Bruce came up with our theme for 1991, which was our year to “think.” Looking back, I have trouble understanding why we took our themes so seriously, but we did. The theme “think” came from a few experiences Bruce had working on the family farm in Idaho. More than once when Bruce had made a mistake, his father had gotten very frustrated and repeated the words “Think, think, think” with great intensity.
I liked the new theme, because it uncovered another one of our marketing problems. By now we were fairly good at making presentations and fairly good at giving customers a little time to talk, but many of our salespeople were not processing the information they were getting from our customers. To some, listening was only a pause in the presentation to give customers a chance to feel important. Our salespeople needed to adapt what they were teaching to what our customers wanted to learn.
The theme was appropriate given all the thinking we had to do to try to solve our problems. Our biggest one was the continued delay in the shipment of WordPerfect for Windows. Just one week after Fall COMDEX in 1990, the Windows programmers informed us that the dates we had given out at COMDEX–a beta test to start in December and a product release at the end of February–would be impossible to meet. The most optimistic in the group felt the end of the second quarter of 1991 was the earliest possible release date. The pessimistic ones were saying August or September. We were in deep trouble.
The situation reminded me of the delays Lotus had gone through with a recent update of 1-2-3. First they announced a six month delay. When the six months were up, they announced another six month delay. I was worried we might suffer the same embarrassment. The right thing to do was to tell our customers immediately that we knew our release date was in trouble, but I wanted to wait. We still did not know for sure when the product would be ready, so I convinced myself we could hold off on announcing the delay until we had better information.
Finally, in early February we broke the news that our release date had slipped, but I was not entirely honest in making the admission. Rather than go with a realistic date or a vague date or no date at all, I announced a hoped-for second quarter release, which was the most optimistic date from our most optimistic developer. My only excuse for announcing the earliest possible yet highly improbable date was that I was going through some sort of temporary insanity caused by a bad case of denial. I was afraid to face the difficulties that a longer delay might cause.
Few of the industry analysts or members of the press believed we could make the second quarter date, and word of our troubles was leaking out of the company. Amy Wohl, one of the more influential consultants, was quoted as saying we would make the second quarter release date if there happened to be 75 days in June. She underestimated the time we would need by about 60 days.
We on the Board had no one to blame for the delays but ourselves. The project directors we had chosen were inexperienced managers, and they made the mistakes inexperienced managers make. They were prone to overly optimistic forecasts and had trouble chewing people out when they missed their deadlines. Another of our mistakes was that we waited too long to add new programmers to the project, always thinking we were so close to a release that we did not have time to train additional programmers and get any significant help from them. We also took too long to make our experienced DOS programmers get involved. They could have helped a little more, but we had a hard time convincing them that the Windows project was more important than anything else. With sales still going up, many thought things were going too well to be concerned.
It was not just the WordPerfect for Windows project that was behind schedule. As Alan and I met with each of the development groups to check on their progress, their stories were all the same. They all were behind schedule, and they all wanted more programmers. The DrawPerfect programmers, for example, were nowhere close to a new release. Though their original product was well received in 1990, their program was already showing its age and starting to get less favorable reviews. Their next version, renamed WordPerfect Presentations, would not be ready for two more years.
The FormsPerfect group, which was working on a project that would later be renamed InForms, was also making little progress. Their program was intended to help companies replace their printed forms with electronic forms. For example, if a company had a form for requesting reimbursement for medical expenses, InForms would allow employees to fill out the form on their computer screens and then route it to the proper department automatically. When Alan and I were meeting with the Forms group, Bruce happened to call from Europe. When we told him what we were doing, he asked us to make sure the forms the group was designing would be compatible with WordPerfect. When we relayed his concern to the group, they all looked sick. Somewhere along the way someone had decided that WP compatibility was not a high priority, and the product, which would not come out until mid-1993, would not create forms which could be used with WordPerfect.
WordPerfect Office was now six years into an eleven year project, and a release that would deliver on all our promises was still a long way off. One big problem was getting all the different Office development groups to work together. By now we had teams for PC networks, for the Macintosh, and for UNIX, DG, and DEC machines. Unfortunately, none of the groups seemed to be willing to work out their differences. For instance, the groups could not decide on the maximum number of enclosures which could be attached to a message. The VAX group wanted an unlimited number of attachments, the Mac programmers felt 100 files was more than enough, and the PC group allowed only 30 files to accompany a message. Of course, if someone from a VAX machine were to send 101 attachments with a message, the Mac and PC Office versions would crash, but that was not incentive enough to get the groups to agree on a standard. The VAX group felt they had to have an unlimited number, because their VAX competitors offered an unlimited number. The Mac programmers were certain 30 was too few, but the PC group already had a product shipping with the 30 file limitation, and they were not about to change.
