Compared to the previous year, 1989 looked easy. We were back in rhythm, planning on another release of WordPerfect for DOS in the fall and expecting sales to rise at least another hundred million dollars or so. Our share of the word processing market was up to 40% and rising. We were in a good position to pull further ahead in the DOS word processing market, and if IBM and Microsoft kept fighting long enough over OS/2, perhaps we would have time to get ready to win in the GUI market as well.
For the first time, we decided to have a theme for the year, which we introduced at our Christmas party. Having a theme may sound a little old fashioned or silly, but we took the idea very seriously. It was a major part of our strategy to increase sales and finish off the competition. After spending more than a year concentrating on our war with Microsoft, we were ready to center our attention on something else.
Our theme for the year was “teach,” and we wanted to teach everyone, including our dealers, our customers, the press, and industry analysts, more about using our products. If people came by the office for a visit, we were determined to get them in front of a computer and teach them something new about WordPerfect before letting them leave. If we visited customers at their offices, we would try our best to show them something new about our products. If dealers asked for our help on sales calls, we were going to make sure they learned enough to help with the demonstrations. We also wanted all our employees, especially our marketing personnel, to become more proficient with our software.
In January we received a nice surprise from InfoWorld. They did a report on the top fourteen word processors and rated the products side by side. Although the reviewer had to call customer support more than once to get through, he eventually got a good answer to his question. Our technical support score improved from a Poor to a Very Good, and our overall score went up to 8.3, the highest of all the products in the review. Microsoft Word 4.0 was given only a 7.0. In the area of printer support, an area where Microsoft at one time could claim a big advantage, we were now rated ahead of Word. Our closest competition was WordStar 2000, but not even a fairly good rating from InfoWorld could help sell the product. The review slammed Displaywrite with a lowly 4.4 rating. Multimate, now known as Multimate Advantage II, received only a 6.3.
Our 5.0 sales for the year started out strong. Before the product’s release, we had purposefully decided not to stress its desktop publishing (DTP) features in our advertising. We knew we could compete fairly well with the other desktop publishing packages, but felt it best to promote WordPerfect as the best word processor and let others mention our desktop publishing capabilities. We were worried about scaring some customers away if we made the product sound too complicated. The strategy seemed to work well. The press quickly picked up on the DTP features, some even chiding us for not promoting them more than we did. In a PC Magazine survey conducted early in the year, almost 30 percent of readers planned to purchase their desktop publishing software from us, compared to 38 percent from Aldus, the DTP market leader. This was a great percentage for us considering we did not have a true DTP product.
January of 1989 marked the end of our crisis growth era, when we moved the last departments of the company into the research park. We now had enough money in the bank to plan and build for the future. Soon we added a cafeteria, which Bruce named the Hard Disk Cafe, and a conference center to the complex.
At this same time we were also building a residential subdivision adjacent to the research park. We purchased 45 acres of an old orchard next door and subdivided it into 40 home sites. We sold most of the lots to WordPerfect employees at cost with attractive financing. My wife and I purchased a lot right next to WordPerfect’s property, just a few hundred feet from customer support. I liked the idea of walking to work and expected to work at at WordPerfect as long as I could push my walker to the office.
With the 5.0 problems behind us, our profits rose back up to the 30% level, and we had enough money to start buying out our distributors in Europe. Our international sales were growing, and Bruce felt a WordPerfect office in each of the European countries would increase sales. We now were the market leader in all the Scandinavian countries as well as Holland and Belgium. Soon we would lead in Great Britain and Spain. We were having only limited success in France, Germany, Austria, and Switzerland, but hoped a more official presence in Europe would help improve sales in those countries.
In February, as was a tradition, we invited a few representatives from our largest accounts to come to Deer Valley to ski with us and tell us how we were doing. All of them were concerned about our lack of a WordPerfect for Presentation Manager, and some of them wanted a Windows version. We tried our best to reassure them that we would eventually run on all important computing environments. We told them that while we could not always be first to release a product for a new environment, we could catch up quickly. Our large accounts were not too impressed with our answers.
