What does this article cover?
- What should we judge them on?
- The reputation that most exceeds the record
- The record that exceeds the reputation
- What happened to the famous three?
- Why do reputations drift from records?
- What does the criterion actually reward?
- Who should this article send away?
- Where does this break down?
- People also ask
- What I sell, and what I don’t
- About the Stagnation Assassin
Ranking executives on impressiveness produces a list of people who were good at being famous. This one runs on a single checkable question applied to each name: a decade after they left, was the company still better off. That criterion is harsh, it produces no clean heroes, and every negative judgment below belongs to somebody who published the analysis with the numbers attached rather than to me.
I run the numbers on protecting people in a downturn in my podcast episode Bob Chapman: No-Layoff Leadership Decoded.
What should we judge them on?
Whether it survived them. Not the share price on the day they retired, not the magazine covers, and not the memoir. The state of the business a decade after the successor took over, which is the only measure that separates a turnaround from a well-timed exit.
Three reasons this is the right test and one reason it is uncomfortable.
It is checkable, which almost nothing else about executive performance is. It is hard to game, because nobody can manage a number they no longer control. And it captures the thing a board actually wants when it hires somebody, which is a business that works rather than a period during which the business appeared to work.
The discomfort is that it takes ten years to apply, which means it is useless at the moment of hiring and devastating in retrospect. That asymmetry is the whole subject.
The reputation that most exceeds the record
Jack Welch, by a distance, and the case against him is published rather than mine. Fortune named him Manager of the Century in 1999. Twenty years later General Electric was broken into three companies, and the numbers underneath the legend had been examined by then.
The record as it was understood. Welch ran GE from 1981 to 2001 and grew its market capitalisation more than thirtyfold. In 1993 GE became the most valuable company on the American stock market. He was, by consensus, the executive other executives studied.
The record as reassessed. Jeffrey Sonnenfeld of Yale, writing in Fortune when the breakup was announced, put the comparison plainly: over the same eighteen-year window, Reuben Mark at Colgate-Palmolive delivered total shareholder return of 3,372 per cent against Welch’s 2,504. Extend the window through his successor and GE returned 745 per cent from 1984, against 3,385 for the S&P 500 and 10,254 for Colgate. GE’s stock fell 25 per cent in Welch’s final year, the year his memoir appeared.
The mechanism that did not survive. GE Capital generated a reported 45 per cent of group profits and functioned as a tool for smoothing quarterly results. That worked while credit was cheap and industrial cash flow was strong, and it was the thing that made GE vulnerable when neither was true. The financial crisis found the structure rather than the operations.
And the succession. This is the part I find most damning and it is the least discussed. Welch’s protégés went on to run Home Depot, Chrysler, 3M and Boeing, and the commentary on that cohort’s subsequent records is not kind. A leader who produced a generation of leaders is being judged on more than one company.
Over the same eighteen years, a less celebrated chief executive at Colgate-Palmolive delivered 3,372 per cent total shareholder return against Welch’s 2,504. That comparison is published, and it is not mine.
The record that exceeds the reputation
Lou Gerstner, who died in December 2025, took IBM in 1993 with roughly an eight billion dollar annual loss and cumulative losses near sixteen billion over three years, refused the board’s plan to break it up, and left a company whose successor was a career insider.
What he inherited. An outsider from RJR Nabisco with no technology background, arriving after the resignation of John Akers, at a company whose survival was openly doubted in print. Revenue had fallen from a late-1980s peak and the mainframe business was collapsing under distributed computing.
The decision. Most accounts identify one: the refusal to dismantle the company. The consensus prescription was to break IBM into pieces and Gerstner concluded that its scale and breadth were the asset, repositioning it as an integrator. IBM Global Services became the largest information technology services business in the world within a decade, which is the concrete form of that bet.
The numbers. Market value from roughly $29 billion to $168 billion. Share price from around $11 to $90 at his 2002 retirement. And the detail that separates this from most cost-led recoveries: IBM ended his tenure employing 65,000 more people than when he arrived.
