FICTION · An Unfair Advantage Intervention
Fiction. Real history. Sourced below.
In the first half of 2026 LEGO paid DKK 1.9 billion to buy 29 LEGO and LEGOLAND Discovery Centres from Merlin Entertainments, the company it sold LEGOLAND to in 2005. The signature move of the most famous focus turnaround in business was reversed twenty-one years later, in the middle of a record year, and almost nobody has noticed.
I have cut a part count in anger, on a line, with the people whose parts I was cutting standing in the room, so this case is not academic for me. What I want to correct is the lesson rather than the story. Everyone repeats that LEGO proves focus beats diversification. LEGO’s own last decade contradicts that sentence, the company’s leadership has said so publicly, and the more useful reading is available in the record if you follow it past 2014.
Full fiction disclosure and sourcing note
Maria Gonzalez is a fictional character from The Unfair Advantage by Todd Hagopian, analyzing the documented history of the LEGO Group from the present day. She has never met, advised, or corresponded with anyone named here, and no dialogue, thought, or motive is invented for any real person. Quoted words are documented public statements. Facts come from the public record, including LEGO’s own results announcements, and are accurate as of 2026; the analysis is invented. LEGO reports in Danish kroner and is privately held, so figures are given as the company reports them.
- The Count
- The Killers
- The Distinction
- The Relapse
- The Repurchase
- By the Numbers
- The Monday Morning Playbook
- Frequently Asked Questions
- About the Stagnation Assassin
The Count
By 2003 LEGO was losing roughly a million dollars a day, sales were falling around thirty percent, the operating margin had gone from about eighteen percent in the late 1990s to about two, debt stood near $800 million, and the company was reportedly some three months from insolvency.
The cause was not neglect. Through the 1990s, convinced that children were moving to screens, LEGO expanded into theme parks, clothing, watches, jewellery, television, video games and non-brick action figures. The unique-element count passed thirteen thousand. Internally, nobody could say reliably which products made money, and some complex electronic sets were selling below their cost to produce.
That element count is the part I would put in front of any operations team, because it is the physical expression of the disease and it is measurable. Every additional unique part is a mould, a validation, a stocking location, a forecast, a changeover, and a place where a schedule can break. None of that appears as a line item. It appears as overhead nobody can allocate and as a plant that is inexplicably harder to run than it was five years ago.
The Killers
The 3-S Method names three killers of capacity: changeover time, organisational complexity, and operational orthodoxies. LEGO’s was unambiguously the second, and the method matters here more than the diagnosis, because 3-S is a procedure you run rather than a condition you reach.
Jørgen Vig Knudstorp took over in October 2004, the first chief executive from outside the family, and ran it. Unique elements were roughly halved, back toward seven thousand. The loss-making LEGOLAND parks were sold in 2005 to Merlin Entertainments, then Blackstone-owned. Clothing, in-house game development and other ventures were shed or outsourced. The organising test was whether a thing strengthened the brick or distracted from it.
The free-fall stopped in 2005 and profitability returned in 2006, and that part of the story is told accurately everywhere. What gets lost is that the sequence, sketch the actual state, streamline what does not belong, then solve the remaining system, is a method rather than an event. Organisations treat it as an event because events end, and a discipline that never ends is much harder to fund and much harder to praise.
The Distinction
Here is the correction that matters most, and the reason the popular lesson is not merely incomplete but misleading. If LEGO’s lesson were that focus beats diversification, the following decade would be inexplicable, because LEGO grew fourfold on a wave of new ventures: licensed sets built on outside properties, a successful video game franchise, LEGO Ideas turning adult fans into designers, and a feature film in 2014.
That is diversification by any honest definition. It worked because every one of those ventures ran through the brick. The licensed products were brick sets. The games were about building. The film was ninety minutes of the core product. The failed ventures of the 1990s ran the other way: clothing, watches and non-brick figures were LEGO attempting to be something else.
So the distinction is leverage against abandonment, not focus against expansion. And it changes the advice materially. Tell a struggling executive that focus beats diversification and you may talk them out of precisely the core-leveraging expansion that would grow them. The test is not how many things you do. It is whether each of them makes the central thing stronger.
The Relapse
In 2017 LEGO reported its first revenue decline in thirteen years and cut roughly 1,400 jobs from a workforce that had grown to around 18,200. The diagnosis came from Knudstorp himself, by then chairman, who attributed it to an organisation that had accumulated too many layers and overlapping functions during a decade of rapid growth.
Read that plainly. The architect of the most celebrated complexity purge in modern business announced, thirteen years later, that the complexity had grown back, in his own company, on his watch. Not because the lesson was wrong or badly learned, but because growth is a complexity generator and the discipline had lapsed.
