Kodak: a defensible decision and a forecast that moved too slowly
A good decision has an expiry date.
Kodak invented the digital camera in 1975 and stayed with film. With what was known in 1981, that was defensible. The map shows the same decision at three points in time: what it depended on, when its reasoning stopped holding, and why Kodak's forecasts showed that too late.
An invention without a market and a study with a deadline
In 1975 the Kodak engineer Steven Sasson built the first digital camera: 3.6 kilograms, 100 by 100 pixels, 23 seconds to store each image [1]. Management saw no market for it [1]. Kodak held about 90 percent of the film market in the 1970s [2], and the prototype was far from a product that could be sold.
In 1981 Sony announced a camera without film. Kodak's head of market research, Vince Barabba, had the implications studied. By his later account the study reached two conclusions: digital photography can replace film, and Kodak has roughly ten years to prepare [3]. It also named what the pace depends on: the price of the equipment, the quality of images and prints, and how well camera, display and printer work together [3].
Kodak stayed with film and did research on sensors and professional cameras on the side. In 1996 revenue peaked at just under 16 billion dollars [5]. In 1998 the company planned more than one billion dollars for China [5]. The annual report for 2001 still says the film business is developing in new markets and is strong [6].
A decision that stayed in force every year
The decision of 1981 was: film remains the core business, digital technology is followed alongside. It was never reviewed in one piece. It stayed in force year after year while its basis changed.
From 1999 the camera manufacturers' association counts digital cameras: 5.1 million in the first year, 10.3 million in the second, 14.8 million in the third [4]. In 2002, at 24.6 million, they overtook film cameras, of which 23.7 million were still shipped [4][7].
On 25 September 2003 chief executive Daniel Carp announced the change of course: up to 3 billion dollars for investments and acquisitions in digital, and a dividend cut from 1.80 to 0.50 dollars [11]. Between the study and that day lay 22 years.
The same arguments, a time horizon, new evidence
A decision about the core business does not apply to the current year. Each row of the map therefore asks what happens over the next five to ten years. The 1981 study reasoned that way: it gave film about ten years [3]. The top row of the map carries this horizon in its name.
Over such a horizon the direction counts for more than the level. In 1998 film was doing better than ever. The direction had already turned: film cameras peaked in 1997 at 36.7 million units and then fell four years in a row, to 27.6 million in 2001 [7]. The question to ask of such a trend is whether a reversal is plausible. For that, something would have had to put film back ahead of the digital camera. The figures showed the opposite: digital cameras grew strongly three years running [4].
The four arguments were fixed in 1981, and none was added up to 2002. Their assumptions were uncertain, but they were named: the study listed price, image quality and the interplay of the devices [3]. With every new figure, the plausibility of a row changes. The map makes this movement visible long before a single year delivers proof.
Kodak did not have to be first in the new market. But a group with chemical plants, photo labs and tens of thousands of employees does not change course within a year. The restructuring Kodak announced in September 2003 was still under way in 2005, with up to 25,000 fewer jobs [11][13]. A company that needs several years to restructure has to act when it is foreseeable that the balance will turn; it cannot wait until the other side is ahead.
The same map at three points in time
The map poses the same choice at three points in time: change course now or stay with film. Changing course means steering the profits from film into a second business and shrinking the film business in a planned way.
The four rows and their weights stay the same in every state. What changes is the plausibility: what the figures of the respective year support.
F1: steer the profits from film into a second business and shrink film in a planned way. F0: film remains the core business, digital technology is followed alongside.
1981: the study is in
Sony has announced a camera without film. Kodak's market research examines what that means for film.
Size the core business: about 90 percent of the film market in the 1970s
How it is scored
- V 10
- The existence of the company is at stake. Nothing weighs more on the scale.
- I 10
- If film collapses and no second business is ready, the earnings that plants, research and tens of thousands of jobs live on are gone. What remains is a company without a business that carries it. That is the largest difference on the table, which is why it stands at 10.
- P 2
- Kodak's own study considers the replacement of film possible and gives the company roughly ten years. The 1975 prototype had 100 by 100 pixels and took 23 seconds per image. A collapse within ten years is therefore not very plausible.
Size market share in digital cameras and photo printing
How it is scored
- V 6
- Market position is a means: it counts for as much as it later earns.
- I 3
- A device business at consumer-electronics margins. Even a market leader would stay far below the earnings of film.
- P 1.5
- The market does not exist yet, and Kodak's strength is chemistry, not electronics.