The criticisms, which are real. Published analyses hold that revenue was flattered by a contested pension reallocation and the share price supported by substantial buybacks. He arranged an $8.9 billion write-off and ordered 35,000 layoffs. Critics argue he imported a superstar-executive compensation culture IBM had never had, and that research spending stagnated against peers in the years that followed.
The survival test. A qualified pass rather than a clean one. Succession went to a long-time IBMer and every major unit was run by insiders, which is what a durable institution looks like. IBM’s subsequent two decades contain a long revenue decline that the services strategy did not prevent. The structure held; the growth did not.
What happened to the famous three?
Mixed, and the criterion treats all three differently from the way the case studies do. I have written elsewhere about the specific mechanisms Mulally, Bethune and Iacocca each used to rescue their companies. This section asks a different question of the same three, and two of them answer it badly.
Iacocca at Chrysler. The operational work was genuine, with costs cut roughly a third and the break-even point halved. Chrysler was sold to Daimler, separated again, and entered bankruptcy in 2009. A turnaround that required an Act of Congress to compel concessions from unions and creditors produced a company that needed rescuing twice more. On this criterion that is a failure, and it is compatible with Iacocca having been good at his job.
Bethune at Continental. The airline went from near a third bankruptcy to profitability and stayed there, then merged into United in 2010. Consolidation is not a verdict on the operating work, which is why I place this case as overtaken rather than failed. The incentive mechanism he built is still the most transferable thing in this article.
Mulally at Ford. The strongest case on this criterion, because Ford alone among the American automakers avoided federal money and remained independent. What followed is contested and the company has had three chief executives since. I am not going to adjudicate Ford’s last decade in a paragraph; the survival test gives him a pass on the specific thing he was hired to prevent.
Notice what the criterion does here. It separates the quality of the work from the durability of the result, and it declines to collapse them, which is what every ranking of executives does by default.
Why do reputations drift from records?
Because the reputation is fixed while the record is still being written. A leader’s standing is established during their tenure and at the moment of departure, and almost nothing afterwards revises it, so the two measures separate quietly for a decade and nobody reconciles them.
Four mechanisms drive the drift and all four are structural rather than anybody’s fault.
The assessment happens too early. Books, awards and magazine covers arrive while the person is still in post, which is necessarily before the evidence exists. Manager of the Century was awarded two years before the tenure ended and twenty-two before the company was dismantled.
Successors absorb the blame. When a business declines after a celebrated leader leaves, the available explanation is that the successor was weaker, and that explanation is always partly true, which makes it hard to dislodge. The alternative reading, that the predecessor left something fragile, requires evidence nobody collects.
The memoir is written by the subject. Most of what circulates about these careers originates with the person, their authorised chronicler, or a business school case written with their cooperation. That is not dishonesty; it is a systematic absence of anybody with an interest in the other account.
And nobody revisits. The genuinely strange thing about this field is that reassessment is rare and newsworthy when it happens. Sonnenfeld’s piece was published because GE announced a breakup; without that trigger the numbers would have sat unexamined.
Which suggests a habit worth adopting for any leader you currently admire: find out what happened at the company they left, and notice how hard the answer is to find.
There is a version of this that matters inside your own business rather than in business history. The people your organisation holds up as its best operators were assessed the same way, on performance during tenure, and almost nobody checks what happened to the plant, the region or the function after they were promoted out of it. That check is available to you, it costs an afternoon with somebody in finance, and it will occasionally tell you that your most celebrated internal turnaround was a set of deferred costs that arrived eighteen months later under somebody else’s name. I would rather find that out than not, and most companies never look.
What does the criterion actually reward?
Institutions over performances, consistently. Every case that passes this test involved building something that continued to run without the person who built it, and every case that fails involved a mechanism depending on conditions the leader did not control and could not hand to a successor.
Three things separate the two groups, offered as observations rather than findings.
Succession from inside. Gerstner left a career insider and a bench of insiders running every major unit. Welch produced a cohort celebrated on departure and criticised in retrospect. A leader who cannot be replaced from within has not built anything, and the succession decision is the clearest available evidence of what they spent their tenure doing.
Structural rather than financial mechanisms. Financial engineering works until the conditions change and then fails all at once, which is the GE Capital story. An operating change survives a downturn because it is not contingent on one.
