IBM — Business Trajectory Research Report
Understanding the World’s Most Resilient Technology Company
EXECUTIVE SUMMARY
What They Do: IBM helps the world’s largest organizations — banks, hospitals, governments, airlines — run their most critical technology, now with an increasing focus on hybrid cloud and AI.
Why They Exist: Most large companies have decades of technology built up in layers — old systems, new systems, cloud systems, all running at once. IBM exists to make that messy reality work, securely and reliably, at scale.
How They Make Money: Primarily through software subscriptions (especially Red Hat and automation tools) and consulting contracts — helping enterprises modernize technology and deploy AI.
Revenue (2025): ~$67.5 billion
What Makes Them Different: A 115-year-old brand that the world’s most regulated industries — banking, healthcare, government — trust to handle systems where failure is not an option. Combined with real AI and hybrid cloud products, not just consulting.
Biggest Strength: Trust, stability, and depth of enterprise relationships. IBM serves 95% of the Fortune 500.
Biggest Weakness: Slow-moving compared to born-in-cloud competitors. Perception as an “old company” hampers talent acquisition.
Biggest Opportunity: Enterprises deploying AI at scale need exactly what IBM offers: governance, security, reliability.
Biggest Risk: Microsoft Azure and AWS continuing to eat into IBM’s territory while also offering AI capabilities.
Most Likely Direction: A profitable, focused software + consulting company — not a tech giant in the traditional sense, but an indispensable partner to the organizations that run the world.
SECTION 1: The Company Story
What Is IBM?
IBM — International Business Machines — is one of the oldest and most continuously operating technology companies in the world. But calling it a “tech company” the way you’d describe Google or Apple misses the point. IBM’s world is different.
Its customers aren’t consumers downloading apps. They’re the CIO of a global bank trying to keep transaction systems running while also adopting AI. They’re a hospital system that needs to modernize software built in 1995 without shutting down patient records. They’re a government agency that stores sensitive citizen data and can never afford a breach.
For these organizations, IBM is not exciting. IBM is necessary. And necessary, in the enterprise world, is often worth more than exciting.
The Founding Story
In 1911, a financier named Charles Ranlett Flint merged four small companies — including the Tabulating Machine Company, founded by Herman Hollerith — into a single entity called the Computing-Tabulating-Recording Corporation (CTR). The problem these companies were solving was real and urgent: businesses and governments were drowning in data they couldn’t process fast enough. The U.S. Census Bureau had used Hollerith’s punch-card machines to tabulate the 1890 census in just one year, versus the eight years it had taken by hand for the 1880 census.
In 1914, Thomas J. Watson Sr. took over the company. He was a salesman through and through, and he understood that big companies would pay handsomely for anything that helped them manage information. In 1924, he renamed the company International Business Machines — signaling global ambition at a time when most companies thought locally.
Watson’s insight was simple but powerful: businesses need to process information to function, and whoever makes that easier wins. For most of the 20th century, IBM was the answer to that problem.
Mission in Plain English
IBM’s official language talks about “hybrid cloud and AI” and “technology and business expertise.” But translated into plain English, IBM’s mission is this:
Help the world’s most complex organizations use technology reliably, securely, and intelligently — even when those organizations have decades of messy legacy systems to deal with.
SECTION 2: The Customer Problem
To understand IBM, you have to understand the specific nightmare it solves.
The Problem: Technology Doesn’t Age Well, But Organizations Can’t Stop
Imagine a bank that was founded in 1970. In 1980, they built their core banking system on IBM mainframes. In 1995, they added a customer portal. In 2010, they launched a mobile app. In 2020, they moved some data to the cloud. Now in 2025, they want to use AI for fraud detection — but that AI needs to talk to the 1980 mainframe, the 1995 portal, the 2010 app, and the 2020 cloud, all at the same time, without anything breaking, leaking data, or violating financial regulations.
