Why IBM bought Lotus

On July 6, 1995, IBM bought Lotus Development for $3.5 billion. Lotus had once been the second largest software publisher in the world and was worth $5.5 billion at its IPO. Its flagship product, the spreadsheet Lotus 1-2-3, had been a key driver in sales of the original IBM PC in the early 1980s. By 1995, Lotus wasn’t trending in the right direction, but IBM had plans and thought Lotus fit into them.

What IBM saw in Lotus

In 1995, IBM still had designs on competing with Microsoft head-on. IBM had operating systems and a database, but they were lacking in the applications software department, both on the desktop and on the server. Lotus had the #3 office suite, behind Microsoft Office and Novell’s Perfect Office. But the key was Lotus’ e-mail/backoffice server called Domino, and a client for it called Notes. Domino was doing what Sharepoint and Exchange do now, but it was doing it in the mid 90s.

Meanwhile, Novell had been doing similar things, building up a collection of software to go head to head with Microsoft. IBM couldn’t compete alone, but with Lotus they could.

How they fit together

In the Computer Chronicles episode covering COMDEX 1994 that aired September 5, 1994, IBM’s Lee Reiswig said OS/2 delivered everything Windows 95 was promising and everything Windows NT had wanted to be, and that you could buy it right then.

The problem was native 32-bit software. You could run 16 bit Windows applications on OS/2, but you didn’t get the full benefit unless you were running native 32-bit software, just like the Windows platforms. Lotus had 32-bit OS/2 versions of its office suite, its Notes client, and Domino server, but didn’t really promote it.

Buying Lotus gave IBM options. In June 1995, Windows 95’s release was no longer a distant thing. Everyone knew Windows 95 and Office 95 would debut near the end of the summer. Everyone knew they would sell well. What wasn’t certain yet was how well it would work once it got into the hands of millions of consumers. There was still a little bit of doubt about it, the possibility that it would go in the direction that Windows Vista later did, being too slow for current hardware, or in the direction that Windows ME did, being too unstable.

If that happened, IBM could pitch OS/2 and Smartsuite as the smarter alternative. That didn’t happen, and  consumers overwhelmingly opted for Windows 95. But IBM still found a willing audience in the business market.

Selling the IBM-Lotus software stack to businesses

Businesses tend to adopt new operating systems more slowly than consumers do. While many businesses did indeed adopt Windows 95, OS/2 and Smartsuite did do well for several years in the business market. I know from experience. In 1998, I was miserable in my job, but I had a couple of years of experience with both Windows NT 4.0 and OS/2. I could still find OS/2 jobs to apply for in 1998, partly because OS/2 ran better on laptops than NT 4 did. NT had no power management to speak of, so OS/2 offered the stability of NT and the battery life of Win95.

I wonder a lot what would have happened if I had taken one of those jobs instead of doing what I did, but I digress.

And with control of Lotus, IBM could ensure the continued availability of the key software it needed to sell into that market, regardless of what the consumer world did. Meanwhile, IBM could sell Lotus software into Microsoft shops as well. Accounting departments liked having Lotus 1-2-3 because it was familiar, and Notes and Domino were useful regardless of whether you ran them on Windows, OS/2, AIX (IBM’s Unix), or any combination of them. In a mixed environment, IBM could thrive. And being a large corporation, IBM understood that large businesses tended to migrate slowly.

IBM even went so far as to offer native versions of Notes and Domino for Windows NT 4.0 running on DEC’s Alpha CPU. While Microsoft pushed Windows running on x86, that left plenty of niches for IBM to find and sell into, and they did. Even when it meant compiling for a CPU that competed with IBM’s own PowerPC CPU.

This idea would have been ludicrous in 1992. But under Lou Gerstner, IBM no longer insisted on selling you the whole product stack.

Thriving in niches

IBM and Lotus thrived in niches, even as the opportunity for mainstream success withered away. Domino was a million times better than Exchange, but Exchange and Outlook worked with cool toys that its client, Lotus Notes, didn’t. Domino’s momentum didn’t go away all at once, but trendy turn-of-the-century gadgets steadily eroded it.

Exchange and Outlook worked just fine with a Palm Pilot. It took IBM forever to get Notes working with a Palm Pilot, and once they did, the cool thing was a Blackberry. The cool toys worked with Exchange, so Notes got relegated to legacy things. Companies that had Notes and had used it for e-mail would inevitably build other corporate intranet applications with it as well. And once they had migrated off to Exchange, there was always something that worked under Notes and had no equivalent under SharePoint.

IBM guessed correctly that Lotus Notes would be sticky like IBM’s mainframes. But IBM wasn’t able to use that to hold onto the rest of the market. Nor was it able to turn those niches where Notes stuck into a growth market. Domino looked like a redundant system running next to Microsoft technologies. And when you’re a redundant system, every incoming CIO wants to cut you out. Even when that fails, you probably won’t be selling many upgrades.

Lotus and IBM: Unhappy together

Lotus never seemed to especially like being part of IBM. This was visible at conferences, where Lotus employees would cross out IBM on their badges and write in “Lotus.” But there isn’t much reason to believe they would have done much better on their own. The momentum was on Microsoft’s side long before IBM acquired Lotus. In 2018, IBM sold what was left of Lotus to HCL, an India-based software conglomerate, for $1.8 billion.

Buying Lotus wasn’t a mistake

My contrarian view is that IBM buying Lotus wasn’t a mistake. Hear me out.

IBM was placing a bet with Lotus. And it’s easy to look at the $3.5 billion that IBM paid for Lotus versus the $1.8 billion IBM sold Lotus for, and say IBM lost. But in the early years, IBM was turning profits of $1-2 billion in software per quarter, and Lotus Notes was a big part of that. The $1.7 billion that IBM lost on Lotus represents one quarter’s worth of software profits from the 90s.

IBM’s Lotus acquisition represents a merger of two past-their-prime companies. The union didn’t return either of them to their early 1980s heights and it’s tempting to think Win95-era Microsoft smashed them like a bug. But that’s a massive oversimplification that ignores a decade of IBM and Lotus quietly doing better together in Microsoft’s shadow than most of us remember. And if Microsoft had timed its missteps a bit differently, it’s possible the Lotus acquisition could have yielded a few more good years than it did.

Dave Farquhar

David Farquhar is a computer security professional, entrepreneur, and author. He has written professionally about computers since 1991, so he was writing about retro computers when they were still new. He has been working in IT professionally since 1994 and has specialized in vulnerability management since 2013. He holds Security+ and CISSP certifications. Today he blogs five times a week, mostly about retro computers and retro gaming covering the time period from 1975 to 2000.