We were having to live with a lot of OS/2 hype from IBM, Microsoft, and the press, and a lot of Windows hype from Bill Gates. The Windows programming group at Microsoft had not been disbanded when the company started work with IBM on OS/2. The rumor was that one of the members of the almost forgotten Windows development team found a way to make DOS and Windows do much of what OS/2 and PM were going to do. When Scott Oki, vice president of sales for Microsoft, learned that a good version of Windows was possible, he started promoting a Windows strategy rather than an OS/2 strategy to Bill Gates and others at the top. A Windows strategy offered a lot of advantages for Microsoft. If Windows was a success, Microsoft would not have to share control of the operating environment with IBM. Even if OS/2 became the eventual winner, given its delays, Microsoft could still make a lot of money selling Windows in the interim. Bill Gates was won over to a Windows strategy and became a one-man Windows promotion team.
Interestingly, early in 1989 Microsoft was still spending more time and money pitching OS/2 than they were pitching Windows, but Windows was now their strategy for the future. Those companies which had won significant market shares in the DOS world were quick to see the danger if Windows was successful. Most of us threw our influence behind IBM and OS/2, hoping that customers would not fall into Gates’s trap. IBM was big, but they were not nearly as dangerous an applications developer as Microsoft. Given the choice of riding on the back of an elephant or a fox, we felt much safer atop the elephant.
I became an opponent of Windows out of desperation. I called it a toll booth on the road to the future. It irked me that many of our customers were willing to buy Windows as an interim solution until the OS/2 Presentation Manager was ready. Not only would it waste a lot of their money, but it was a huge amount of extra work for us to create two GUI versions for IBM PC’s. Windows was an ugly piece of software to work with and not an easy platform to support. Even Microsoft had trouble working with the product, needing two years to debug their Windows word processor.
The timing for us was awful. The beta copies of the new version of Windows would not be available to us until early 1990, so we could not start work on a Windows product unless we were willing to work with Windows 2.0. Writing for 2.0 meant redoing a lot of the work once 3.0 was ready. We were also reluctant to start work until version 5.1 of WordPerfect for DOS was finished. We did not want to release a Windows product that would be a release behind our DOS product, especially since the expectations of our customers were running high that a GUI product would have many advantages. We had little choice but to hope that IBM could somehow make OS/2 a success, so Microsoft would not have too big an advantage over us in the GUI market.
I took our theme to teach seriously. In February I started spending one half hour each Tuesday morning talking to all the new hires. Many weeks I would speak to twenty or more people. Part of my reason for meeting the new hires was to try to improve my image a little, but mostly I wanted to teach them a little about the company. Many new employees came with incredibly high expectations, thinking they would have unlimited career opportunities and salaries which would soon grow to enormous levels. I hoped that by talking to them I could bring these unrealistic notions into line with the reality we had to offer.
I explained to the new hires a little of the company’s history, its purpose, a little of what it was we did, and how we were different from other companies. I explained that although we were a wonderful company, our definition of wonderful was not necessarily the same as theirs. If, for example, the company chose not to sponsor a little league team, no one should get upset. The purpose, objectives, benefits, salaries, and charities supported by the company were defined by the owners and were sometimes different from what previous employers might have decided.
I promised the new employees that they would have a good working environment, plenty of meaningful work, good benefits and salaries (although not as high as was rumored around the valley), a Board which made careful decisions, and access to anyone in management whenever they had a problem.
In return I told them we expected them to work hard, to be reliable and honest, to avoid gossip, and to keep their salaries confidential. The last request was probably asking too much, but we were trying to pay people based on their contribution, and we did not like them wasting a lot of time comparing their salaries to others’. I also encouraged them not to make WordPerfect Corporation the most important thing in their lives. I believed that family and friends should come first, and that their jobs should not become their lives. I did not want people neglecting their families to try to get ahead at WPCorp. I was like the drunk who warned others not to drink. I was beginning to recover from my obsession with WordPerfect Corporation and feeling guilty about all the time I had spent away from home over the last nine years. I now believed employees could be more effective if their lives were well-rounded.
I should have taken a lot more time taken to explain how the company did its work and what people could expect. Too many people, especially young college graduates with no other job experience, came to work with the wrong expectations and some bad habits. Some could not come on time, some could not put in a full eight hour day, some could not divorce themselves from their personal lives once they got to work, and some were always looking for a ladder to climb and a way to move past someone else. We did not need people who were interested in a series of promotions that would help them reach a goal of fame and fortune; we needed people to answer phones, to take orders, to keep the networks running, to write software, to test software, and to call on customers. We wanted to specifically avoid those ambitious individuals seeking “challenging entry-level management positions to improve their interpersonal skills.” We were hoping to find people who would take a job and do it well, who were not just passing through on their way to the top. Unfortunately, many people became unhappy once they understood how difficult it was to move up in our flat organization.