This is not a hypothetical. This is the lived reality of most large banks, insurers, hospitals, and governments on Earth.
Before IBM (or a company like it): These organizations either hire thousands of internal IT staff to manage the chaos, or they patch things together with duct tape and hope nothing breaks.
What IBM Offers: A company that understands both the old technology and the new technology, can consult on how to modernize, and sells the software platforms needed to do so securely. IBM also has the credibility to walk into a regulated industry and say, “We’ve been doing this since before your CIO was born. Trust us.”
What Customers Are Willing to Pay For: Reliability. When a hospital’s patient management system goes down, people can die. When a bank’s transaction system fails, billions of dollars freeze. IBM’s customers pay a premium not for the coolest technology, but for the most dependable technology.
SECTION 3: Product Portfolio
IBM today has three major business areas. Think of them as three legs of a stool.
Leg 1: Software (~$30B revenue, largest and fastest-growing segment)
This is IBM’s most important business and the one growing fastest.
Red Hat — the crown jewel of IBM’s software portfolio, acquired for $34 billion in 2019. Red Hat makes the enterprise version of Linux (the operating system that quietly runs most of the internet) and OpenShift (a platform that lets companies run software across multiple cloud environments without being locked into AWS or Azure). The key word is open source — companies love it because it doesn’t chain them to one vendor.
Watsonx — IBM’s AI platform, launched in 2023. Unlike ChatGPT, which is a consumer chatbot, watsonx is an industrial AI toolkit for enterprises. It has three layers:
- watsonx.ai: where you build and train AI models
- watsonx.data: where you manage the data those models run on
- watsonx.governance: where you make sure the AI behaves legally, ethically, and consistently
Automation Tools — software that helps companies automate repetitive business processes (think: automatically routing insurance claims, generating invoices, managing IT tickets).
Transaction Processing Software — the unglamorous but extremely sticky software that runs inside banks and insurance companies to process millions of transactions per day. Customers rarely leave because switching is terrifyingly risky.
Leg 2: Consulting (~$20B revenue)
IBM Consulting (formerly IBM Global Services) deploys about 160,000 consultants worldwide to help companies transform their technology. These aren’t just PowerPoint strategists — they actually implement the technology.
What they do: Design and execute multi-year digital transformation programs. A typical engagement might look like: “Help this airline modernize its reservation system, migrate to hybrid cloud, and deploy AI for maintenance prediction — over 3 years.”
Who pays: The same Fortune 500 companies that buy IBM’s software, which creates a virtuous cycle.
Leg 3: Infrastructure (~$12B revenue, declining but profitable)
This is IBM’s oldest business — the mainframe computers and supporting systems that still quietly power an enormous amount of global commerce. IBM mainframes process about $10 trillion in transactions per day worldwide. This business is shrinking as customers gradually modernize, but the margins are high and the switching costs are enormous.
Simple Comparison Table
| Segment | Revenue | Growth | Margin | Stickiness |
|---|---|---|---|---|
| Software (Red Hat, watsonx, automation) | ~$30B | High (+10%) | Very High | Very High |
| Consulting | ~$20B | Flat/Low | Medium | Medium |
| Infrastructure (Mainframe) | ~$12B | Declining | High | Extremely High |
SECTION 4: Business Model
How IBM Makes Money
IBM’s business model is built around a simple idea: get large organizations dependent on your software and services, then keep them.
Step 1 — Land with a consulting engagement. A bank hires IBM Consulting to help them think through a cloud migration strategy. IBM earns a consulting fee, and more importantly, it gets inside the organization.
Step 2 — Sell the software platform. Once IBM understands the client’s environment, it recommends Red Hat OpenShift for the cloud layer, watsonx for the AI layer, and automation tools for the efficiency layer. These are sold on subscription — recurring monthly or annual fees.
Step 3 — Expand the footprint. Once a company is running Red Hat in one division, IBM helps expand it to other divisions. Each new use case adds subscription revenue.