Size investments in the billions that earn more in film than in any replacement
How it is scored
- V 8
- Putting capital where it earns the most is the core task of a board.
- I 6
- Film delivers almost all of the earnings. Every dollar that goes early into a new business earns less at first.
- P 9
- Supported by Kodak's own figures: it holds about 90 percent of the film market, and nothing threatens sales.
Size chemical plants, photo labs and tens of thousands of employees
How it is scored
- V 7
- Jobs and accumulated know-how weigh heavily, but less than the existence of the company.
- I 4
- A company that changes course closes chemical plants and labs, cuts tens of thousands of jobs and writes off equipment. That affects almost every site. It is expensive, but it is a one-off burden.
- P 9
- Certain: a company that does not change course does not restructure.
1998: film at its peak
Film cameras sell better than ever before. The first digital cameras for consumers are on the market.
Size the core business: about 90 percent of the film market in the 1970s
How it is scored
- V 10
- The existence of the company is at stake. Nothing weighs more on the scale.
- I 10
- If film collapses and no second business is ready, the earnings that plants, research and tens of thousands of jobs live on are gone. What remains is a company without a business that carries it. That is the largest difference on the table, which is why it stands at 10.
- P 3.5
- Closer than in 1981, but not yet measurable. Film cameras peaked in 1997 at 36.7 million units, Kodak's revenue in 1996. The manufacturers' statistics do not yet list digital cameras.
Size market share in digital cameras and photo printing
How it is scored
- V 6
- Market position is a means: it counts for as much as it later earns.
- I 3
- A device business at consumer-electronics margins. Even a market leader would stay far below the earnings of film.
- P 3
- Kodak builds sensors and cameras for professionals. Whether that becomes a lead in the mass market is open: electronics groups compete there.
Size investments in the billions that earn more in film than in any replacement
How it is scored
- V 8
- Putting capital where it earns the most is the core task of a board.
- I 6
- Film delivers almost all of the earnings. Every dollar that goes early into a new business earns less at first.
- P 8
- Film still carries the company. Kodak plans more than one billion dollars for China because sales are growing in new markets.
Size chemical plants, photo labs and tens of thousands of employees
How it is scored
- V 7
- Jobs and accumulated know-how weigh heavily, but less than the existence of the company.
- I 4
- A company that changes course closes chemical plants and labs, cuts tens of thousands of jobs and writes off equipment. That affects almost every site. It is expensive, but it is a one-off burden.
- P 9
- Still certain.
Early 2002: the curve is visible
The figures for 2001 are in. Kodak writes in its annual report that the film business is strong.
Size the core business: about 90 percent of the film market in the 1970s
How it is scored
- V 10
- The existence of the company is at stake. Nothing weighs more on the scale.
- I 10
- If film collapses and no second business is ready, the earnings that plants, research and tens of thousands of jobs live on are gone. What remains is a company without a business that carries it. That is the largest difference on the table, which is why it stands at 10.
- P 8
- Shipments of digital cameras have nearly tripled in two years: 5.1 million (1999), 10.3 million (2000), 14.8 million (2001). Film cameras fell by 13 percent in 2001, and film sales in the US declined for the first time. At that pace digital overtakes film within one to two years.
Size market share in digital cameras and photo printing
How it is scored
- V 6
- Market position is a means: it counts for as much as it later earns.
- I 3
- A device business at consumer-electronics margins. Even a market leader would stay far below the earnings of film.
- P 4
- The market is there and growing. That an early place in it brings profit is still weakly supported.
Size investments in the billions that earn more in film than in any replacement
How it is scored
- V 8
- Putting capital where it earns the most is the core task of a board.
- I 6
- Film delivers almost all of the earnings. Every dollar that goes early into a new business earns less at first.
- P 4
- The return depends on volume, and volume is starting to fall. Money in film only earns for as long as the digital camera curve does not continue.
Size chemical plants, photo labs and tens of thousands of employees
How it is scored
- V 7
- Jobs and accumulated know-how weigh heavily, but less than the existence of the company.
- I 4
- A company that changes course closes chemical plants and labs, cuts tens of thousands of jobs and writes off equipment. That affects almost every site. It is expensive, but it is a one-off burden.
- P 5
- Staying no longer avoids the restructuring, it postpones it. If film sales keep falling it comes anyway: later and under pressure.
Decision map: the same four arguments at three points in time. Value and size stay the same; plausibility follows the evidence of the year.