Unfortunately, Alan did not have a lot of time to work with the different development groups. He was spending a good deal of his time giving speeches, collecting awards, and attending industry conferences. Even if he had been around the office, we had so many projects going that he could not have given all of them much attention. In his absence, his development leaders had to do a lot more on their own and sometimes got off track. To try to help get things back under control, Freida, his assistant, and I conspired to limit the awards that Alan could accept to two per month. I wondered if all the attention we were receiving had anything to do with Forbes Magazine having identified Alan and Bruce as among the richest 400 individuals in the United States.
Although my thoughts did not carry a lot of weight with the programmers, I spent a lot of time trying to figure out what needed to be done to get things under control. For years we had tried to produce a successful family of products, and the result was more software projects than we could handle. In trying to figure out which products to continue and which products to cut, I realized that our product successes seemed to be those that were most closely related to word processing. WordPerfect Office had to do with distributing messages or documents, and DrawPerfect had a lot to do with creating figures and charts for documents. Perhaps as a result of the document connection, both products sold fairly well. DataPerfect and PlanPerfect were applications which had little to do with documents, and neither had sold very well. I came to the conclusion that the focus of our development should be on word processing and those products which were closely associated with word processing, and from then on I never hesitated to promote my new theory.
At the end of February, IBM asked me to come to Boca Raton for a secret meeting. This was yet another one of those times when someone was pulling our chain. The appointment was for 8:00 a.m., which was 6:00 a.m. Utah time, so I was in a bad mood when the meetings began. It was almost more than I could bear when I found out that I was there for two days of intensive OS/2 training. There were five or six of us present representing all the largest software companies except Microsoft, and we were to be IBM’s audience for a dress rehearsal of an OS/2 seminar they were about to give. I listened respectfully for as long as I could, which was about thirty minutes, and then raised my hand. The presenter asked me to hold my question until the end of the presentation, but I told him I would leave if I could not interrupt. He reluctantly agreed to let me speak, and I announced that I would rather be at an Amway recruiting dinner than listen to two days of hour long technical presentations. I wanted to know why we were wasting our time talking about OS/2 when Microsoft had clearly won the war of the GUI operating systems.
What I really wanted to know was why IBM was trying to copy Microsoft with a technical approach to marketing. I could remember back when IBM introduced their new Personal Computer AT model with a huge party in Dallas called “Partners in Pride.” They had invited almost everyone in the industry to a beautiful hotel to give us great food and entertainment and to introduce their new product with a few short presentations. That was the IBM I understood. Now, however, they were playing the part of the nerd, trying to be like Microsoft by offering boring classes on graphics capabilities and memory management. This was not the IBM I used to know.
IBM tried to keep me happy by cutting down on the amount of technical information included in the meetings, but most of what they presented went over my head. The one thing I did understand was that IBM was not ready to give up on OS/2. If they had to turn into nerds to win the operating system business, they were willing to give it a try. They were willing to do almost anything to win back control of the market, including giving their operating system away for a time and paying Microsoft to include a copy of Windows with OS/2, so computer users could run DOS, Windows, and OS/2 programs by buying only OS/2.
As I sat through the meetings, I wanted to believe that IBM had a chance. Our long term success was, I thought, dependent on diversity. If the world was filled only with Windows machines, then Microsoft would have a tremendous advantage. If instead the world was filled with DOS, Windows, OS/2, Macintosh, and UNIX machines, we could maintain our advantage in the personal computer word processing market.
My advice to IBM was to not play the role of the nerd. I hoped they would hang on to their old polished and professional image, and not walk around with uncombed hair and calculators on their belts. I felt their seminar would be much better if their presentations were more exciting and they took the participants down to the beach to eat a great dinner and listen to a Beach Boys concert as they had in the old days.
Microsoft was very good at making it tough for IBM to push OS/2. They were promising to deliver tools to port Windows applications to OS/2 and encouraging developers interested in OS/2 to write Windows versions first. Later, long after most companies had decided to go with a “Windows first” strategy, Microsoft decided not to deliver the OS/2 porting tools. Worse still, once IBM told the world that OS/2 would run DOS programs better than DOS, and Windows programs better than Windows, Microsoft reminded the world that IBM’s right to distribute the Windows portion of OS/2 would expire in a few years, leading some people to conclude that OS/2 would eventually be a dead end. To this day, I have little hope that IBM will be able to win out over Microsoft, but I admire their tenacity.