Step 4 — Stay forever. The embedded software becomes infrastructure. Removing IBM from a bank’s technology stack is like removing a load-bearing wall from a skyscraper. It’s theoretically possible but practically terrifying. This is called “lock-in,” and it’s the foundation of IBM’s profitability.
Revenue Mix: IBM has deliberately shifted toward recurring software revenue, which is more predictable and profitable than project-based consulting. Today, over 50% of IBM’s revenue is recurring — a massive change from a decade ago when it was primarily a services company.
Why This Business Is Attractive
- Recurring revenue: Software subscriptions renew automatically.
- High switching costs: Replacing IBM software in a regulated bank or hospital is a multi-year, multi-hundred-million-dollar project. Few companies do it voluntarily.
- Premium pricing: Regulated industries pay for trust and compliance support, not just functionality.
- Cross-sell potential: A client using Red Hat is a natural candidate for watsonx.
SECTION 5: The Growth Journey
Stage 1: The Golden Age (1914–1980s)
Thomas Watson turned IBM into the dominant force in business computing. In the 1950s and 60s, IBM’s mainframes were the computers — not just one option, the option. The U.S. government used them. Airlines used them. Banks used them. IBM was so dominant that the U.S. Justice Department filed an antitrust lawsuit against it in 1969 (it was eventually dropped in 1982).
In 1981, IBM invented the Personal Computer (PC) — and then made a decision it would regret for decades: it used an open architecture, letting companies like Microsoft (for the operating system) and Intel (for the chip) become the real winners of the PC revolution.
Stage 2: The Painful Decline (1990s–2000s)
As PCs replaced mainframes in many business applications, and as companies like Dell commoditized PC hardware, IBM’s dominance eroded. The company that had once defined business computing was losing relevance.
IBM sold its PC business to Lenovo in 2005. It sold its semiconductor business. It shed division after division. The stock was propped up by financial engineering — stock buybacks — rather than genuine business growth. For nearly a decade, revenues declined.
Stage 3: The Bet on Services (2000s–2018)
IBM tried to reinvent itself as a services company — the consultants who help enterprises use technology, rather than the company that makes the technology. This worked reasonably well but created a new problem: IBM was now competing with Accenture, Infosys, and TCS on price and talent, rather than competing on technology superiority.
Then came Watson. IBM’s AI beat Jeopardy champions in 2011 and the world declared IBM relevant again. But Watson struggled to find commercial traction — it was impressive in a lab, less impressive when deployed in messy real-world environments.
Stage 4: The Pivot That Actually Worked (2019–Present)
Two decisions changed IBM’s trajectory:
Decision 1: Buy Red Hat for $34 billion (2019). This was IBM’s largest acquisition ever and many analysts thought IBM overpaid. They were wrong. Red Hat gave IBM the most important piece of the hybrid cloud market — the technology that lets companies run software across multiple clouds without vendor lock-in. This turned out to be exactly what large enterprises needed.
Decision 2: Spin off Kyndryl (2021). IBM carved out its managed infrastructure services business (90,000 employees, $19B in revenue) and spun it off as a separate public company called Kyndryl. This was a surgical removal of a slow-growing, low-margin business that was dragging down IBM’s profile. The remaining IBM was now primarily a software and consulting company — cleaner, higher-margin, and more focused.
Arvind Krishna, who became CEO in 2020 (and had led the Red Hat acquisition), executed both moves. Under his leadership, IBM has delivered consistent mid-single-digit revenue growth and expanding margins. The watsonx AI platform, launched in 2023, has accumulated over $12 billion in bookings — far outpacing early analyst skepticism.
