Free Who Says Elephants Can't Dance?: Leading a Great Enterprise through Dramatic Change by Jr. Louis V. Gerstner Page A
Authors: Jr. Louis V. Gerstner
This was the model created by IBM, and over time, only a handful of computer competitors, all fully integrated, emerged in the United States (often described as the BUNCH—Burroughs, Univac, NCR, Control Data, and Honeywell). The same model emerged in Japan and, to a lesser extent, in Europe.
In the mid-1980s a new model started to appear. It argued that vertical integration was no longer the way to go. The new breed of successful information technology companies would provide a narrow, horizontal slice of the total package. So companies that sold only databases began to emerge, as well as companies that sold only operating systems, that sold only storage devices, and so on. Suddenly the industry went from a handful of competitors to thousands and then tens of thousands, many of which sold a single, tiny piece of a computer solution.
It was in this new environment that IBM faltered, and so it was logical for many of the visionaries and pundits, both inside and outside the company, to argue that the solution lay in splitting IBM
into individual segments. This conclusion, however, appeared to me to be a knee-jerk reaction to what new competitors were doing without understanding what created fragmentation in the industry.
Two things really drove the customer to support this new, fragmented supplier environment:
• Customers wanted to break IBM’s grip on the economics of the industry—to rip apart IBM’s pricing umbrella, which allowed it to bundle prices and achieve significantly high margins.
• The customer was increasingly interested in delivering computing power to individual employees (the term was “distributed computing,” in contrast to the mainframe’s “centralized computing”).
WHO SAYS ELEPHANTS CAN’T DANCE? / 59
IBM was slow, very slow, in delivering distributed computing, and many small companies moved in to fill the gap. These companies were in no position to deliver an entire integrated solution, so they offered add-ons to the basic IBM system and built around IBM’s central processing hub. This is clearly what Microsoft and Intel did when IBM reluctantly moved into the PC business.
So it really wasn’t that the customer desired a whole bunch of fragmented suppliers. The broad objective was to bring more competition into the marketplace and to seek suppliers for a new model of computing.
And it worked. By the early 1990s there were tens of thousands of companies in the computer industry, many of which lived for a few months or years, then disappeared. But the impact of all this dynamism was lower prices and more choices (with the notable exception of the PC industry, where Microsoft re-created the IBM choke hold, only this time around it was in the desktop operating environment rather than in the mainframe computer).
While there were good consequences to this fundamental reshaping of the computer industry, there was also one very undesirable outcome. The customer now had to be the integrator of the technology into a usable solution to meet his or her business requirements.
Before, there was a general contractor called IBM or Burroughs or Honeywell. Now, in the new industry structure, the burden was on the customer to make everything work together.
This was all complicated by the absence of uniform standards throughout the computer industry. Competitors in the computer industry, unlike any other industry I know, try to create unique standards that they “own.” They do not make it easy for other companies in the industry to connect with or understand these standards without extracting a huge price. Thus, there is a cacophony of standards and interconnect requirements that makes the creation of a single solution very difficult. (I’ll discuss this in more detail later.)
60 / LOUIS V. GERSTNER, JR.
As a big customer of the information technology industry in the early 1990s, I knew firsthand that integration was becoming a gigantic problem. At American Express, our wallet-sized piece
In the mid-1980s a new model started to appear. It argued that vertical integration was no longer the way to go. The new breed of successful information technology companies would provide a narrow, horizontal slice of the total package. So companies that sold only databases began to emerge, as well as companies that sold only operating systems, that sold only storage devices, and so on. Suddenly the industry went from a handful of competitors to thousands and then tens of thousands, many of which sold a single, tiny piece of a computer solution.
It was in this new environment that IBM faltered, and so it was logical for many of the visionaries and pundits, both inside and outside the company, to argue that the solution lay in splitting IBM
into individual segments. This conclusion, however, appeared to me to be a knee-jerk reaction to what new competitors were doing without understanding what created fragmentation in the industry.
Two things really drove the customer to support this new, fragmented supplier environment:
• Customers wanted to break IBM’s grip on the economics of the industry—to rip apart IBM’s pricing umbrella, which allowed it to bundle prices and achieve significantly high margins.
• The customer was increasingly interested in delivering computing power to individual employees (the term was “distributed computing,” in contrast to the mainframe’s “centralized computing”).
WHO SAYS ELEPHANTS CAN’T DANCE? / 59
IBM was slow, very slow, in delivering distributed computing, and many small companies moved in to fill the gap. These companies were in no position to deliver an entire integrated solution, so they offered add-ons to the basic IBM system and built around IBM’s central processing hub. This is clearly what Microsoft and Intel did when IBM reluctantly moved into the PC business.
So it really wasn’t that the customer desired a whole bunch of fragmented suppliers. The broad objective was to bring more competition into the marketplace and to seek suppliers for a new model of computing.
And it worked. By the early 1990s there were tens of thousands of companies in the computer industry, many of which lived for a few months or years, then disappeared. But the impact of all this dynamism was lower prices and more choices (with the notable exception of the PC industry, where Microsoft re-created the IBM choke hold, only this time around it was in the desktop operating environment rather than in the mainframe computer).
While there were good consequences to this fundamental reshaping of the computer industry, there was also one very undesirable outcome. The customer now had to be the integrator of the technology into a usable solution to meet his or her business requirements.
Before, there was a general contractor called IBM or Burroughs or Honeywell. Now, in the new industry structure, the burden was on the customer to make everything work together.
This was all complicated by the absence of uniform standards throughout the computer industry. Competitors in the computer industry, unlike any other industry I know, try to create unique standards that they “own.” They do not make it easy for other companies in the industry to connect with or understand these standards without extracting a huge price. Thus, there is a cacophony of standards and interconnect requirements that makes the creation of a single solution very difficult. (I’ll discuss this in more detail later.)
60 / LOUIS V. GERSTNER, JR.
As a big customer of the information technology industry in the early 1990s, I knew firsthand that integration was becoming a gigantic problem. At American Express, our wallet-sized piece
Similar Books
After
Marita Golden
The Star King
Susan Grant
ISOF
Pete Townsend
Rockalicious
Alexandra V
Tropic of Capricorn
Henry Miller
The Whiskey Tide
M. Ruth Myers
Things We Never Say
Sheila O'Flanagan
Just One Spark
Jenna Bayley-Burke
Island of Silence (Unwanteds)
Lisa McMann
The Venice Code
J Robert Kennedy