"All you have to do is rattle off a few technical terms, and the customers look at you like deer caught in the headlights," one consultant says. "If you know more about computers than they do, they think you know everything about computers. Nobody knows everything about computers." There are all kinds of subspecialties these days, and a consultant who might be outstanding in one area can be hopelessly lost in another. It's like asking a knee specialist to do a heart transplant. What customers need is not computer expertise, just skepticism. "You don't have to know how an engine works to buy a car," the consultant says, "but you need the gumption to control the buying process. You can't be intimidated."
So powerful is the guru effect, however, that even sophisticated companies with strong negotiating skills can get burned. Not long ago the partners of a Los Angeles law firm met a consultant who promised to bring them "up to speed" on computers. "He was trying to convince us that all the lawyers needed PCs," recalls a former associate at the firm. "He said we could be the first guys on the block to distribute laptops to our clients, so they could be networked into our system, with passwords to their files, and we could be really high tech and paperless."
The "guru" put a $200,000 price tag on the project. Some of the senior managers found his ideas so innovative that they decided to make the firm a proving ground for a complete package, which they and the consultant could jointly market to other law firms.
As the deal unfolded the lawyers spent numerous hours studying the proposal, designing a plan to fit the strategy, and deciding how to finance the job. It looked so promising that the firm invested tens of thousands of dollars in the consultant's small company. In the end, for unknown reasons, he never delivered the goods. And only later did the lawyers learn that everything in the package was available--off-the-shelf--for about $100,000 and could have been installed with minimal help. "This happens all the time," says the former associate.
For any given computer problem, there exists a solution. This stuff isn't magic, but it is complex and demanding. It sometimes happens that outside contractors, hungry for income, overstate their expertise and fast find themselves out of their depth, floundering and creating an aftermath of misery. We're not talking outright fraud, just false confidence by the consultant that he or she can manage the task. Witness the crisis that beset Motherwear, a catalog business, after a sales-driven software developer installed a calamitous program.
A $4.8-million company based in Northampton, Mass., Motherwear had outgrown its mail-order software, which handled everything from merchandising and credit-card sales to invoicing, shipping, and inventory control. After searching for a beefed-up replacement, CEO Prakash Laufer contracted with a small midwestern company we'll call Avion Systems (not its real name).
"Their people seemed friendly and knowledgeable," Laufer recalls, "and when they did a presentation at a catalog conference, I thought they were on the cutting edge. They came in to run a demo for us with their president and chief programmer and said their system could do everything we needed. I was impressed." In December 1995 he signed Avion aboard for a $35,000 project, hoping to have the system running by the beginning of March for his spring season. In early February, Avion sent a few programmers to Motherwear's location for the installation.
They left after eight days. The troubles began soon after. "We had major problems in every area of the company," Laufer says. The new "system" left Motherwear barely able to process credit-card orders--its lifeblood--and returns. It couldn't track customers; in some cases they were billed five times for the same thing or never billed at all. For several months the company could not ship packages by air. It had no accurate reports on what stock was in inventory, what had been shipped, or what items needed replenishing.
"Our people were literally crying and freaking out," Laufer recalls. "Customers were yelling at our call-center operators because their returns were a month late, and the call center couldn't give any answers, because it couldn't access their files. People had to run around to manually enter orders, which took us two or three times longer than before. That meant callers were on hold for a long time, and they'd hang up and not call back."
Ironically, the Avion software had substantial reporting features, so Motherwear could calculate the damage from canceled orders--about $300,000, according to financial manager Holly Smith -Bove. "But you don't know how many of the people you angered will never order again," she says. Payroll costs, moreover, shot up by $100,000. "It was taking us so much longer to process orders that we had to bring in more people," Laufer says. "And we also had heavy turnover. Employees got upset because customers were taking out their frustration on them."
Two months after the initial installation, and only after Motherwear brought in a lawyer, Avion finally sent in a second team, but most of the problems remained until Laufer hired a competent programmer to untangle the mess. That work cost him $50,000 last year and $40,000 this year. "We're finally getting the system we want, but it's been very expensive," he says. "What we learned was 'Buyer beware.' These guys were convincing salesmen, but obviously, they couldn't back it up."
Don't get the wrong impression--there are plenty of good, honest consultants and programmers out there. The key to successful office automation is identifying the right one for the job and then giving the project the same focus and attention you'd give any core business operation. Consultants emphasize that the key is defining the problem to be solved and planning out the system in painstaking detail.
"There's a rule that professionals go by," says Gordon MacDonald, "which is that for every hour you spend planning--meaning not touching a computer--you save five hours in implementation. Before anybody gets to write software, you should have mock-ups of all the screens the program is going to produce, with the menus and options. You should have a flow chart explaining how the program is going to proceed, in plain English, so the nontechnical people can see how it will react to specific situations. And there should be a clear statement as to what the program will and won't do. The kiss of death is when a programmer spends an hour with an end user and then sits down to write."