I also spent a lot of time teaching our reps. By now we had about one hundred marketing representatives scattered across the United States and Canada. Most of them were young, recently married, and recently out of college. They were an attractive, hard working, self-motivated bunch of salespeople who loved the company and its products. They were very different from the poorly equipped and poorly trained group we sent out the first summer of the program. They were paid a good salary, given good benefits, and provided with good equipment. They were given one month of training before going out into the field and were brought back to Utah for training twice each year.
I liked getting involved in what the reps were doing. When they published their monthly reports electronically on CompuServe, I always read them and answered many of their questions. The reps were a good source of information, because they talked to real customers and knew which of our marketing programs and strategies were working. I attended and spoke at their semi-annual training meetings and invited their spouses and children to swim in our pool.
The rep program was something of a case study for me, where I could test out my management theories and try out new programs. The sales director who ran the rep program had to put up with a lot of my meddling. Together we wrote the first rep handbook, which explained the purpose, objectives, and duties of a rep. We experimented with different salary schedules, none of which included a commission, and made sure that the reps were given regular evaluations. We kept the department organization flat. For a time, every one of the one hundred plus reps reported directly to the sales director, who reported directly to me. This proved to be too many reps for one person to supervise, however. We tried using directors out in the field to help supervise them, but when the results were not very good, we used directors working out of the home office.
We tried opening offices in New York and Washington, DC, with assistants to help them, to see if they would be more effective working out of an office. Eventually we closed the offices, believing the reps were more effective working out their cars and homes. The offices did little to promote the face-to-face meetings with customers and dealers, which was the purpose of the program. Offices tended to promote the wrong types of activities, like staff meetings, paperwork, and telephone tag. We did not want our reps sitting behind desks shuffling papers. We wanted them to be with the customers, shaking their hands, looking them in the eye, and teaching them about our products.
Instead of offices, we equipped the reps with laptop computers and cellular phones. Instead of a local number, the reps’ business cards listed an 800 number which their dealers and large accounts could call to get help or information. We provided assistants, who worked at our Utah office, to answer the calls, take messages, and schedule appointments. Eventually the rep assistants in Utah knew the customers in their assigned areas almost as well as someone who would have worked out of an office in the field. We saved an enormous amount of money by not opening offices in the major cities.
The rep program had a lean budget and so did the rest of the marketing department. The industry average for marketings expenses for a company our size was about sixteen cents of every sales dollar. We spent only eleven cents. Of those eleven, five went to pay for the customer support phone and personnel costs, leaving us with six cents for marketing salaries, advertising, shows, promotions, and travel. Our competition typically spent only two cents of every dollar on support, leaving them with fourteen cents for everything else. We were very careful about spending money, relying on the strengths of our products and word of mouth advertising to do much of our marketing.
As we expected when we started the year, 1989 followed our normal script closely. Sales went from $57 million the first quarter, to $67 million the second quarter, and $68 million the third. With 5.1 on schedule to come out in the fall, we were set to have another record-breaking year.
Of course, even though overall sales were good, many of our other products were having a tough time. By now the Amiga and Atari products were unprofitable, but we could not see an easy way of getting out of those markets. Customers who purchased those products and developers who worked on them were extremely loyal to their favorite machines. If we made the slightest move to decrease our development efforts, we made instant enemies. Over the next couple of years we would lose anywhere from $20,000 to $50,000 a month on those two divisions, because we were afraid to abandon even a few of our customers. We were also publishing too many international versions of our low-selling products and ordering them in much higher quantities than we could sell. We were, however, having fairly good success with WordPerfect Office for PC Networks.
In addition to our Amiga and Atari losses, we were consistently losing money on our Macintosh, IBM 370, and OS/2 groups. The Macintosh group lost money because we were putting a lot of development resources into getting their next version ready. Our OS/2 group released a 5.0 non-GUI version, which brought in a little money, but it was not enough to come close to covering the costs of development of the upcoming GUI product. Our DG and VAX groups were fairly profitable, but they were also attempting to develop too many products. It gave us some comfort to believe that our losers were helping support our winners. We liked to think that a few Amiga customers were influential in getting their employers to choose WordPerfect as the company standard.
