TURNS AND RETURNSBy Larry Barrett, Sean Gallagher | Posted 2003-07-02 Email Print
McDonald's planned to spend $1 billion over five years to tie all its operations in to a real-time digital network, but the project failed before it even got off the ground.
A more immediate problem for McDonald's and Cantalupo was dealing with the diminishing profits and sluggish sales that had conspired to erode the stock price. Innovate, he said, didn't address the most pressing needs of the company—improving the speed and quality of restaurant operations. As former CIO Dill says, "McDonald's didn't understand how to use technology to improve its day-to-day operations."
When Cantalupo canceled the Innovate project, he said McDonald's would reduce its capital spending in 2003 by $700 million, to $1.2 billion, which is down from $2 billion in fiscal 2002, ended December 2002. Some of this cash will be used to repurchase McDonald's shares and for increasing shareholder dividends as well as paying off some of the $9.7 billion in long-term debt on its balance sheet.
McDonald's lackluster financial performance in recent years has led three of the 15 analysts tracking the stock to issue "sell" recommendations. Only six of the 15 analysts rate it a "buy" or "strong buy" despite its precipitous plunge.
In fiscal 2002, McDonald's posted a profit of $893 million on sales of $15.4 billion. That sales figure represents only a 4% improvement from fiscal 2001 when it earned $1.64 billion on sales of $14.8 billion.
"We've had a neutral rating on the stock for five years," says Alan Hickok, an analyst at U.S. Bancorp Piper Jaffray. "Its financial performance has been deteriorating for years without arrest. That it finally reported a loss wasn't so much a shock as an inevitability."
An internal memo sent out in December from Vice Chairman Jim Skinner to employees working on the Innovate project states that McDonald's decision to terminate Innovate was based on the company's recent financial difficulties.
"We have been screening our initiatives based on their customer impact and their ability to deliver financial benefits to the system in the near term," Skinner wrote in the memo. "With that in mind, we have made the very difficult decision to immediately stop all work on Innovate."
Immediately canceling Innovate meant that between 100 and 200 contractors and McDonald's employees working on the project would be pink-slipped or reassigned. And the $170 million spent in 2002 specifically for the Innovate project was lost.
McDonald's current CIO Dave Weick says most of that $170 million was spent on the "research and development" of Innovate. BearingPoint and PricewaterhouseCoopers (now part of IBM Global Services) consulted on the project and provided the initial analysis and framing used to determine the project's viability and likely return on investment (ROI).
"As far as the ROI, candidly, I don't want to go into specifics," says Weick of measuring Innovate's possible profit return. "We did have some numbers but I'm not able to discuss them. I can tell you that we viewed Innovate as a five-year plan to get [implemented] in our major markets. Clearly, with that kind of expense, we were sure we would have seen substantial ROI had the project gone forward."
PricewaterhouseCoopers (PWC) would not comment about its work on Innovate. BearingPoint spokesman John Schneidawind says the consulting services firm will not comment on McDonald's nor will it say whether or not McDonald's is even a customer. However, McDonald's CIO Weick did say that BearingPoint played a key role in Innovate's development (see Dossier, p. 54).
Jim Carlini, president of information technology consultancy Carlini & Associates, says BearingPoint, PWC and McDonald's executives likely sat down in the initial stages of Innovate's development to determine what benefits the company hoped to gain from the system.
"I'm sure they wanted to know basic information like sales per restaurant, service times, the number of crew members working a particular shift and all the other data you'd want to improve your operations," he says. "I'm sure they spent millions just determining what platform would work best in all the stores and, possibly, some prototypes and initial software licenses to test it in some capacity."
Carlini, however, is skeptical that $170 million was spent entirely on consulting fees and some early-stage hardware and software. "I'm guessing they buried several other pet technology projects in there as well," he says. "But those consulting fees can add up in a hurry."
The rest was apparently spent on Sun Fire-class enterprise server hardware from Sun Microsystems to host the Oracle 11i software that would form the backbone of Innovate. The fate of that hardware is unknown, though sources at Sun indicate that another project is under way at McDonald's involving Sun hardware. "We're not at a point where we can talk about anything because it's still in the middle of things," says Aaron Cohen, a Sun customer public relations spokesperson.
Oracle officials declined to comment on Innovate or McDonald's.
As one former McDonald's information systems manager put it, "No one wants to talk about this project. If you're doing business with McDonald's, you're not going to talk. And if you're not, you're not going to talk because eventually you're going to want to do business with them."
McAfee, the assistant professor at Harvard Business School, says McDonald's likely fell victim to classic pitfalls that befall corporations trying to implement and justify information systems projects of this size for the first time.
"Companies attempting implementations of this size have very different ways of determining their return on investment," he says. "I've seen companies take the sharp-pencil approach and break everything down to the penny. Others aren't looking for any kind of immediate return because they're simply making a strategic investment and will worry about the ROI down the road."
Had McDonald's made the $1 billion investment in Innovate to streamline its supply chain and improve its day-to-day operations, it would have needed to achieve at least a 1.5% improvement in sales, or roughly $231 million a year, to pay for the initial rollout. And that's an additional 1.5% above the 3% to 5% in annual sales McDonald's was already projecting.
McDonald's Weick says the project would have reached the break-even point as soon as eight of the McDonald's top-performing locations were live. However, he was unable to elaborate on the exact dollars McDonald's hoped to save.
McDonald's certainly made efforts to manage the cost and risks up-front for Innovate. After an initial exploration of the ideas behind the project with PWC Consulting, McDonald's brought in a sourcing consulting firm, the MPower Group (based a little more than a mile from the sprawling McDonald's corporate campus in Oak Brook) to help create processes for selecting technology vendors and integrators for the project (see Gotcha! above). According to white papers published by the MPower Group about its groundwork for Innovate, McDonald's saved 40% against existing professional services rates with its integrators, and negotiated technology vendors down 60% from their initial quotes as a result of the sourcing process they put in place.
That McDonald's first foray into large-scale, real-time data systems failed comes as no shock to some experts.
"It doesn't surprise me a bit that a company like McDonald's, with its relative lack of experience in this area, spent so much money and has so little to show for it," McAfee says. "It happens all the time. The fact that McDonald's isn't exactly on the cutting edge in terms of technology or the executive-level appreciation and understanding of technology surely made matters worse."