Key Turning Points
| Year | Event | Significance |
|---|---|---|
| 1911 | Founded as CTR | Origin |
| 1924 | Renamed IBM | Global ambition declared |
| 1981 | Invented PC | Short-term win, long-term mistake |
| 1993 | Lou Gerstner becomes CEO | Saved IBM from collapse |
| 2011 | Watson wins Jeopardy | AI ambition signaled |
| 2019 | Red Hat acquired for $34B | Hybrid cloud foundation |
| 2020 | Arvind Krishna becomes CEO | Strategic pivot accelerated |
| 2021 | Kyndryl spun off | IBM refocused on software + consulting |
| 2023 | watsonx launched | Enterprise AI platform established |
SECTION 6: The Competitive Landscape
IBM competes on multiple fronts simultaneously, which is both a strength (diversification) and a weakness (no single clear identity).
The Cloud Giants: AWS, Microsoft Azure, Google Cloud
These are IBM’s most dangerous competitors — and also the companies IBM often has to partner with. A bank using IBM consulting to deploy AI might do it on Azure or AWS, not IBM Cloud.
The honest truth: IBM has largely conceded the public cloud infrastructure battle to AWS and Azure. IBM Cloud exists but has a small market share. IBM’s strategy is instead to be the software and consulting layer on top of whatever cloud a customer uses — the Switzerland of cloud, working with everyone.
Why customers choose AWS/Azure over IBM Cloud: More services, more innovation, lower prices for pure infrastructure, stronger developer communities.
Consulting Competitors: Accenture, TCS, Infosys, Capgemini
Accenture is IBM Consulting’s most direct rival — both fight for the same large digital transformation contracts. Accenture has generally been seen as more agile and better at management consulting; IBM as deeper in technology implementation, especially involving IBM’s own products.
Why customers choose Accenture: Broader strategy capabilities, faster delivery, strong relationships with Microsoft/AWS. Why customers choose IBM Consulting: Deeper technology integration when IBM software is involved; stronger credentials in regulated industries; combined hardware/software/consulting offering.
Enterprise Software: SAP, Oracle, Salesforce, Microsoft
These companies compete with IBM’s software portfolio in areas like automation, data management, and AI tools.
The key differentiator: IBM’s software is specifically designed to work across multiple clouds and on-premises environments — the hybrid reality that most large enterprises live in. SAP and Oracle have their own cloud platforms but are increasingly trying to be open as well.
Competitive Comparison Matrix
| Competitor | Strength vs IBM | IBM’s Edge |
|---|---|---|
| AWS | Vast cloud infrastructure | IBM’s hybrid/multi-cloud flexibility |
| Microsoft Azure | Office 365 + AI integration | IBM’s neutrality (non-Microsoft shops) |
| Accenture | Agility, strategy consulting | IBM’s technology depth + own software |
| Red Hat competitors | Limited (Red Hat leads OSS enterprise) | Market leadership position |
| Oracle/SAP | Established enterprise footprints | Open-source, hybrid positioning |
SECTION 7: The Customer Perspective
Who IBM’s Ideal Customer Is
IBM’s sweet spot is a specific type of organization:
- Large enterprise (5,000+ employees), ideally Fortune 500
- Operates in a regulated industry (banking, healthcare, insurance, government, utilities)
- Has been around long enough to have significant “legacy” technology (systems 10–30 years old)
- Cannot afford downtime or data breaches
- Has budget to pay premium prices for trusted solutions
- Is trying to modernize without “ripping and replacing” everything
In short: large, old, regulated, and terrified of risk.
What Customers Love
- Trust and stability. IBM has been around for 115 years and will be around for 115 more. For a bank or hospital, that matters enormously.
- Deep expertise in regulated industries. IBM understands HIPAA, SOX, GDPR, and the regulations that keep regulated-industry CTOs up at night.
- The hybrid cloud philosophy. Customers love Red Hat’s approach: run your workloads anywhere (AWS, Azure, on-premises) without being trapped with any one provider.
- Enterprise-grade AI. watsonx’s focus on governance, auditability, and responsible AI resonates with legal and compliance teams.