And it is still being run. As recently as June 2026, the current chief executive Niels Christiansen was publicly walking through how LEGO built too much complexity and how it got back on track, and observing that the company is now probably competing more with the entertainment industry than with the toy industry. LEGO’s own 2025 results statement describes strengthening its supply chain to manage growing complexity, in the most profitable year the company has ever had. That is what a chronic discipline looks like from the outside: the subject keeps coming up when nothing is wrong.
The Repurchase
Which brings me to the fact I have not seen discussed anywhere, and I think it is the most interesting thing in the case.
In the first half of 2026, LEGO reported revenue up 21 percent to a record DKK 41.9 billion and operating profit up 22 percent to DKK 10.9 billion. In the same period it invested DKK 1.9 billion to acquire 29 LEGO and LEGOLAND Discovery Centres from Merlin Entertainments. Merlin is the company it sold LEGOLAND to in 2005, in the transaction that symbolised the entire turnaround.
I want to be careful and precise rather than clever about this. Discovery Centres are small-format indoor attractions, not the capital-intensive destination parks that were bleeding the company in 2004, and the two are not the same asset class. So this is not simply LEGO undoing its own decision.
Run the actual test instead. Does a branded indoor attraction full of bricks strengthen the core or distract from it? A reasonable case says it strengthens it, and belongs alongside the film and the video games as expansion that runs through the product. A reasonable case also says that attractions are property, that property is exactly the category that nearly killed the company, and that “it runs through the brick” is a test elastic enough to justify almost anything once you are winning. I do not know which is right and I would not pretend to from outside. What I would say is that this is precisely the moment the method exists for. Twenty-one years and one record half-year after the sale, the company is buying attractions from the same counterparty, and the only honest position is that the question is open and being answered in real time.
By the Numbers
- 13,000 to about 7,000 — unique elements before and after the turnaround, the measurable form of the disease.
- Roughly $1 million a day — the cash burn in 2004, against debt near $800 million and a margin down from about 18 percent to about 2.
- 2005 — LEGOLAND sold to Merlin Entertainments as part of the purge.
- 1,400 jobs from about 18,200 — the 2017 relapse, after the first revenue decline in thirteen years.
- DKK 41.9 billion, up 21 percent — H1 2026 revenue, a record, with operating profit up 22 percent.
- DKK 1.9 billion for 29 Discovery Centres — bought back from Merlin in 2026, twenty-one years after the sale.
The Monday Morning Playbook
Four things a sitting operator can do this week:
- Count something physical. Complexity hides in overhead and shows up in unit counts. Pull your unique part, SKU or configuration count for the last five years and look at the slope. If it only goes up, you already have your answer.
- Apply the leverage test to additions, not just to cuts. The question for any new product, market or venture is whether it makes the central thing stronger. Most failed expansion fails that test on the day it is proposed, before any money is spent.
- Verify the core before you concentrate on it. Simplification multiplies an advantage that already exists and cannot manufacture one. Concentrating resources on a mediocre core produces a more efficient decline.
- Put simplification on a calendar. Growth regenerates complexity, so treat the purge as a recurring procedure with an owner and a date rather than a crisis response. LEGO ran it in 2004, needed it again by 2017, and is still discussing it in a record year.
Frequently Asked Questions
Why did LEGO almost go bankrupt?
Through over-expansion away from its core. Fearing that children were moving to digital play, LEGO diversified into theme parks, clothing, watches, television and non-brick figures during the 1990s, pushing its unique-element count past thirteen thousand. By 2003 sales were falling around thirty percent, debt approached $800 million, and the company was losing roughly a million dollars a day.
How did LEGO turn itself around?
Under Jørgen Vig Knudstorp from October 2004 it roughly halved its unique-element count, sold the LEGOLAND parks to Merlin Entertainments in 2005, and shed or outsourced non-core ventures, using a single test of whether an activity strengthened the brick. Profitability returned by 2006 and revenue grew fourfold over the following decade.
Does LEGO prove that focus beats diversification?
No, and its own record contradicts that reading. LEGO’s growth decade was built on new ventures including licensed sets, video games, LEGO Ideas and a feature film. The difference is that those ran through the brick, while the failed ventures of the 1990s led away from it. The distinction is leverage against abandonment rather than focus against expansion.
Did LEGO’s complexity problem come back?
Yes. In 2017 it reported its first revenue decline in thirteen years and cut about 1,400 jobs, with Knudstorp attributing it to layers and overlapping functions accumulated during rapid growth. The subject remains live: the current chief executive was still discussing how the company built too much complexity as recently as June 2026, in the middle of record results.
About the Stagnation Assassin
Todd Hagopian writes as The Stagnation Assassin on complexity, product-line simplification, and the operating disciplines that have to be run repeatedly rather than achieved once. He is VP of Product Strategy and Marketing at JBT Marel and the author of the Stagnation Assassin series.
Growth is a complexity generator. Run the purge once and declare victory, and you have not solved the problem, you have scheduled it.
The 3-S Method and the rest of the capacity frameworks are set out in The Unfair Advantage.