four decisions. The plan tells you what the organization intends to do. The four decisions tell you whether it was ever going to happen.

. . .

Luciano de Castro Carvalho — Chief Transformation Officer. São Paulo. 20+ years leading large-scale transformation and turnaround programs across 30+ countries. Former McKinsey & Company and Alvarez & Marsal.

Read more: Medium · Profile

Every board has a strategy. Most have a deck from a top-tier firm. The strategy is rarely the problem.

The problem is the operating model — the actual machinery of how the company decides, executes, allocates capital, and measures itself — has nothing to do with the strategy in the deck.

This gap is where competitive advantage accrues. Or doesn’t.

What an operating model actually is

Most executives use the term loosely. They mean the org chart. Or the process map. Or the IT architecture.

The operating model is the integrated system of five things:

Decision rights — who decides what, with what authority.

Resource allocation — how capital, talent, and attention flow to opportunities.

Performance measurement — what gets tracked and rewarded.

Capability investment — what’s built, bought, or ignored.

Cultural defaults — what people do when the playbook doesn’t tell them.

When these five are designed coherently around the strategy, execution happens. When they aren’t, the strategy stays in the deck.

Most underperforming portfolio companies have a reasonable strategy and an operating model designed for a different strategy three CEOs ago.

A 10-minute diagnostic

Before any analytical work, I ask the top eight executives — separately — three questions.

“Describe our strategy in three sentences.”

Materially different answers mean the strategy isn’t operating. It’s just communicating. There’s a difference.

“What are the three most important decisions you must make this quarter — and who has authority to make them?”

Unclear decision rights mean decisions happen in hallways, not in the model.

“If a senior leader proposed a $5M investment aligned with strategy but with no precedent, how would it get approved — and how long would it take?”

Two weeks means the model absorbs new moves. Six months means it can’t.

After 30+ engagements, the correlation between fast, clear, consistent answers and companies that execute well is close to 1.0.

Where operating model alpha comes from

Three design choices appear consistently in outperformers.

Decision rights pushed lower than peers. Not delegation for its own sake — pushed to where information is best, with explicit authority and accountability. Most companies over-centralize because senior leaders prefer optionality. The companies that win accept slightly more variance in exchange for materially more speed.

Resource allocation is dynamic, not annual. Quarterly capital reviews. Talent fluidity protocols. Project kill switches. Reallocation built into the DNA, not the budget cycle.

Performance management is leading, not lagging. Pipeline conversion, customer concentration trends, working capital trajectory. By the time EBITDA reflects a problem, behavior has been off-track for nine months.

The PE implication

The conventional value creation model: buy a company with a decent strategy, install operational improvements, ride EBITDA expansion to exit.

The differentiated model: redesign the operating model itself. Push decision rights down. Build continuous reallocation. Replace lagging KPIs with leading ones.

The first generates incremental EBITDA during hold. The second generates multiple uplift at exit — because the buyer is acquiring a better operating system, not just a better P&L.

In a market where deal multiples are compressed, the operating model is no longer the back office.

It is the asset.

Luciano de Castro Carvalho — Chief Transformation Officer. São Paulo. 20+ years leading PE-backed and sovereign-backed transformations across 30+ countries. Former McKinsey & Alvarez & Marsal.

Related articles

• Why Corporate Transformation is No Longer Optional
• The Hidden Cost of Delayed Transformation

Contact

· Website · LinkedIn · Newsletter

This essay originally appeared on Medium.
A CEO once told me, three months into a turnaround, that he wished he had started the transformation two years earlier.

He didn’t say it because it would have been easier.

He said it because, two years earlier, he still had the choice.

By the time we met, he didn’t.

The bank covenants were tight. The board had lost patience. Strategic clients were renegotiating terms downward. Transformation was no longer a strategic initiative — it had become the only remaining path to survival.

Over twenty years leading transformation and turnaround engagements across more than thirty countries, I’ve heard the same regret repeatedly:

Not that transformation happened.

But that it happened too late.

The Illusion of Stability

Most executives still operate under the assumption that the business model which worked over the last decade will continue working in the next one.

The data says otherwise.

According to research by McKinsey, cited by IMD Business School, the average lifespan of companies listed in the S&P 500 was 61 years in 1958. Today, it is less than 18 years. McKinsey projects that by 2027, 75% of the companies currently listed on the S&P 500 will have disappeared — bought out, merged, or gone bankrupt.