In September we reached our tenth anniversary, if you were counting from the time SSI first incorporated. Our PR staff made a big deal of the occasion and we received a lot of local recognition. In addition to the news stories in the financial sections, it seemed that every week we were getting a new award or special presentation. Alan was Utah’s exporter of the year and businessperson of the year for the local Chamber of Commerce. Kelly Services made us their employer of the year in Utah.
Probably the most amazing statistic in those ten years was our perfect revenue history. For every quarter during the first ten years we were in business, our collections were higher than the previous quarter. For one quarter our sales were slightly lower than the previous quarter, but the money we collected from quarter to quarter always went up. In an editorial for the quarterly newsletter that marked the anniversary, I wrote, “We intend to offer a complete, integrated, productive, easy-to-use family of business software products, and we want them to run on many computing platforms.” Looking back, I wish we had not tried to do so much. Those few customers who purchased PlanPerfect for the VAX, for example, would be very disappointed when we stopped development for that product. Very soon we would realize that we could not write every product for every machine, and eventually we would have to cancel some projects and break many of our “family of products on many platforms” promises. We did not have the resources to do everything, especially with our GUI problems.
Rolling out 5.1 should have gone smoothly, but it did not. Dan and I had always taken care of the WordPerfect roll outs, but this time we assigned the work to others. I assumed the others had watched us enough to know how to price the updates, come up with a marketing plan, produce the brochures and packaging, get the advertising ready, etc. Of course, we had neglected to provide any formal training for our successors, so as the 5.1 release day approached not much was ready.
I realized then that our training was still very deficient. When someone was given a new responsibility, I had expected them to generally do what others had done before them. Instead of learning from previous mistakes, however, the new people always seemed to want to do things on their own. The result was almost always the same. Every time responsibilities were passed from one person to another, we had to live with the blunders, many times repeating the same ones over and over again. It seemed to me that in a well-run company we should not have to let everyone learn by making their own mistakes.
Once Dan and I got involved, the roll out went smoothly. Dan took care of the advertising, and I wrote a positioning statement. We now had about 60% of the word processing market, so positioning 5.1 as an “all things to all people” product made sense. We stressed that although not everyone needed every new feature, there were a few people in every organization who needed at least a few of the features. The engineering department would like the equations. New users would like the pull down menus and mouse support. Secretaries would love the tables and labels features. Executives would like the spreadsheet links. Educators would enjoy the larger character set. Someone was bound to get excited about the improved hyphenation, the improved merge, and the more flexible tabs. Everyone would like the easier to use installation program. The result of this new major release would be that an entire organization could use just one word processor to satisfy all users, even those with very specific requirements.
WordPerfect 5.1 was network ready right out of the box, so we eliminated our higher priced network version. Because we were effectively lowering the price of the first station on a network, we used this as an excuse to raise the price for additional network stations. It made customers upset when we raised the price, but we felt we had to do it. The old $150 price for a computer attached to a network was way too low compared to the $495 price for a computer not attached to a network. Because more customers were now installing networks, we felt we had to raise the price of network stations to avoid a revenue decline.
PC Week gave us a lukewarm first review. By now we knew that the GUI had taken over the minds and hearts of many of the writers and editors of the trade publications, but we were still surprised to see complaints that WP had “quirky” commands and was difficult to use. It was clear the reviewer did not like the product when he wrote, “[WordPerfect] has some things new, some things borrowed (from the competition, that is) and some things old that will leave the users feeling blue.” InfoWorld, however, was much more complimentary. Their headline, “WordPerfect 5.1 jumps ahead of competition,” was all we needed to reassure our customers that 5.1 was a good product.
We celebrated our ten year anniversary in our COMDEX booth with a seven foot tall photo album. As the actors flipped through the pictures, they came to the highlight of our history–the new WordPerfect 5.1. In spite of the fact that everything else about COMDEX concerned either OS/2 or Windows, 5.1 was a big hit with customers. Sales would jump to $89 million for that last quarter. The tables, labels, and equations were magical, easy to demonstrate, and easy to get excited about. The visitors to the booth were a mob of happy people.
Alan and I had lunch at COMDEX with some high-placed officers from IBM. They assured us that Windows would not succeed and that an agreement had been struck between IBM and Microsoft which would effectively stop Microsoft from making a success of Windows. It seems funny now, but at the time we were impressed by what they told us. They said Windows was neutered, and we believed it.