What Customers Dislike
- Pricing. IBM is expensive. Enterprise contracts can run into tens of millions annually.
- Speed of innovation. Compared to younger cloud companies, IBM can feel slow to release new features.
- Sales complexity. IBM’s sales process is notoriously complex — multiple teams, long negotiation cycles, hard-to-understand pricing.
- Legacy perception. Some internal IT stakeholders push back on IBM because it’s seen as the “old guard” rather than the exciting new option.
- Past overpromises. Watson’s early healthcare hype (IBM promised it would cure cancer; it didn’t) left some buyers skeptical of IBM’s AI claims.
What Customers Want More Of
- Simpler pricing and faster deployment
- Better integration between IBM’s many software products
- More modern developer experience tools to compete with AWS and Azure
- Faster AI results with less consulting overhead
SECTION 8: The Challenges
Challenge 1: The “Late to Cloud” Problem (Past, Partially Resolved)
IBM was slow to invest in public cloud and spent years playing catch-up to AWS and Azure. The Red Hat acquisition was the most effective response — it gave IBM a differentiated hybrid cloud story rather than trying to be a smaller AWS. The challenge is largely addressed strategically, but IBM still lacks mindshare among cloud-native developers.
Confidence: High
Challenge 2: The Revenue Transition (Ongoing)
IBM is actively transitioning from project-based consulting and hardware revenue (lumpy, declining) to software subscriptions (recurring, growing). This transition creates short-term pain — consulting revenue has been flat or slightly declining — while the software segment grows. Managing this transition without missing analyst expectations is a constant tightrope walk.
Confidence: High
Challenge 3: The Watson Hangover (Partially Resolved)
IBM’s original Watson AI was over-hyped and under-delivered, particularly in healthcare. This has made some enterprise buyers skeptical of IBM’s AI claims. watsonx is a fundamentally different product — more focused, more practical, less hype — but IBM has to overcome the institutional memory of failed Watson pilots.
Confidence: High
Challenge 4: Talent Acquisition (Ongoing)
IBM competes for AI and cloud engineers against Google, Meta, Amazon, and hundreds of well-funded startups offering higher salaries, more exciting problems, and better perks. IBM’s reputation as an “old company” makes recruiting top talent difficult.
Confidence: High — this is a structural challenge without an easy solution
Challenge 5: Consulting Margin Pressure (Ongoing)
IBM’s consulting business faces price pressure from lower-cost Indian IT firms (TCS, Infosys, Wipro) who can often do similar work at 30-50% lower rates. IBM tries to differentiate on complexity and technology integration, but price remains a real battleground.
Confidence: High
SECTION 9: Organizational Understanding
(Based on public information — labeled accordingly)
IBM is a large, mature organization with approximately 250,000+ employees globally after the Kyndryl spin-off.
Leadership: Arvind Krishna (Chairman & CEO since 2020) came up through IBM’s technology side — he built IBM Cloud and led the Red Hat acquisition. This is notable: IBM’s CEO is a technologist, not a salesperson or financial engineer, which is a change from previous leadership.
Segments (post-2025 restructure):
- Software — product development, Red Hat integration, watsonx
- Consulting — Strategy & Technology, Intelligent Operations (client-facing delivery)
- Infrastructure — mainframe, storage, IBM Cloud
Organizational Strengths:
- Deep industry specialization (financial services, healthcare, government verticals)
- Strong research arm (IBM Research has produced more patents than any company for decades)
- Global delivery model across 175+ countries
Organizational Weaknesses:
- Large bureaucracy can slow decision-making
- Historically siloed between hardware, software, and services divisions
- Cultural tension between “old IBM” and the post-Red Hat, startup-culture teams
SECTION 10: Future Opportunities
Opportunity 1: Enterprise AI Governance ⭐ (Highest Potential)
Every large company is trying to deploy AI, and almost every legal, risk, and compliance team is terrified of it. What happens when an AI makes a discriminatory lending decision? What if an AI hallucinates in a medical context? IBM’s watsonx.governance — which helps companies track, audit, and control AI behavior — is positioned perfectly for this anxiety.