Arguments for changing course Arguments for staying
| Argument | 1981 | 1998 | Early 2002 | |
|---|---|---|---|---|
| 1 | Film collapses within ten years, and without a second business the earnings are gone | 20.0 P 2 | 35.0 P 3.5 | 80.0 P 8 |
| 2 | An early entrant has a place in the new market | 2.7 P 1.5 | 5.4 P 3 | 7.2 P 4 |
| 3 | The money stays in the business with the highest return | 43.2 P 9 | 38.4 P 8 | 19.2 P 4 |
| 4 | No restructuring of plants, labs and workforce | 25.2 P 9 | 25.2 P 9 | 14.0 P 5 |
| Change course total | 22.7 | 40.4 | 87.2 | |
| Stay total | 68.4 | 63.6 | 33.2 |
Course of the map: the same arguments at each point in time. Value and size stay the same; only the plausibility has changed. Where the two thick lines cross, the side that is ahead changes.
How to read the map
Each row names exactly one impact: one difference between the two futures. V is the weight of the good at stake (0 to 10), I is the size of the difference (10 is the largest on the table, all others are in proportion to it), P is the plausibility given the evidence on the day of the decision (0 to 10). The numbers are assumptions of this reconstruction, not measurements. Each one carries its reason so that it can be challenged.
The scoring uses what was known on the day of the decision. What became known later is in the last section and enters no number on the map.
A curve that doubles and a forecast that follows behind
Film cameras Digital cameras
Data as a table
| Year | Film cameras (m) | Digital cameras (m) |
|---|---|---|
| 1995 | 29.6 | |
| 1996 | 31.1 | |
| 1997 | 36.7 | |
| 1998 | 36.0 | |
| 1999 | 33.9 | 5.1 |
| 2000 | 31.7 | 10.3 |
| 2001 | 27.6 | 14.8 |
| 2002 | 23.7 | 24.6 |
| 2003 | 16.3 | 43.4 |
| 2004 | 10.1 | 59.8 |
| 2005 | 5.4 | 64.8 |
| 2006 | 1.6 | 79.0 |
| 2007 | 0.8 | 100.4 |
| 2008 | 119.8 | |
| 2009 | 105.9 | |
| 2010 | 121.5 |
Fig. 1: Digital cameras overtake film cameras in 2002. When the figures for 2001 were in, digital cameras had nearly tripled in two years. Data: CIPA, shipments by member companies.
| Year | Kodak's forecast | Result |
|---|---|---|
| 2003 | Start of year: 4 to 5 points less growth. Mid-year: −7 to −8% | about −8% |
| 2004 | −10 to −12% | about −18% |
| 2005 | up to −30% | about −25% |
Fig. 2: For two years the result was at the unfavorable edge of the forecast or below it. Source: Kodak, annual reports 2002 to 2005.
The figure shows the shipments of the camera manufacturers. Digital cameras doubled from 1999 to 2000 and grew by 43, 66 and 77 percent in the three years after that [4]. Film cameras lost almost all of their volume between 2000 and 2007: from 31.7 million to 0.8 million [7].
Kodak's forecasts are in its annual reports. In early 2003 the company expected digital photography to reduce the growth of US film sales by 4 to 5 percentage points. By mid-year it was a decline of 7 to 8 percent [8]. For 2004 Kodak predicted 10 to 12 percent; it turned out to be 18 percent [9][10]. Only the forecast for 2005 was above the result [10][12].
Each of these forecasts extended the previous year and added a little. A quantity that grows by half or more every year cannot be caught that way. Extended at the pace of 1999 to 2001, it was foreseeable in early 2002 that digital would overtake film within one to two years.
What would have changed the decision
The map of 1981 sides with Kodak. Giving up the core business then would have meant forgoing billions in profit [15], for a technology that did not yet exist as a product for sale.
The study delivered what a recorded decision needs: the conditions under which it stops holding. Price, image quality, how the devices work together. On the map these three quantities hang on one row, and every year calls for the same reading: is the plausibility still at 2?
Kodak did watch the change. Willy Shih, a Harvard professor and a former Kodak executive, writes that management constantly tracked how fast digital was replacing film [15]. What was apparently missing was a procedure linking observation and decision: a threshold beyond which the reasoning of 1981 counts as outdated, and a forecast that extends strong growth as strong growth.
The map also shows what changing course does not mean. The row on a place in the new market stays weak in all three states. In 2004 and 2005 Kodak was the US market leader in digital cameras [17], and in 2005 digital revenue was higher than film revenue [12]. That did not carry the company. After 2010 the whole camera market lost 94 percent of its volume to the smartphone [4]. The same collapse hit the competitor Fujifilm. It steered the money from film into businesses that build on its chemistry, such as cosmetics from 2006 [16].