During the COMDEX week, we held a joint press conference with Lotus to announce an alliance of sorts, one in which we agreed to work together on our OS/2 products. Lotus had developed a platform which ran on top of OS/2, and they were allowing us to use it without charge if WordPerfect would support it. IBM liked what Lotus was trying to do and had given them permission to build links between the new platform and IBM mainframes. By taking advantage of the Lotus platform, we felt we could shorten the time it would take to deliver an OS/2 product and while gaining direct access to IBM’s large machines and 1-2-3 for OS/2 spreadsheets.
Jim Cannavino, head of IBM’s PC division, came and sat on the front row to demonstrate IBM’s support for the alliance. We specifically did not invite Microsoft, even though they ached to be involved. In spite of their new Windows maneuvering, Microsoft was still trying to look like they were leading the way in promoting OS/2.
A VP from Lotus demonstrated 1-2-3 for OS/2. Alan Ashton demonstrated a very fragile, unfinished version of WordPerfect for OS/2, announcing 1990 release date. Although we were not ready to announce a date, IBM and Lotus were anxious for us to show our support for OS/2 with some announcement of when we would ship. Jim Manzi announced our Agreement of Cooperation (I persuaded Lotus not to call the relationship an alliance, because I liked Thomas Jefferson’s advice to avoid entangling alliances). I spoke last.
The part I enjoyed most about the whole experience was driving the Lotus public relations department crazy. They wanted to know what I was going to say before the press conference started, afraid I might ruin things and waste the $30,000 we were all spending on the event. They tried to get me to give them a transcript of my speech, but I explained to them that I preferred to speak extemporaneously, without a written script. Next they tried to get me to rehearse, but I told them I was not prepared. They were used to executives who used the standard speech writer, make-up artist, bodyguard, and publicist entourage, and did not know what to do with someone who did not want all the help. To further tease them, I made sure they saw me writing my speech on an envelope as Jim Manzi made his comments, just moments before I walked on stage.
Of course, I thought I did a good job. I have always loved to get up in front of people, to get their attention, to hold it, and to make them laugh. I made sure everyone knew we were not “getting into bed” with Lotus. This would have been out of character for a company from Utah. Jim Manzi said we were dating, but I tried to make it clear that we were just good friends. Much like two fathers-to-be who struck up a conversation in the maternity waiting room, we had become friends while waiting for our OS/2 babies to be born.
The friendship did not last too long, however. Microsoft would embarrass IBM greatly by making Windows a very capable product and putting more work into their Windows applications. The Lotus platform, which looked so attractive to our programmers in the fall, would not look as good to them by the next spring. Soon Lotus would acquire a word processor called Amí Pro and become a direct competitor to us. Our first try for an OS/2 baby would eventually abort, and their first version would be a sickly thing, never amounting to much.
At COMDEX Computer Reseller News always published its list of the 25 most influential computer industry executives. I had appeared on their list the two previous years, but this time I disappeared from its ranks. They had called to tell me I would be on the list and to get a quote, but I suggested they put Alan on the list instead. I was a little embarrassed about making the list when Alan and Bruce did not. I was always trying to explain that the three of us ran the company together, but I usually received more of the industry press attention because I went to more industry conferences. Although Alan was considered the CEO in Utah, and Bruce the CEO in Europe, I was generally thought to be in charge everywhere else. No one wanted to believe that the three of us ran the company together as was apparent when, despite my suggestion, Alan failed to make the Top 25 list.
I was particularly proud of one promotion we came up with for the 5.1 release. We offered to let 5.1 update customers donate their old copies of 5.0 to their favorite elementary, junior high, or high school. The offer required some paperwork, which gave us a chance to identify schools which were interested in using our products. The donations saved schools a lot of money and allowed them to use software which was fairly current. The program cost us almost nothing and convinced many more schools to use our products. The part I liked most about the promotion was that Microsoft could not duplicate the program. Their installed base of Word users was so much smaller than ours, that a similar program would have had very little effect.
We ended the year in great shape financially, with sales of $281 million and pre-tax profits back up in the 33% range. 5.1 was a great product and selling very well. We were a little disappointed, however, that the technical advances in the product were largely ignored because of all the attention paid to GUI. The focus of the industry was directed toward Windows and OS/2. Although we were having trouble delivering our OS/2 baby, our employees were having great success on their own. That year 222 real babies were born, which was more than one for every eight of our 1,612 employees.