Turnarounds: how companies came back from the edge, and where the plans stall

The day Ford pawned the blue oval

In 2006 Ford borrowed $23.5 billion and put up almost everything it owned as collateral: its main U.S. factories, its shares in Ford Credit, and its trademarks, including the blue oval itself.

It was an odd thing to do before the storm arrived, and it hurt. Bill Ford later put it plainly.

“When we had to hock the blue oval, that was a very tough thing.”Bill Ford, as recorded in the Ford case

Then the storm came. Ford lost $14.6 billion in 2008, a record at the time. But the money was already in hand, and Ford rode out 2008 and 2009 without a federal bailout while its rivals could not. In 2009 it made $2.7 billion, its first annual profit in four years, and it has posted a profit every year since.

Even that story has an asterisk worth keeping: the 2009 profit leaned on a $3.4 billion one-time gain from cutting debt. A turnaround is rarely one clean decision. The comebacks in this guide are built from the same few moves: buy time, cut back to what works, fix how the business runs, and bet on something new.

What turnaround is

It is not the same as a reinvention, where a healthy company changes direction by choice. In a turnaround someone else sets the clock: lenders, the cash balance, or a buyer circling at a low price. The decisions have to work before the money runs out, and they have to keep working after the crisis has passed.

Why it is so hard

On paper the steps are obvious. In a real company, three things get in the way.

The ways companies did it

The comebacks on record use four kinds of move. Most used more than one.

What 42 companies did

Here is every decision on record, placed by the year it was made and marked by how it turned out. Hover a mark, or open the table, for the company and the outcome.

Of the 40 turnarounds that can be judged, 37 held, 2 held only in part, and 1 reversed.

That is not a success rate. A turnaround is hard to reverse because there is nothing to go back to: no one returns to the strategy that nearly killed the company. When a turnaround fails it is not undone. It stalls, or the next plan replaces it. So the cases to study are the ones marked partial.

Lego's reversal is a sale made to raise cash, not the plan itself. It sold its Legoland parks for about £250 million in 2005 to pay down debt; in 2019 the family's holding company joined a consortium buying back Merlin Entertainments, the parks' operator, in a deal valued at roughly £4.8 billion. The refocus on the brick stood; buying back what had been sold cost far more than the sale raised.

Intel's plan to build chips for other companies was paid for by a product business that was shrinking: revenue fell from $79.0 billion in 2021 to $53.1 billion in 2024. In 2023 the foundry business brought in $952 million and lost $482 million. Its architect, Pat Gelsinger, left in December 2024 with the factories still rising and the funding question unresolved.

Burberry's 2006 decision to walk away from profitable low-end sales of its check pattern grew revenue 102% over the tenure of the CEO who led it. But the tension came back, and every CEO since has needed a fresh recovery plan.

What the share price did

For many of the decisions there is a clean market record: the company's total return over the three years after the decision, against the S&P 500.

The chart shows how loosely a share price follows a recovery. Apple's 1997 deal with Microsoft was followed by three years in which its shares beat the index by 425.1 points. Target cleared its excess stock in one quarter in 2022, and its operating income rose 48.3% the following year, yet its shares trailed the index by 86.6 points. Morgan Stanley's 2009 move into wealth management became a lasting change in how the firm makes money, and its shares still trailed by 64.1 points. A turnaround can work and still outlast the market's patience.

How to run this decision

If you are weighing a version of this, here is what the record suggests doing, in order.

1. Raise the money before you need it

   A turnaround runs on time, and time is bought with cash. Borrow, sell or raise while lenders still see a company worth backing. Ford pledged its trademarks in 2006, before the storm, and that is why it could ride out 2008 and 2009 without a federal bailout.

   Seen in: Ford, 2006 · New York Times, 2009 · Marvel, 2005
2. Decide what to stop before deciding what to start

   Write the list of products, stores, parts and businesses that go. Lego cut its unique elements roughly in half. McDonald's stopped counting new restaurants as success. Sony sold its PC business and split off its TV business. Each made what remained simpler to fix.

   Seen in: Lego, 2004 · McDonald's, 2003 · Sony, 2012
3. Take the loss in one go

   Spread a write-down across many quarters and every quarter becomes a bad-news quarter. Target cleared its excess stock in a single quarter in 2022: operating income that quarter fell 87%, and the next year it rose 48.3%. Sony took its restructuring charges up front and turned durably profitable in fiscal 2016.

   Seen in: Target, 2022 · Sony, 2014
4. Pick one operating number customers can feel

   McDonald's chose sales at restaurants already open, a number a ribbon-cutting cannot fake. T-Mobile watched how many customers left each month. Chipotle watched the share of sales ordered digitally. A number like that tells you within months whether the fix is working.

   Seen in: McDonald's, 2003 · T-Mobile, 2013 · Chipotle, 2018
5. Don't fund the new engine only from the one that is shrinking

   When the new business depends on cash from the old one, every bad quarter in the old business becomes a question about the new one. Intel's foundry was paid for by a product business whose revenue fell to $53.1 billion in 2024. If the structure can't be avoided, agree at the start what happens when the funding dips.

   Seen in: Intel, 2021
6. Build it to survive a change of CEO

   Put the parts that must last into structures, contracts and capacity, not into one executive's strategy. Burberry's 2006 repositioning worked under the CEO who led it; every CEO since has needed a fresh recovery plan for the same tension. Intel's plan lost its architect before its factories were finished.

   Seen in: Burberry, 2006 · Intel, 2021
7. Separate the one-off from the recovery

   The first good year often leans on something that won't happen again: a gain on debt, an asset sale, a last batch of old stock. Ford's 2009 profit included a $3.4 billion gain from cutting debt. Adidas said roughly €200M of its 2024 profit came from selling off Yeezy stock and would not recur.

   Seen in: Ford, 2009 · Adidas, 2023

To work the numbers for your own case, read Turnaround plan: how to tell a real turnaround from a cost program.

Warning signs

What it has looked like when this decision was going wrong.

• The new recovery plan is aimed at the same problem as the last one.Burberry, 2006 and 2024
• The new engine is paid for by the business that is shrinking.Intel, 2021
• The first good year rests on a gain that will not recur.Ford, 2009 · Adidas, 2023
• The comeback is announced before it is earned.General Motors, 2010
• Success is measured from the bottom of the crisis, not from where the company stood before it.Starbucks, 2008

All 42 cases

Each case is one company at one decision: what was on the table, what it chose, why, and what followed, with every fact sourced.

Questions

What is a business turnaround?

The set of decisions that takes a company from a decline that threatens its survival back to a business that can pay its own way. It usually combines buying time, cutting back to what works, fixing how the business runs and betting on one new source of growth.

What are examples of successful corporate turnarounds?

Ford borrowed against its assets in 2006 and avoided a federal bailout. Lego cut its unique parts roughly in half and became the world's largest toymaker. Apple's 1997 deal with Microsoft restored faith in the Mac. Chipotle rebuilt its kitchens and digital ordering after 2018, and Microsoft bet on its cloud business in 2014.

How long does a turnaround take?

Longer than the headlines suggest. Hyundai backed its cars with a ten-year warranty from the 1999 model year, then spent two decades making the cars good enough to deserve it: it ranked 32nd of 37 makers in initial quality in 2001 and 10th by 2011. Morgan Stanley's move to wealth management took 16 years to look finished.

Why do turnarounds fail?

In this record they rarely get reversed. They stall: the leader who started the plan leaves, the new business is paid for by an old one that keeps shrinking, or the problem comes back and the next chief executive starts a fresh plan.