Why it matters: As AI regulation increases globally (EU AI Act, emerging U.S. frameworks), companies will need exactly this capability. IBM is one of the few AI companies with a credible governance story.
Difficulty: Medium. Requires continuous investment and credibility-building.
Opportunity 2: Mainframe Modernization (Conversion to Gold)
The world’s mainframes aren’t going away — but they can be made smarter. IBM is actively helping mainframe customers add hybrid cloud connectivity and AI workloads to their existing mainframe infrastructure, rather than replacing it. This turns the “dying” infrastructure business into an AI delivery platform.
Why it matters: Mainframe customers have enormous data assets sitting idle. Making that data AI-ready is a multi-billion dollar opportunity.
Difficulty: Medium.
Opportunity 3: Quantum Computing (Long-Term Bet)
IBM has the most advanced commercial quantum computing program in the world. It targets a full-scale, error-corrected quantum computer by around 2029. Quantum could revolutionize drug discovery, financial risk modeling, and materials science — all areas where IBM’s existing clients operate.
Why it matters: First-mover advantage in quantum could be transformative.
Difficulty: Very High. Timeline is uncertain; commercial applications are still years away.
Opportunity 4: AI Agent Automation for Enterprises
The next wave of enterprise AI isn’t chatbots — it’s autonomous agents that can execute multi-step business processes (book a flight, process a claim, reconcile accounts). IBM is investing heavily in “digital labor” solutions — AI that actually does work rather than just advising on it.
Why it matters: This is a massive market that plays directly to IBM’s strength in understanding complex enterprise workflows.
Difficulty: Medium-High.
SECTION 11: Future Risks
🔴 High Risk: Microsoft’s Bundling Power
Microsoft sells Azure, Office 365, Teams, GitHub, and now Copilot AI as a single suite to enterprises already using Microsoft products. A company whose employees use Outlook and Teams every day is heavily incentivized to also use Azure for cloud and Microsoft Copilot for AI — even if IBM’s offering is technically superior. This “bundling” effect is IBM’s most persistent threat.
🔴 High Risk: Consulting Commoditization
If AI tools make it easier for enterprises to self-serve on technology modernization — or if lower-cost offshore firms get better at complex work — IBM’s consulting margins could compress significantly.
🟡 Medium Risk: Watson Credibility Hangover
If watsonx fails to deliver measurable results for early enterprise customers, the narrative could quickly become “IBM overpromised AI again.” The $12B+ in watsonx bookings is encouraging but deployment results will matter more in 2025–2026.
🟡 Medium Risk: Geopolitical/Regulatory Disruption
IBM operates in 175 countries. Trade tensions, data sovereignty laws (particularly in the EU and China), and government procurement rules can all create headwinds for global contracts.
🟢 Low Risk: Core Customer Abandonment
IBM’s most embedded customers — the banks running mainframes, the airlines using IBM transaction software — are extremely unlikely to leave. The switching costs are too high. This core is highly stable.
SECTION 12: The Business Trajectory — The Story
IBM’s story is one of the most remarkable reinvention narratives in business history. But to understand where it’s going, you have to understand what it nearly became.
By the early 2010s, IBM was a cautionary tale in the making. Its legendary consulting business was growing slower than the market. Its Watson AI had been hailed as the dawn of a new era and then quietly failed to live up to the hype. Hardware revenues were collapsing. The stock had been “managed” with buybacks for so long that critics questioned whether IBM was a real technology company anymore or just a financial engineering exercise in a technology costume.
Then came two unlikely heroes: Linux and a Canadian tech company named Red Hat.