By May we still were not showing much progress with a WPwin release. We were a little closer to getting into a beta test, but nowhere close to releasing a product. I dreaded making the announcement of another delay. While I was thinking about how to break the news, I came up with an idea for an ad campaign. I was tired of having everyone focus on our release date, so I proposed to our ad agency that we have a contest for our customers to guess how many copies of WPwin we would ship in the first month the product was released. I hoped to shift the focus from our weakness, which was our tardiness, to our strength, which was the great demand that existed for the product.
The first ad in my proposed campaign would announce the contest, giving the customers details about expected demand for the product and our capacity to manufacture it. We would explain that the purpose of the contest was to get as many people as possible to help us come up with an accurate sales forecast. The second ad would announce various statistics about the guesses we received. We would publish the highest estimate, the lowest, and the average and mean for all estimates. We would also let people know how the product was progressing. A third ad would announce our preparations for the shipment of the product, a fourth the product release, and a fifth the winner of the contest. Hopefully, in the sixth and last ad of the campaign, we could announce to the world that we had the best-selling word processor for Windows.
At first our ad agency had trouble with the idea. Of course, since I was the executive vice president, they told me they liked the idea, but when they showed us a possible layout for the first ad, they had changed the contest into a sweepstakes. The customers only had to send in an entry (not a sales estimate), with the lucky winners chosen from a random drawing. It took a lot of persuading and a little emotion to get them to understand that it had to be a contest. I wanted customers to think about the huge number of packages we were going to ship, and I was frankly very curious as to what they had to say. In the end I got my way, and we went ahead with the campaign.
Getting the agency to go with my idea reminded me that while ad firms can be a great help, they sometimes overestimate their competence and underestimate the taste of some of their clients. When asking them for ideas, I sometimes felt like they were using a trick I learned in my drapery days. When parents decided to let children choose their own drapes, I found the children almost always chose colors which were too bright. To prevent them from choosing something I knew their mothers would hate, I would give them only three choices, making sure that two of the choices were so bad that they would take the one I wanted them to take. When an ad agency gave a presentation, I felt like they were doing the same thing to me. I would see only one option that was reasonable and a few others that were always obviously horrible. More than once I wished they would have treated me like a thinking adult, rather than an incompetent executive who was not to be taken seriously.
Our first ad in the WPwin series had a picture of Clive Winn asking customers to call an 800 number to help him forecast the first month’s sales of WPwin. He offered $25,000 cash to the person with the estimate closest to actual sales.
We held auditions for operators from our Information Services group and selected those who were friendliest to take the calls which came in from the ad. We trained them not only to ask for names, addresses, phone numbers, and estimates, but also to take time to express appreciation to the callers for their help and to engage them in a conversation about their plans for Windows and our Windows product.
From the thousands of people who called, we selected a few and sent a photographer out to take pictures of them, which we included in the second ad. In addition to the high, low, and average of all estimates, we included explanations of some of the different ways people had arrived at their forecasts. Some of the calculations were very complicated.
One of the estimators was a Microsoft employee. He happened to guess that our sales would be many times greater than the current sales for Microsoft Word for Windows, so we put his estimate in the second ad along with a map of Washington State. Microsoft called in a huff to complain about the low blow, which made the whole thing a lot more fun. Instead of wringing our hands and spending the summer worrying about how late we were, we spent it thinking about how many copies of the product we would sell and how much Microsoft hated all the hype.
It was very gratifying for me to read Stewart Alsop’s comments about the ad campaign. Mr. Alsop published a well-read industry newsletter and later would become Editor-in-Chief for InfoWorld. In describing the first ad in the campaign he wrote, “This is one of those incredibly creative marketing ideas that can only be done once…By asking customers to predict the success of the product, WordPerfect manages to get the customers involved in trying to make the thing successful.” In another article written for a trade publication in England, he called the first ad “one of the most creative advertisements ever seen in the PC industry.” That was a nice reward considering the struggle it had been to get our agency to support the campaign.
In July I had a very strange experience. My dad, who by now had retired from Harris Corporation and had come to work in our sales department in Utah, called to ask me for an appointment. This was unusual, considering his house was right next door to mine and we saw each other almost every day. When we met he looked very sad and explained to me that Bruce had asked him meet with me. Bruce wanted him to tell me that I would have to change, because I was too hard on people and too many people were afraid of me.
I had trouble understanding why my dad had been asked to convey the message, so I called Bruce and requested a meeting. By the time I reached his house, Alan was there as well. They explained their concerns, and I told them I did not know how to do my job any other way. I asked them if I was any more difficult or any harder to work with than I had been during the last five or ten years. I had my share of enemies and some people were afraid of me, but this was nothing new. I asked them to play me or trade me and told them there was no way I could do my job without upsetting a few people.
