Ready to transform artificial intelligence from a buzzword into your personal revenue generator

HubSpot’s groundbreaking guide "200+ AI-Powered Income Ideas" is your gateway to financial innovation in the digital age.

Inside you'll discover:

Download your guide today and unlock a future where artificial intelligence powers your success. Your next income stream is waiting.

Get Your Guide

👋 Hi, it’s Rohit Malhotra and welcome to the FREE edition of Partner Growth Newsletter, my weekly newsletter doing deep dives into the fastest-growing startups and S1 briefs. Subscribe to join readers who get Partner Growth delivered to their inbox every Wednesday morning.

If you’re new, not yet a subscriber, or just plain missed it, here are some of our recent editions.

Ready to transform artificial intelligence from a buzzword into your personal revenue generator

HubSpot’s groundbreaking guide "200+ AI-Powered Income Ideas" is your gateway to financial innovation in the digital age.

Inside you'll discover:

Download your guide today and unlock a future where artificial intelligence powers your success. Your next income stream is waiting.

Get Your Guide

_{Interested in sponsoring these emails? See our partnership options here.}

Subscribe to the Life Self Mastery podcast, which guides you on getting funding and allowing your business to grow rocketship. 

Previous guests include Guy Kawasaki, Brad Feld, James Clear, Nick Huber, Shu Nyatta and 350+ incredible guests.

IBM acquiring Confluent isn't a data play. It's a structural move from a company that watched AI agents stall at the data layer and decided it couldn't keep selling intelligence without owning the infrastructure underneath it.

On the surface: one of enterprise tech's oldest names buying the streaming platform that 40% of the Fortune 500 already runs on. A logical tuck-in. A capability acquisition. The kind of deal that gets filed under "strategic."

Look closer, and the logic gets sharper.

Confluent wasn't struggling. It was scaling. Built on Apache Kafka, the de facto standard for data streaming, it had quietly become the operational backbone for Michelin, BMW, L'Oréal, and Ticketmaster. Not a proof-of-concept. Real enterprises, real volume, real-time data moving across hundreds of systems simultaneously.

IBM, meanwhile, had watsonx. It had hybrid cloud. It had consulting relationships with every large enterprise that matters. It had the AI stack. It needed the live data layer to sit underneath it.

So instead of spending years building real-time streaming infrastructure from scratch, IBM bought the team that already did it, for $11 billion, at $31 per share.

The deal closed March 17. But the signal it sends is bigger than the price tag. AI agents don't run on yesterday's data. And IBM just made sure its don't have to.

Confluent didn't start as an $11 billion acquisition target with 6,500 enterprise customers, Fortune 500 penetration, and a streaming platform embedded in the operational fabric of global industry. It started as a problem statement. Data was moving too slowly. Enterprises were making decisions on information that was already hours old. And someone needed to fix the plumbing before the next wave of computing made that lag catastrophic.

Jay Kreps didn't build Confluent inside a database hype cycle chasing analytics dashboards and BI tools. He built it around Apache Kafka, a system he had already helped create at LinkedIn to solve one of the hardest problems in distributed systems: how do you move enormous volumes of data across an enterprise in real time, reliably, at scale? Kafka became the answer. Confluent became the company built to bring that answer to the rest of the world.

The timing looked infrastructure-nerdy. It was actually foundational.

While the rest of the industry debated cloud migration strategies and data lake architectures, Confluent was quietly becoming the nervous system of enterprise operations. Michelin ran its global supply chain across 170 countries on it. BMW streamed IoT data from 30 production sites through it. Ticketmaster processed live inventory and customer activity across hundreds of systems using it. These weren't pilots. These were mission-critical workloads that couldn't afford to fail.

The company built its customer base the hard way. Enterprise by enterprise. Integration by integration. With a product that required real technical commitment to adopt and delivered real operational leverage once deployed.

By the time IBM came calling in early 2026, Confluent had over 6,500 enterprise customers, a leadership position in the Forrester streaming data platform Wave, and infrastructure that the world's largest hybrid cloud and AI company had decided it needed more than it needed to build.

Here's what makes the structure interesting: IBM is acquiring a company with 6,500 enterprise customers, a streaming platform processing mission-critical workloads across global industry, and a technical architecture that would take years to rebuild from scratch. The question isn't whether $11 billion is the right price. The question is whether IBM had any real alternative.

The timeline tells the story.

2014: Jay Kreps co-founds Confluent to commercialize Apache Kafka, a distributed streaming system he helped build at LinkedIn. Enterprise data infrastructure is not a glamorous market. The team builds anyway.

2019 to 2024: The customer base compounds. Michelin, BMW, L'Oréal, Ticketmaster, and thousands of others embed Confluent into their operational core. The 6,500 enterprise customer figure stops being a projection and starts being a renewal cycle.

March 17, 2026: IBM closes the acquisition for $11 billion at $31 per share. The enterprise AI market gets the message.

But here's what IBM actually bought.

The infrastructure: A real-time data streaming platform built on Apache Kafka, the industry standard, with the technical architecture to plug directly into watsonx, IBM Z, IBM MQ, and the hybrid cloud stack IBM has spent a decade assembling.

The customers: 40% of the Fortune 500 already running on Confluent. Not evaluations. Not proofs of concept. Production workloads that move every time Confluent moves.

The AI data layer: The missing piece. IBM had models. IBM had agents. IBM had consulting and cloud. It did not have a governed, real-time data fabric capable of feeding those agents with live operational signals. Confluent was exactly that.