The numbers from the Fortune 500 are equally stark. According to AEI’s analysis, only 52 of the 500 companies on the original Fortune 500 list from 1955 — just 10.4% — were still on it in 2021. Nearly nine of every ten have gone bankrupt, merged, been acquired, or simply faded into irrelevance.

Innosight’s 2021 Corporate Longevity Forecast confirms the trend: average tenure in the S&P 500 has fallen from 30–35 years in the late 1970s to a projected 15–20 years this decade, with about half of today’s S&P 500 firms expected to be replaced over the next ten years.

Most of these companies didn’t disappear because of a sudden black swan event.

They disappeared because they failed to transform while transformation was still affordable.

The real danger is rarely disruption itself.

The danger is waiting too long to react to it.

Transformation is no longer a strategic option. It is the baseline cost of survival.

Transformation Is Not a Project

This is where many boards still get it wrong.

Transformation is not something companies do temporarily.

It is something resilient companies become permanently.

The organizations that survive long term treat transformation as an ongoing operating discipline:

• continuously simplifying operations;

• reallocating capital;

• pruning portfolios;

• redesigning incentives;

• and questioning whether yesterday’s model still makes sense tomorrow.

The companies that fail usually treat transformation as an event.

A three-year program.

A steering committee.

A consulting deck.

A slogan.

By the time the program ends, the market has already moved again.

A Simple Diagnostic

One of the first questions I ask CEOs during transformation diagnostics is:

If you had to remove 30% of your cost base within the next 18 months without destroying the business, where would it come from — and why hasn’t it come from there already?

The answers are usually revealing.

Leaders running resilient organizations typically know exactly where the inefficiencies are.

What they lack is not analytical capability.

It is political timing.

Leaders running fragile organizations often cannot answer the question at all.

Not because the inefficiencies don’t exist.

Every large company has them.

But because nobody inside the organization has truly been authorized to confront them.

That gap is often the difference between resilience and fragility.

The Four Conditions Behind Real Transformation

In my experience, successful transformations usually share four characteristics.

1. The CEO owns the transformation personally.

Not the CFO.

Not the transformation office.

Not external advisors.

The CEO.

Organizations quickly recognize whether senior leadership is genuinely committed to difficult decisions or merely delegating them.

2. The case for change is brutally specific.

“We need to become more agile” is not a transformation thesis.

“We operate with a structurally uncompetitive cost base and cannot sustain margins at current pricing levels” is.

Specificity creates urgency.

Generic language creates theater.

3. Incentives change before behavior does.

If compensation systems, KPIs, promotion criteria, and governance structures continue rewarding the behaviors that created the problem, transformation becomes performative.

Organizations always follow incentives more consistently than speeches.

4. Early wins become visible quickly.

Not for public relations purposes.

But to prove internally that the transformation is real, operational, and irreversible.

Most failed transformations collapse because at least two of these four conditions were missing from the start.

The most common failure is the first one:

A CEO who genuinely wants transformation, but still believes transformation can be delegated safely.

It rarely can.

What Is Really at Stake

Almost every transformation I have led started with the same hidden reality:

Someone inside the organization had already identified the problem years earlier.

And was ignored.

The cost of delay compounds quietly.

By the time boards finally decide to act, the available strategic options have narrowed dramatically — and the cost of generating the same value has multiplied several times over.

Transformation procrastination does not only destroy EBITDA. It destroys optionality.

And once optionality disappears, survival becomes the primary conversation.

The New Reality

Corporate transformation in 2026 is no longer a competitive advantage.

It is the baseline requirement for staying relevant.

The real question for leadership teams is no longer whether transformation will happen.

It is whether it will happen:

• proactively;

• with time;

• with capital;

• and with strategic flexibility;

or reactively, under pressure from banks, shareholders, auditors, or collapsing market conditions.

I have seen both scenarios repeatedly.

They produce very different outcomes.

One creates value.

The other simply tries to survive.

And by then, survival is often the best remaining objective anyone is still willing to discuss.

. . .

Luciano de Castro Carvalho — Chief Transformation Officer. São Paulo. 20+ years leading large-scale transformation and turnaround programs across 30+ countries. Former McKinsey & Company and Alvarez & Marsal.

Related articles

• Operating Model As A Competitive Advantage
• The Hidden Cost of Delayed Transformation

Contact

· Website · Linkedin · Newsletter