A breakup is a bet on what the parent becomes. Only 7 of 24 clearly paid off.
Netflix announced a spin-off called Qwikster, and canceled it just 23 days after announcing it. It is the breakup everyone remembers, because it never happened.
The ones that did happen are harder to judge. Take every spin-off, split and sale of a major business in our record whose ending is known, 24 of them. 7 clearly paid off. 6 went wrong: 3 went badly for the parent and 3 were called off before they were done. The largest group, 11, was completed with results that were mixed or never clearly stated.
A breakup is judged by what the parent becomes, not by the piece it lets go.
Clean wins, real regrets and a crowded middle
Of 29 separations, 5 are too recent to call and are left out of these counts.
The wins share a pattern: the parent came out of the separation as something clearer. When GE Vernova launched in April 2024, GE's stock was near a seven-year high and GE HealthCare's stock was up about 50% from its 2023 debut. After Merck spun off Organon, the parent leaned on its biggest drug: by 2025 Keytruda was generating more than $30 billion, nearly half of all of Merck's revenue. GE: breakup · Merck: Organon
Older wins look the same. PepsiCo let its restaurant chains go, and the snack business came to carry the corporation's profit pool. Hilton separated its real estate and timeshare arms, and the asset-light model it kept was followed by growing fee revenue and units through 2024. PepsiCo: restaurant businesses · Hilton: real estate and timeshare arms
The regrets share a pattern too, and it is the mirror image: the parent found it had given away something it needed. Palm split off its operating system and licensed it to rivals, then had to buy its future back and never regained the advantage it had when the two were one. The family that controls LEGO sold the LEGOLAND parks, then bought the same parks operator back fourteen years after selling it. Palm: PalmSource · LEGO: LEGOLAND theme parks
Completed, and still arguable
The middle is the largest group, and it is where the idea of this page does its work. IBM's spin-off of Kyndryl stood, and Kyndryl kept shrinking and posting losses after the split. That is a bad story for Kyndryl. It is not, on its own, a verdict on IBM, which is why the case is counted as mixed rather than as a failure. IBM: Kyndryl
Other middles turn on the relationship a parent keeps with the piece it lets go. About 2.5 years after eBay separated PayPal, eBay announced it would replace PayPal with Adyen as its main payments provider, and PayPal's shares dropped about 10% in after-hours trading. HP's 2015 split promised two equal twins; only one kept coming apart. eBay: PayPal · Hewlett-Packard: Hewlett Packard Enterprise
Changing your mind has a price, and it rises with time
3 separations in the record were called off, and the cost grew with the delay. Netflix reversed Qwikster in 23 days. Gap took three hundred and twenty-two days to change its mind about Old Navy, kept the business, and booked $189 million in charges for a divorce that never happened. At Toshiba, management's plan to split the company in two failed to win the required majority. Netflix: Qwikster · Gap: Old Navy · Toshiba: devices/semiconductor business
A called-off breakup is never free. The work of preparing two companies, the charges, and the months of uncertainty for the people inside them are spent either way.
The breakups too young to judge
5 more separations are too recent to call and are left out of the counts. Medtronic listed its diabetes business, retained about 90% of it, and promptly trimmed its EPS guidance after a $157 million charge the new company owed for development funding. Comcast separated its cable networks into Versant Media, while Peacock, which it kept, posted a $432 million quarterly loss on 46 million paying subscribers. Medtronic: diabetes business · Comcast: Versant Media
If the record is a guide, most of them will end up arguable rather than clearly right or wrong, and the question to watch for each is the one this page keeps asking: what did the parent become?
Judge the parent, and write the test down first
Before a separation, the debate is usually about the piece: what it is worth on its own, who will run it, how the market will price it. The record says the verdict comes from somewhere else. The breakups that paid off left a parent that was simpler and stronger; the ones that went badly left a parent missing something it needed.
So write down, before the announcement, what the parent should become and how you will know: the number, and the year you will check it. And be honest about the evidence: these are separations documented well enough to verify, which leans toward the famous ones. The quiet middle is, if anything, larger than it looks here.
Every deal, and how it ended
Each one in the words of its Decision Case. Open a group to read them.
Called off3
- Netflix: Qwikster (2011)
Netflix canceled the Qwikster spin-off just 23 days after announcing it.
- Gap: Old Navy (2019)
Three hundred and twenty-two days later, the company changed its mind, kept Old Navy, and booked $189 million in charges for a divorce that never happened.
- Toshiba: devices/semiconductor business (2022)
Management's plan to split Toshiba into two companies failed to win the required majority.
Completed, and the stated rationale paid off7
- PepsiCo: restaurant businesses (1997)
The chains stayed divested and the snack business came to carry the corporation's profit pool.
- Vodafone: Verizon Wireless (2013)
The sale closed and the stated proceeds and tax outcomes followed.
- Novartis: Sandoz (2023)
The divestitures completed and the focused business grew as the article documents.
- General Electric: breakup (2015)
FSOC voted on June 28, 2016 to rescind the designation of GE Capital Global Holdings, LLC as a nonbank systemically important financial institution
- Hilton: real estate and timeshare arms (2017)
The asset-light model stood and fee revenue and unit growth followed from it through 2024.
- Merck: Organon (2020)
By 2025 Keytruda was generating over $30 billion — nearly half of all of Merck's revenue.