That an earlier decision would have let Kodak dominate the camera market cannot be claimed. What can be said is that it would have entered the transformation in a better position: years earlier, while the film business could still pay for the restructuring, and under less pressure. It would have had a fighting chance.
What happened afterwards
The 2003 plan set a target of 16 billion dollars in revenue for 2006. The result was 13.3 billion [11][18]. In July 2005 chief executive Antonio Perez said that traditional products were declining faster than expected and raised the job cuts to as many as 25,000 [13].
On 19 January 2012 Kodak filed for bankruptcy protection [14]. Of more than 145,000 employees in 1988, about 17,100 were left at the end of 2011 [2][14]. Fujifilm puts the cost of its own restructuring at about 200 billion yen and came through the collapse [16].
A good decision has an expiry date.
A core business shrinks, a new one grows, and the figures are in the company's own reports. Many market leaders saw this and reviewed yesterday's decision too late.
- 2001
Polaroid
In the second quarter of 2001 revenue fell by 31 percent. Polaroid's own quarterly report named growing demand for digital formats as one reason. In October 2001 the company filed for Chapter 11 protection [23][24].
- 2010 to 2013
Nokia
Smartphone revenue fell by 27 percent in 2011 and by half in 2012. In 2013 Nokia sold its devices business to Microsoft and licensed its patents, for 5.44 billion euros in total. The devices business had brought in almost half of revenue in 2012 [19][20].
- 2013 to 2016
BlackBerry
In fiscal 2014 revenue fell by 38.5 percent, and devices sold dropped from 28.1 to 13.7 million. In September 2016 BlackBerry announced that it would stop developing its own devices [21][22].
Sources
- IEEE Spectrum: the first digital camera (Steven Sasson, 1975)
- The Economist (14 January 2012): The last Kodak moment?
- Chunka Mui, Forbes (18 January 2012): How Kodak Failed. With Vince Barabba's account of the 1981 study
- CIPA: shipments of digital still cameras, 1999 to 2024
- Eastman Kodak: annual report (10-K) for 1998
- Eastman Kodak: annual report (10-K) for 2001
- CIPA: shipments of film cameras up to 2007
- Eastman Kodak: annual report (10-K) for 2002 and quarterly report for Q2 2003
- Eastman Kodak: annual report (10-K) for 2003
- Eastman Kodak: annual report (10-K) for 2004
- Eastman Kodak: digital strategy announcement of 25 September 2003 (reproduced by Imaging Resource)
- Eastman Kodak: annual report (10-K) for 2005
- Associated Press (20 July 2005): Kodak raises its job cuts
- Eastman Kodak: annual report (10-K) for 2011
- Willy Shih (2016): The Real Lessons From Kodak's Decline. MIT Sloan Management Review 57 (4)
- Fujifilm Holdings: investor presentation 2015 (demand for color film, restructuring costs)
- Vinokurova, Kapoor (2025): The Problem of Sustaining a Successful Enterprise: Kodak's Multiple Takes at Strategic Renewal That Culminated in Failure. Business History Review
- Eastman Kodak: annual report (10-K) for 2006
- Nokia: annual report (Form 20-F) for 2012
- Microsoft (3 September 2013): Microsoft to acquire Nokia's devices and services business
- BlackBerry: management's discussion and analysis for fiscal 2014 (exhibit to Form 40-F)
- BlackBerry (28 September 2016): Q2 fiscal 2017 results and new strategic direction for devices
- Polaroid: quarterly report (Form 10-Q) for Q2 2001
- Polaroid (12 October 2001): voluntary Chapter 11 petition
Decision under uncertainty Challenger, 1986
A risk without a number is not a risk of zero.
The engineers could not quantify the risk. The map shows they did not have to: it was enough to know on which side of the tipping point it lay.
Read the case →
Transparency of a decision Netflix and Qwikster, 2011
What is not open inside cannot be explained outside.
The customers were not told why. Inside the company nobody voiced the doubts. A decision that is not transparent inside cannot be explained outside.
Read the case →
Deciding by one's own values Volkswagen, 2006 to 2015
Values count only when they are part of the decision.
The objections were on the table, the values were on paper. Under pressure of time and cost, neither counted. As a row in a decision, a value is harder to pass over in silence or to talk down.
Read the case → The method behind the mapsThe full derivation of V × I × P is on the page Reason Based Decision Making for Business.
To the derivation →