When the Board asks if you can commit for another five years, and you can’t answer — that’s the signal
MongoDB announced that Dev Ittycheria is stepping down as CEO after 11 years at the helm. CJ Desai, formerly of Cloudflare, takes over as the next CEO. The company beat Q3 estimates. The stock popped. Growth reaccelerated this year. The board praised the transition.
But what also happened: A 59-year-old CEO who took MongoDB public in 2017, grew it into a multi-billion dollar database company, and navigated every twist and turn of the cloud era just looked at five more years in the AI Age and said, “I’m not sure I can do that.”
And he’s not alone.
Brian Halligan had a similar moment in 2021. After 15 years building HubSpot from a Cambridge apartment into a $30B+ marketing automation juggernaut, he stepped back to Executive Chairman and handed the CEO role to Yamini Rangan. The catalyst? A snowmobile accident that gave him months to think while she ran the company. When he recovered, he realized: she should just keep running it.
These are successful CEOs making rational decisions about whether they want to do a fundamentally different job for another five years.
The 10+ Year Club is Having a Moment
Dev’s 11-year run isn’t unusual in SaaS — it’s actually becoming the new normal at the top tier. Marc Benioff has been running Salesforce for 26 years. That’s right. Twenty-six. He co-founded it in 1999 and is still going strong, now betting the company on AI agents while personally texting with Dario Amodei about deployment strategies.
Look at the data: While average CEO tenure across all industries has dropped to 7.2 years in 2023 (down 34% from 10.9 years in 2017), a significant subset of public company CEOs are actually serving longer. About 19% of all CEOs stay in position for 10+ years, with median tenure in this “long-serving” group hitting 14 years. In B2B specifically, founder-CEOs and early executives who took companies public often stick around much longer than the averages.
Still in the seat:
- Marc Benioff at Salesforce (co-founded 1999, 26 years and counting)
- Aaron Levie at Box (co-founded 2005, still CEO at age 40 after nearly 20 years)
- Drew Houston at Dropbox (co-founded 2007, still CEO at age 42 after 18 years)
- Reggie Aggarwal at Cvent (co-founded 1999, 26 years and counting – took public twice, sold for $1.65B, went public again, sold again for $4.6B)
Recently stepped down:
- Brian Halligan at HubSpot (co-founded 2006, CEO for 15 years until 2021, now Executive Chairman)
- Frank Slootman at Snowflake (CEO 2019-2024, stepped down Feb 2024 at age 66 after taking it from $200M to $3B)
- Jeff Lawson at Twilio (co-founded 2008, CEO for 16 years until pushed out by shareholder activists in Jan 2024)
- Dustin Moskovitz at Asana (co-founded 2008, CEO for 17 years until stepping down March 2025)
- Daniel Dines at UiPath (co-founded 2005, CEO for 19 years, stepped back Feb 2024 — but returned!)
- Jeff Bezos at Amazon (CEO 1994-2021, 27 years before stepping to Executive Chairman)
But What About the “Young” CEOs?
Here’s the more interesting question: What about Aaron Levie and Drew Houston?
Aaron is 40. He’s been CEO of Box for nearly 20 years — his entire adult life. He dropped out of USC to start the company. Drew is 42. He co-founded Dropbox at 24 and has been CEO for 18 years. Both took their companies public (Box in 2015, Dropbox in 2018). Both are worth more than they will ever need personally. Both are now facing the AI transformation of their core businesses.
How much longer do they want to keep doing this?
These guys are theoretically in their prime. They’re not 60-year-old founders looking at retirement. But they’ve also been doing the same job for nearly two decades. They’ve navigated:
- The transition from consumer to enterprise
- The shift from on-premise to cloud
- The battles with Google, Microsoft, and AWS
- Multiple activist investor campaigns
- The pandemic work-from-home surge
- And now, the radical AI transformation
Aaron turned down a $600M acquisition offer from Citrix in 2011 when he was 28. Drew turned down multiple acquisition offers over the years including from Steve Jobs. They made the right calls. They built real, sustainable businesses.
But at what point does being CEO stop being the thing you’re most excited about?
When you’re 25 and building a rocketship, the CEO job is everything. When you’re 40+ with generational wealth, a track record of success, and maybe kids who are getting older — do you really want to spend the next 10 years reimagining your entire product around AI agents? Or would you rather do something else with your one wild and precious life?
I’m not saying Aaron or Drew are about to step down. But I am saying: if I were on their boards, I’d be having the same conversation Dev’s board had with him. “Can you commit to doing this for another five years?”
Because if the answer is anything other than an immediate “yes,” you probably need to start planning for a transition.