- GE: breakup (2021)
by the time GE Vernova launched in April 2024, GE's stock was near a seven-year high and GE HealthCare's stock was up about 50% from its 2023 debut.
Completed, and the record reports it went badly3
- Pan Am: Pacific Division (1985)
The sale stood and the company continued its decline into bankruptcy and shutdown.
- Palm: PalmSource (2003)
Once the OS was split off and licensed to rivals, Palm had to buy its future back and never regained the fused advantage.
- LEGO: LEGOLAND theme parks (2005)
The family bought the same parks operator back fourteen years after selling it.
Completed, results mixed or not stated11
- Intel: XScale (2006)
Intel stayed out of mobile and its later subsidized attempt confirmed the exit; the cash cow held for years before a structural share slide.
- Alibaba: Alipay (2011)
Then, in late 2020, Chinese regulators suspended the IPO days before it priced, and over the following years forced Ant to restructure into something far more like a regulated financial holding company than an integrated super-app.
- Kraft Foods Inc.: North American grocery business (2012)
The split stood, revenue grew and diversified internationally, though a ~9% non-snacks tail remains.
- The Washington Post Company: newspaper publishing businesses (2013)
The sale closed, the paper changed owners, and the company continued as an education-focused Graham Holdings.
- Hewlett-Packard: Hewlett Packard Enterprise (2015)
HP's 2015 split promised two equal twins — only one kept coming apart
- eBay: PayPal (2015)
On January 31, 2018 — approximately 2.5 years after the separation — eBay announced it would replace PayPal with Dutch processor Adyen as its primary payments provider; PayPal shares dropped ~10% in after-hours trading
- DowDuPont: three-way split (2019)
Both separations landed inside the promised 18-to-24-month window, delivering the planned three-way split.
- Yum! Brands: Yum China (2016)
The separation was completed and outside investors put capital into the standalone Yum China.
- IBM: Kyndryl (2021)
The separation stood and Kyndryl continued shrinking and posting losses after the split.
- Daimler: Daimler Truck (2021)
The truck arm listed and the parent renamed itself, both completed as chosen.
- J&J: Consumer Health (2021)
The spin proceeded through IPO and a subsequent announced split-off of the majority stake.
Too early to call5
- Comcast: Versant Media (2026)
Peacock reached 46 million paying subscribers and posted a $432 million quarterly loss; Comcast co-CEO Cavanagh told analysts 'we're not assuming this gets easier any time soon.'
- DuPont: Electronics (2025)
DuPont announced it is targeting November 1, 2025 for the completion of the separation of its Electronics business as Qnity Electronics, Inc.
- Unilever: ice cream (2025)
TMICC debuted on 8 December 2025 at €12.96/share implying a market cap of ~€7.93bn, controls €7.9bn revenue, claims 21% global ice cream market share, and operates the world's largest fleet of ice cream cabinets (3 million units).
- Medtronic: diabetes business (2026)
Medtronic retained about 90% of it and promptly trimmed FY2026 EPS guidance by twelve cents after a $157 million charge MiniMed owed Blackstone for development funding on its Flex insulin pump.
- WBD: Global Networks (2025)
The article states the split was announced but reports no dated outcome after it completed.
Asked about spin-offs and breakups
Do spin-offs work?
Some do, clearly. Of 24 spin-offs, splits and sales of a major business with a known ending, 7 (29%) paid off as intended, 3 went badly for the parent and 3 were called off. The largest group, 11 (46%), was completed with results that were mixed or never clearly stated.
How often is a company breakup called off?
In the record, 3 of 24 separations with a known ending were called off before they were completed. Every one of them cost the company time, and some cost real money.
How do you know if a spin-off worked?
Judge the parent, not the piece. A separation counts as having gone badly here only when it went against the parent's own stated reason: the parent lost value or declined because of it, or later bought the business back. The separated business struggling on its own does not count.
Which spin-offs paid off?
PepsiCo: restaurant businesses, Vodafone: Verizon Wireless, Novartis: Sandoz and General Electric: breakup, among 7 in the record whose stated rationale paid off.
Is the company worth more in pieces, or only worth reporting in pieces? The The Separation pack works it through, with the cases behind it.
Stratrix analysis of 29 separations (spin-offs, split-offs, breakups and sales of a major business) in its verified Decision Cases, each placed by its case's own words; a separation told in more than one case counts once. 5 too recent to call are left out of the shares. As of November 5, 2026.
Keep going
Tools to try
- Ansoff move finderFramework: Ansoff Matrix
- Business model canvas builderFramework: Business Model Canvas
- Change readiness checkFramework: Kotter's 8-Step Change Model
- Five Forces scorerFramework: Porter's Five Forces
- SMART goal builderFramework: SMART Goals
- SWOT builderFramework: SWOT Analysis
More from Stratrix
The Business Model Wizard
Make the seven choices that define your model, test your numbers against real companies, and meet the rival AI would build against you.
Free · about twelve minutesBuild your Model Thesis →Test a decision, free
Set a live decision against the companies that made it before you.
AcquisitionsThe Distance TestDivestituresThe Divestiture TestDiversificationThe Diversification TestMake or buyThe Make-or-Buy TestWeekly email
Stratonomics, every Tuesday
One data story every Tuesday: what happened, what it means, and the figures you can quote. Free.