Red Hat had built a business on something the tech industry had initially laughed at — giving away the operating system for free (Linux) and charging enterprises for support, security updates, and enterprise-grade packaging. It turned out this was exactly what the enterprise market needed. And Red Hat’s OpenShift product was perfectly positioned for the hybrid cloud era — a world where companies wanted to run software in their own data centers, on AWS, on Azure, and on Google Cloud, all at the same time, without being trapped.
When IBM’s then-cloud chief Arvind Krishna pushed IBM to acquire Red Hat for $34 billion in 2019, many analysts winced. The price was steep. The logic wasn’t immediately obvious. But Krishna understood something crucial: the future of enterprise technology wasn’t “everyone moves to AWS.” The future was messier — a tangle of old and new, cloud and on-premises, regulated and unregulated. And IBM, with Red Hat, was the only major vendor specifically designed to thrive in that tangle.
Then came the Kyndryl move. By spinning off the low-margin infrastructure services unit — the thousands of technicians who staffed corporate data centers — IBM shed what amounted to a $20 billion drag on its growth story and emerged as a leaner, higher-margin, software-first company.
Then came AI. And this time, IBM didn’t make the Watson mistake of promising to cure cancer. Instead, it built watsonx as industrial plumbing — unglamorous, governance-focused, designed for the compliance officer as much as the data scientist. And it worked. Over $12 billion in bookings before the product was even two years old.
Today’s IBM is not the IBM of the 1990s mainframe era, or the confused IBM of the 2010s. It is a company that has, after a decade of painful surgery, positioned itself at the intersection of three trends it did not create but is uniquely equipped to capitalize on: the hybrid cloud reality of large enterprises, the AI adoption imperative, and the regulatory pressure around both.
What could make IBM succeed: Continued execution on watsonx, AI regulation increasing demand for governance tools, and enterprises finding that Red Hat’s hybrid cloud approach is the only sane solution to their multi-cloud complexity.
What could make IBM fail: Microsoft successfully selling Copilot + Azure as a bundle that obviates IBM’s role; a major watsonx failure that revives the “IBM overpromises AI” narrative; or a talent exodus that leaves IBM unable to innovate fast enough.
The most likely outcome? IBM won’t be the flashiest name in technology. It will continue to be the most necessary one — the company that makes the world’s most important organizations run, reliably, year after year after year.
That has been IBM’s story for 115 years. There’s little reason to think the next 15 will be different.
SECTION 13: Key Strategic Takeaways
For a new employee, consultant, investor, or strategic partner, here are the most important things to internalize about IBM:
- IBM’s competitive moat is trust in high-stakes environments. Banks, hospitals, and governments will pay premium prices for a vendor they trust not to fail them. IBM has 115 years of that trust.
- The Red Hat acquisition was the right move. The hybrid cloud thesis has proven correct. Enterprises aren’t moving entirely to AWS — they’re navigating a messy multi-cloud, multi-legacy reality. Red Hat OpenShift is the infrastructure of that reality.
- watsonx is IBM’s best AI bet ever — because it’s not trying to win the chatbot race. IBM has learned from Watson’s failures. watsonx targets the unsexy but lucrative problem of enterprise AI governance, deployment, and data management.
- The Kyndryl spin-off cleaned up the income statement. IBM is now a higher-margin, software-first company. The transformation isn’t complete, but the direction is clear.
- Microsoft is the existential threat. Not because Microsoft is smarter or better, but because enterprises that already use Office, Azure, and Teams face enormous friction in also buying IBM AI tools. IBM needs to win in environments where Microsoft is not already embedded.
- The mainframe is not a liability — it’s a Trojan horse. Those ancient mainframes are the gateway to enormous enterprise data estates. IBM is converting mainframe customers into AI customers, one workload at a time.
- Revenue growth will be modest but margins will expand. IBM is not trying to grow at 20% per year. It is trying to shift its revenue mix toward high-margin recurring software, while AI consulting drives enterprise expansion. This is a profitable, defensible business — not a hypergrowth story.