Whether anybody else could run it. The Continental incentive mechanism works because it is a rule, not a relationship. Anybody could operate it. That is the property to look for and it is almost never what a board is buying when it hires a name.
Who should this article send away?
Anyone hiring for a reputation. Every case here shows the gap between reputation and record running in some direction, and the direction is not knowable at the moment of appointment, which is precisely when boards rely on it most heavily.
The practical substitute is unglamorous. Ask a candidate what happened at their previous employer in the three years after they left, and whether the people they promoted are still there. Neither question is about them, which is why both are informative.
A second group: anyone who wants a leader to emulate. The three things this criterion rewards are institutional rather than personal, and there is nothing in any of these careers to copy that is not also available from reading about the mechanism.
Where to go instead, and all of it beats my summary. Sonnenfeld’s Fortune commentary is short, numbered and written by somebody who has studied executives professionally for decades. David Gelles at the New York Times has done the extended work on Welch’s influence on the wider culture of management. And Gerstner’s own account of IBM is worth reading with the published criticisms of it alongside, which is a better exercise than accepting either.
Where does this break down?
Nobody has applied this test to my own record, including me. The bio at the foot of this page claims three billion dollars of shareholder value across four companies, and I have never published what happened at any one of them in the decade after I left it.
That is the same omission I have just used against other people, and I am not going to pretend the asymmetry is comfortable. The honest position is that my claim is of the type this article treats sceptically, and a reader is entitled to apply the criterion to it.
Three further limits. The criterion is unfair to anybody whose company was consolidated or sold for reasons unconnected to their work, which is why Continental is placed as overtaken rather than failed, and that placement is a judgment call somebody could reasonably dispute. A decade is arbitrary; five years would flatter more people and twenty would condemn nearly everybody. And this article covers five people, all American, all men, all at large public companies, which is a selection driven by what leaves a documentary record rather than by where good turnaround leadership happens.
The claim I would keep: the question of what happened after somebody left is checkable, and the near-total absence of it from how executives are assessed is the reason reputations and records drift so far apart.
Ask a candidate what happened at their previous employer in the three years after they left, and whether the people they promoted are still there. Neither question is about them, which is why both are informative.
People also ask
Who is the most overrated turnaround CEO? Jack Welch, on the published evidence. Yale’s Jeffrey Sonnenfeld showed a less celebrated Colgate-Palmolive chief executive outperformed him over the same period, and GE was broken into three companies in 2021.
Who is the most underrated? Lou Gerstner, who refused the consensus advice to break IBM apart, took it from roughly an $8 billion annual loss to a market value near $168 billion, and left with 65,000 more employees than he found.
How should turnaround leaders be judged? By whether the result survived them. The state of the business a decade after the successor took over is checkable, hard to game, and the only measure that separates a turnaround from a well-timed exit.
What separates the leaders whose work lasted? Succession from inside, operating rather than financial mechanisms, and changes that work as rules anybody could run rather than as relationships only they could maintain.
What should a board ask instead of checking reputation? What happened at the candidate’s previous employer in the three years after they left, and whether the people they promoted are still in place.
What I sell, and what I don’t
I don’t sell consulting. My employment contract doesn’t allow it, so there is no engagement waiting at the end of this article and no proposal coming. Two things here are commercial: I speak for a fee, and I sell books. Everything else, including the audit, is free, and it stays free because it has to be. If you are hiring into a turnaround and want a second read on the two questions above, ask at my contact page.
About the Stagnation Assassin
Todd Hagopian is a Fortune 500 transformation executive who has generated $3B in shareholder value across Berkshire Hathaway, Illinois Tool Works, Whirlpool, and JBT Marel, where he serves as Global VP of Product Strategy and Marketing. Known as The Stagnation Assassin, he is the author of two published books, The Unfair Advantage: Weaponizing the Hypomanic Toolbox and Stagnation Assassin: The Anti-Consultant Manifesto. His blog is published in 15+ languages and read by operators worldwide. Bring him to your stage via his speaking page or connect with him on LinkedIn.
The talk version asks a room to name a famous turnaround chief executive, then asks what happened to the company ten years later. Usually nobody knows, which is the point. Formats and fee are on my speaking page.