Latest / Boardroom Conversations with Fexingo: CEOs, Strategy, and Corporate Leadership Explained / How the IBM Board Turned Around the Company with Red Hat
Transcript
- Lucas: When IBM announced it was buying Red Hat in October 2018 for $34 billion in cash, the Wall Street reaction was pretty brutal. At $190 a share, it was a 63 percent premium. I remember analysts calling it an act of desperation. Luna: And honestly, IBM had been stuck in revenue decline for almost six years at that point. Cloud was the only growth story, but they were way behind Amazon and Microsoft. Lucas: Right. So the board had a choice: keep cutting costs and shrinking the company, or place a massive bet that the next decade of enterprise computing would be hybrid — that is, companies would run workloads across their own data centers, private clouds, and multiple public clouds. That thesis was the core of the Red Hat deal. Luna: And the board bought into it. But what's interesting to me is that this wasn't just a dramatic acquisition and then a hands-off approach. The board had to oversee a really difficult cultural integration. IBM, the quintessential old-school enterprise vendor, absorbing Red Hat, the poster child of open source. Lucas: Exactly. I spoke to someone who was in the room for the integration discussions. He told me the board emphasized that Red Hat had to remain operationally independent — keep its own branding, its own management, its own developer culture. The fear was that IBM's bureaucracy would kill the very thing they paid for. Luna: That's a pretty disciplined approach. Normally in big acquisitions, the acquiring company's executives will talk about 'synergies' and 'integration' — meaning cost cuts. But the IBM board explicitly decided not to force Red Hat onto a common ERP system or shared sales quota. Lucas: Yeah, they treated it almost like a subsidiary of a holding company, not a division. And that decision went hand in hand with choosing a new CEO. Ginni Rometty had led the Red Hat deal, but the board knew that the execution phase required a different skill set. So in April 2020, at the height of the pandemic, they named Arvind Krishna as CEO. Luna: Krishna came from inside IBM — he ran the cloud and cognitive software unit — but he also had deep ties to the open source community. He was the one who negotiated the Red Hat deal from an operational side. So the board basically anointed the guy who understood the asset they had just bought. Lucas: Right. And within a year, Krishna made a bold move: he spun off the managed infrastructure services business — that was a big chunk of IBM's old IT services — into a separate company called Kyndryl. That freed IBM to focus entirely on hybrid cloud and artificial intelligence. Luna: The board backed that too, even though it meant shrinking revenue and headcount significantly. Kyndryl represented about a quarter of IBM's revenue at the time. It was a bet that smaller, faster, and higher-margin was better than larger and declining. Lucas: Let's put some numbers to it. Before the Red Hat deal, IBM's cloud revenue was around $17 billion, but most of it was lower-margin infrastructure services. By the end of 2025, IBM's hybrid cloud revenue reached about $27 billion, and the software part of that — Red Hat plus the broader cloud platform — was growing at a compound annual growth rate of about 10 percent. Luna: And the stock price took a while to catch up. Mid-2022, IBM was still trading around $130. But by mid-2026, it's been above $200. Not a home run on the scale of Microsoft's cloud pivot, but a solid turnaround for a company many had written off. Lucas: A big part of that is the board's patience. They didn't panic when the pandemic hit and corporate spending froze for a quarter. They didn't abandon the hybrid cloud thesis when Amazon and Microsoft kept winning hyperscale deals. They stuck with the vision that most enterprises would never move everything to a single public cloud. Luna: And that thesis has been proven right. Surveys from 2025 show that 87 percent of large enterprises have a multi-cloud strategy. IBM's advantage is that they can manage workloads across all those environments, including on-premise legacy systems that banks and governments can't easily move. Lucas: But let's not sugarcoat it — the board made missteps too. One of the biggest was the Watson AI initiative in the 2010s. The board poured billions into building a healthcare AI platform that never really commercialized. It became a cautionary tale about overpromising AI capabilities. Luna: To be fair, Watson was kind of ahead of its time. The technology wasn't mature, and the use cases were too vague. The board owns that oversight — they should have forced more discipline earlier. But they also learned from it. When the generative AI boom hit in the 2020s, IBM was more measured—focusing on enterprise-grade tools for code generation and customer service, not trying to cure cancer. Lucas: Right. The board's composition itself evolved. By 2025, IBM's board had a much higher share of directors with tech backgrounds — former CTOs from other large enterprises, a former head of the National Security Agency, a few venture capitalists. The old days of white-shoe lawyers and former CEOs of unrelated industrial companies were fading. Luna: That matters because a board that truly understands technology can ask better questions. When Krishna came back to the board in 2023 and said we need to invest $4 billion in AI research over the next three years, the board didn't just nod. They challenged him on how he'd measure return, which segments to prioritize, and how to avoid a repeat of Watson. Lucas: And Krishna's response was pretty concrete. He said we're targeting specific verticals — financial services, healthcare, retail — and we're going to charge per transaction, not a massive upfront license. That's a much lower-risk commercial model. Luna: I also think the board deserves credit for how they handled CEO succession. Rometty had been CEO for eight years, and the transition to Krishna was smooth. There was no palace intrigue, no activist investor forcing a change. The board started the search process two years before the actual change, and they communicated it internally and externally well. Lucas: That's rare. A lot of boards wait until there's a crisis. Look at what happened at Boeing or Disney. But IBM's board had a culture of planning. They even had a documented succession philosophy that they updated annually. It's boring, but it works. Luna: So when you step back, the IBM turnaround story is really a board story — a board that was willing to bet big, integrate carefully, spin off distractions, and support the right CEO. Not every board has that discipline. Lucas: And that's exactly the kind of boardroom conversation we love to unpack on this show. If these discussions have helped you think more clearly about corporate strategy or governance, I'll mention something quickly — this podcast stays ad-free and independent because of listeners like you. Luna: Yeah, it genuinely makes a difference. Even a couple of dollars a month helps cover the research time and the hosting fees. If you've gotten value out of these episodes, you can support the show at buy me a coffee dot com slash fexingo. Lucas: That's buy me a coffee dot com slash fexingo. All lowercase. No pressure at all — just if it's worth something to you, it keeps us going. Luna: So back to IBM — one final thought. What do you think the board is most focused on for the next two to three years? Lucas: I think it's going to be about maintaining that 10 percent cloud growth rate while AI becomes a bigger revenue driver. They also need to make sure the Kyndryl separation stays clean — no lingering cost overlaps. And then there's the threat that at some point, a big tech competitor might try to eat their lunch in hybrid cloud, even though it's a tough market to crack. Luna: Right, but that's a good problem to have. A decade ago, the question was whether IBM would survive. Now it's about how it competes. That's a boardroom achievement worth studying.