I left Bruce’s house thinking things were settled. I was not very impressed that they had put my father in the middle of the discussion, but I knew they were all under a lot of stress because of the Windows delays. Bruce was getting a lot of pressure from his international offices to release a Windows product in a hurry, and Microsoft was beginning to cut into our market shares in Europe.
Alan was in a difficult situation as well. Many of the development project dates were slipping, and, like me, he was in the position of having to disappoint a few people. He was very determined to get a WordPerfect 6.0 project started right away. The programmers doing the 5.1 Windows version also wanted responsibility for WPwin 6.0, but they could not think of working on a new project for another few months. Alan felt he had no choice but to disappoint the 5.1 Windows programmers and give the 6.0 Windows project to another group.
There was a lot more to what Bruce was trying to tell me than I knew at the time. The vice president of development in charge of WordPerfect Office was convinced I needed to change, and he had been lobbying with Alan, Bruce, and others to have another person elected to the Board of Directors in order to dilute my influence. He was sure that Bruce and Alan went along with my views too much of the time, and that I was leading the company to ruin.
Even without the lobbying, I was not too happy with this vice president. Office was not moving forward very quickly, and instead of resolving the problems, this well-intentioned leader was giving lectures from a book by Stephen Covey about seven habits of effective leaders. He was requiring his teams to take time away from their programming efforts to attend lectures and do homework concerning the book. This vice president was also teaching from a book which advocated using conflict as a way to get things done. I wondered how conflict could be the answer when what the Office group needed was some conflict resolution. Unless a way could be found to get those groups back on track, we were soon to be seven years into a twelve year project.
When I heard about the lectures, I asked Alan what was going on. I wondered why this vice president was allowed to teach whatever he wanted. As I had explained in my management training, I expected the Board to define the correct principles, and not any and every vice president in the company. After listening to me, Alan told me not to worry. He had given the go ahead for the vice president to teach out of Covey’s book, and he would take care of the problem. The conversation was very disappointing to me. I realized then that Alan was not very involved in what I had been trying to teach. He just wanted everyone to be happy.
I did learn a lot from the incident, however. For years we had used “we teach correct principles and let employees govern themselves” as the basis for our management philosophy. Although we had not always done a good job of implementing the philosophy, I felt we had at least understood it. After the incident, I realized we were not even close.
If we were to actually adopt the philosophy, I realized that we would have to complete at least four steps. First we had to define the correct principles for WordPerfect Corporation. Then we had to teach them. Next we had to trust the employees to govern themselves. Finally the employees had to be accountable to the correct principles in governing their actions.
If any one of the four parts was missing, then the management system was not going to work. If the principles were not defined, then any teaching was likely to be in error. If the principles were defined but not taught, the employees would be in the dark. If the employees were supervised so completely that they had no freedom, then there was no opportunity for the learning and individual initiative that come with self-government. If the employees refused to be accountable to the principles when governing their actions, then the first three steps were useless. Although we had done a fairly good job of turning people loose over the years, or fulfilling step three, we always had trouble with steps one, two, and four. Even all the work I had done with my management training courses was not very effective. Without the full support of Alan and Bruce, I could not be certain that my principles were the correct ones for the company.
The fundamental problem I had with Bruce and Alan and others in the company who were defining their own principles was the difference in our definitions of the term “correct.” I thought correct meant “as defined by the Board,” but many others in the company thought it meant “principles which are recognized as true or good.” That a certain principle might be declared true or valid by a certain writer, a certain authority, or a certain business school was totally irrelevant to my way of thinking. What should have mattered most was how the owners of the company wanted to conduct their business. Only the owners had the authority to define right and wrong within the company.
A day or so later I had a chance to explain why I thought it was not possible for anyone in the company to define the correct principles to Alan and the vice president in question, but the vice president was not convinced I was right. He felt he should be given enough responsibility and authority to define and adapt principles for his own group. He felt the title of vice president carried with it certain rights and privileges, because that was the way the job was defined in the normal business world. Of course, I felt that the duties of a vice president should be defined by the employer, not by the rest of the world. There were no published rules of business which required the owners to meet a particular set of expectations.
In spite of Alan’s feeling that our disagreement was not a big problem, people started visiting me to tell me about a petition that was being circulated. Claims were being made that I was ruining careers without good reason and manipulating the Board to my own advantage. I was more hurt than angry with the discovery. This vice president and a few people around him were assuming that if someone’s advancement was impeded by my actions, it was because of my personal feelings for that person. The fact that a particular employee may have messed up badly or that there may have been a good reason for what happened had never occurred to them. My guilt was presumed without a full view of the evidence.