The team: Jay Kreps and the engineers who built the category. Unlike the Red Hat playbook, IBM has made clear this is an integration, not an independent subsidiary. The roadmap tilts toward watsonx, IBM Z, and IBM services from day one.

The real play: A fully integrated agentic architecture where events flow through Confluent, trigger agents running on watsonx, and are governed end to end across every environment IBM touches.

What changes: Confluent keeps operating. The product keeps shipping. The only visible shift is IBM's global enterprise relationships sitting behind every future deployment, removing the ceiling on how far the streaming infrastructure can reach.

To understand why IBM acquired Confluent, you need to understand what IBM actually is. Not just a legacy technology company. Not just a consulting firm with a cloud business. A global enterprise infrastructure company whose entire strategic position depends on remaining the default layer through which enterprise AI gets built and deployed. And the gap between where IBM was and where it needed to be was exactly the real-time data infrastructure Confluent had already built.

Confluent did one thing obsessively well: make data streaming usable for enterprises that couldn't afford to get it wrong. While cloud vendors debated data lake architectures and AI companies chased model benchmarks, Confluent was quietly building the operational fabric that Michelin and BMW and Ticketmaster could actually run mission-critical workloads through without downtime, data loss, or governance risk. IBM's AI future required exactly that layer. It just didn't own it yet.

Here's the strategic advantage: data infrastructure compounds over time. The relationships, integrations, and operational dependencies accumulating inside enterprise environments now will define who controls the AI data layer a decade from now. Confluent had already accumulated them at scale.

The real world difference wasn't subtle. Enterprises don't just need AI models. They need AI models running on live, governed, continuously flowing operational data that connects seamlessly to every system they already run. That isn't a chatbot integration. That is a full stack data infrastructure commitment. Confluent wasn't just a streaming company. It was a builder with the technical depth and enterprise trust oriented around exactly that requirement.

And this mattered more as AI shifted from experimental to operational.

IBM saw all of it. The capability gap. The timing gap. The strategic position in a market where whoever owns the real-time data layer owns the infrastructure on which the next generation of enterprise AI gets built.

Confluent wasn't building toward an independent exit. It was building toward combination.

The IBM-Confluent deal exposes a quiet truth about data infrastructure companies: building world-class streaming technology without a global enterprise distribution network behind you is an acquisition target, not a permanently independent business.

Most streaming and data platform companies are fighting over the same ground: better Kafka compatibility, faster throughput, cleaner cloud integrations. It's the most obvious place to compete, which makes it the most crowded. You're racing against the next well-funded open source project, competing for enterprise contracts against cloud hyperscalers who've finally woken up, and hoping your platform stays differentiated before an AWS or a Google decides managed streaming is their next default offering.

Confluent went one level higher: the enterprise trust layer. Governance-first architecture, deep hybrid cloud compatibility, and a customer base that included some of the most operationally demanding companies in the world. They didn't just build streaming technology. They built streaming infrastructure that risk-averse enterprises could actually run mission-critical workloads through, which is a fundamentally harder problem than the technology alone.

That positioning made them acquirable, not invincible. There was no independent path to the distribution scale required to make real-time data infrastructure truly universal across every enterprise environment, every mainframe, every hybrid cloud deployment globally. Confluent had the infrastructure and the customers. It didn't have IBM's 175-country enterprise relationships. That's exactly why the deal got done, before the real-time data layer consolidated around operators who did.

Now Jay Kreps builds inside an organization where IBM's global consulting relationships, hybrid cloud footprint, and century-old enterprise credibility define the strategic context. He built something enterprises trusted by staying focused and moving with discipline. He's walking into a company where distribution scale and institutional depth matter more than technical differentiation alone.

The acquisition sounds like validation. It is. But it's also a ceiling made visible. The best streaming infrastructure in the world still needed the world's largest enterprise AI company to reach its actual potential.

Build the data layer. But know who owns the enterprise relationships.

The IBM-Confluent deal will get filed under enterprise tech M&A. It belongs in a different category.

This isn't a large company absorbing a smaller one to fill a product gap. It's the moment the enterprise AI establishment formally acknowledged that real-time data infrastructure isn't a nice-to-have sitting underneath the AI stack. It is the stack. And the race to own that infrastructure, the streaming layer, the governance controls, the enterprise relationships, is already underway.

The companies that built early and built right are getting absorbed into the institutions that need them. The companies that waited are now building from scratch against a shrinking window and a more expensive competitive landscape.

For the broader enterprise AI industry, the signal is unambiguous. AI agents need live data. AI models need continuously flowing operational context. The gap between where data is generated and where intelligence acts on it is about to get significantly shorter, significantly more governed, and significantly more embedded inside the platforms enterprises already run.

IBM didn't acquire Confluent because real-time AI won. It acquired Confluent because it couldn't afford to find out what losing the data layer looked like.

That's the kind of conviction that closes an $11 billion deal. It's also the kind of signal that rewrites an industry.

The data layer just changed hands. Everything built on top of it changes next.

Here is my interview with Stefan Bader, CEO and co-founder of Cello, a referral automation platform used by Typeform, Miro, and VEED.

If you enjoyed our analysis, we’d very much appreciate you sharing with a friend.

Here are the options I have for us to work together. If any of them are interesting to you - hit me up!

And that’s it from me. See you next week.

What do you think about my bi-weekly Newsletter? Love it | Okay-ish | Stop it