Later in a design review meeting with this vice president and a few programmers from the Macintosh Office group, Alan finally realized the problem was more than a simple disagreement. After the programmers explained what they were doing and what their priorities were, I asked why their 4.0 version had a higher priority than their 3.1 version. Since 3.1 was still unreleased, I was curious to understand why a future version would get the higher priority. The vice president, who had been fairly quiet to this point, lost his temper, raised his voice, and answered so angrily that he bared his teeth and spit as he spoke. The intensity of his emotion finally convinced Alan that this was not a problem which would go away quickly. My question did not deserve such an emotional response. Soon Alan would put someone else in charge of WordPerfect Office and move the vice president to another position in the company.
I did not enjoy my job too much after this time. For close to ten years I was the person most responsible for running the company. It came as somewhat of a shock to learn as a result of my strange experience with Bruce and my problems with this vice president that a lot of what I did was not wanted or appreciated. Although Bruce and Alan were quick to admit that many of the things I did were brilliant, I could now see that we had a lot of disagreements on how the company should be run. I had reached that point where they were beginning to wonder if I wasn’t more of a bother than I was worth.
It was in the fall that we started to talk seriously about going public. I had gone to Bruce and Alan asking for help with a potential estate tax problem, and this led to a discussion of their estate tax problems. Their situation was much more complicated than mine, because their stock was worth so much more money. If both parents in one of the families were to die, then the only way to pay the estate taxes without risking the financial security of their children would be to sell off enough stock to pay the taxes within six months of the deaths. Selling what could be half a billion dollars worth of stock within a six month period was not something they could count on, however. As we talked about the problem, we came to the conclusion that the best answer was to go public or to put ourselves in a position where we could go public at any time. Duff Thompson, our general counsel, and I were asked to start looking into the possibility of a public stock offering.
The WPwin roll-out came off without a major problem. As soon as the product shipped in November, our ads showed trucks lining up to pick up the software and software packages arriving at dealer shelves. As we had hoped, sales were terrific, and for the month of November we outsold Word for Windows. Thirty days after the release, Price Waterhouse sifted through all our invoices and all the estimates and came up with the winner of our contest. As luck would have it, the $25,000 was won by a single mom from the Midwest with a newly adopted baby. We could not have asked for a better ending to our ad campaign script.
Our success was not cause for a huge celebration, however. Microsoft had introduced a new version of WindWord at COMDEX (their official abbreviation for Word for Windows was WinWord, but I liked to add the extra d), and the new version included some new features we did not have. Among other things, they included a grammar checker and feature called Word Art. I liked to think the features were not very useful, but they did look very nice in a demonstration.
Soon after COMDEX, Bruce sent Alan and me a memo. He asked if the three of us were still the ones to run the company. He thought it might be time to delegate more responsibility and rely on the opinions of some other people in the company, rather than only on our own feelings and judgment. He wrote that he was a little embarrassed by our image, which seemed to him to be unprofessional at times. He also felt it was time to go public.
I thought the memo was a little odd and assumed he was venting his frustration over Microsoft’s new features. I told him I thought we knew what we were doing and that I was still able and willing to do my job. I also asked him if he still wanted to do his part. I heard later that he interpreted my question as an insult and decided then that one of us had to go. At the time I was not thinking that he was incompetent. I was asking the question, because I wondered if he was trying to tell us that we was ready to slow down a little. It never occurred to me that he was trying to tell me he wanted me to take a less active role in running the company.
I did agree with Bruce’s desire to go public, but it was not because it might improve our image or solve the estate tax issues. For years our international offices had operated with a lot of independence. I was hoping the initial public offering (IPO) preparations would force us to reign in the international branches. At the very least, I hoped we would get better reports back from the offices so we could get our tax returns finished earlier than the last minute before the end of the last filing extension.
On the surface, 1991 looked like a great year with sales of $533 million and our normally high profits. We were so sure that our record sales levels would continue that we gave out large bonuses to all of our 2,894 employees. From my point of view, however, the year was not a great one. I was worn out, not only because of our ongoing conflict with Microsoft, but from all the infighting and the politics. At a time when I finally thought I knew how to run the company, I realized that not too many people cared.
I did have one satisfying moment at the end of the year. Duff Thompson reminded me that when we recruited him in 1987, I had told him we would have sales of half a billion dollars within five years. Although he had trouble believing me at the time, he noted in December that we had made it with a year to spare.