Why the Long Tenures Made Sense (Until Now)
There were good reasons why SaaS CEOs stuck around longer than other industries:
The compounding advantage: SaaS isn’t like other businesses where you can parachute in, cut costs, and show a quarter of improved margins. Everything compounds. Your retention math compounds. Your expansion revenue compounds. Your customer success motion compounds. Having a CEO who understands these mechanics at a cellular level — because they literally built them — was worth its weight in gold.
Founder-market fit on steroids: Many of these long-tenured CEOs weren’t just founders. They were early cloud pioneers who saw the SaaS model when everyone else was still selling perpetual licenses. Dev had been at BladeLogic (which BMC bought for $854M in 2008) before MongoDB. He knew the playbook because he helped write it.
The board stability effect: Once you’ve taken a company public, survived your first few earnings calls, and proven you can execute — boards don’t want to rock the boat. Why would they? The data shows successors to long-serving CEOs typically perform worse than their predecessors. About 69% of successors who replaced a top-performing, long-tenured CEO ended up in the bottom two performance quartiles.
Relationships and Continuity Matter in B2B: You can’t swap CEOs every 3-4 years when your sales cycles are 6-18 months and your customer relationships span decades. MongoDB’s biggest customers have been with them for years. Those relationships were built on personal connections and trust, much of it flowing through Dev himself.
But AI Changes Everything
Here’s what’s different now, and why I think we’re at an inflection point:
The game changed overnight: For a decade, SaaS CEOs got really, really good at one game: land-and-expand, PLG motions, consumption pricing, usage-based revenue. Now, seemingly suddenly, you’re competing with AI-native startups that can build in weeks what took you years. Your roadmap is constantly under threat. Your customers are asking about AI in every call.
The pace is exhausting: I talk to B2B CEOs every week. Every single one is implementing AI agents, rebuilding products around LLMs, competing with AI-first startups, and explaining to boards why they need to invest millions in AI R&D while maintaining margins. The cognitive load is insane. And if you’re 55+, you’ve already done this dance for 10-15 years through multiple cycles.
The energy equation is real: Running a public SaaS company has always been a grind. But now? You’re dealing with AI disruption on top of the usual pressures. Activist investors are circling anyone who shows weakness. The market punishes any sign of slowing growth. And your best engineers are getting recruited by AI labs offering $1M+ packages.
Dev’s quote to CNBC says it all: “Earlier this year, as part of our normal succession planning process, the board asked me about my long-term plans and whether I could commit for another five years as CEO. I thought about it long and hard, I talked to my family, I talked to the board and ultimately I realized I couldn’t make a decision like that.”
Translation: I’m 59 years old. I’ve been doing this for 11 years. I’ve made my money (MongoDB’s market cap is ~$25B). Do I really want to spend ages 59-64 fighting AI-native database startups, reimagining our entire product around vector search and AI workloads, and dealing with quarterly earnings calls where every analyst asks about our AI strategy?
The honest answer was: No.
And that’s completely rational.
But What Happens When the Handoff Doesn’t Work?
Here’s the uncomfortable question: What happens when these CEO transitions fail?
Daniel Dines founder of UiPath gave us the answer.
In February 2024, Daniel stepped back as CEO of UiPath after 19 years building it from a bootstrapped Romanian startup to a $10B+ public company. Rob Enslin, a former Google Cloud exec, took over. Daniel moved to Chief Innovation Officer and Executive Chairman, focusing on AI and product.
Four months later — May 2024 — Daniel was back as CEO. Rob was out.
What happened? Growth slowed. The stock crashed. Daniel told Sequoia: “The last two months were probably some of the toughest — maybe the toughest… Talking to investors after our earnings call, it was really some of the toughest moments. It’s the disappointment of people that put the trust in you.”
Here’s what he said about coming back: “I am returning as the CEO of UiPath after a stint in product and engineering… Coming back as the CEO will help me bring back to UiPath the fanatical customer centricity — maybe some of you still remember the UiPath versions from 2016, 17 and 18 where we’ve done everything for the sake of the customer.”
Translation: The professional CEO couldn’t maintain what the founder had built. The company lost its edge. And when the board panicked, they went back to the founder.
This is going to happen a lot more.
Think about it: You’re replacing CEOs who built category-defining companies in the SaaS 2.0 era with new CEOs who need to rebuild them for the AI Era. The odds of getting it right the first time are low. The data on successor performance is already terrible. Add in the AI transformation complexity, and it gets worse.
We’ll see more boomerangs:
- Founders who step back, watch their successors struggle, and have to return
- Professional CEOs who get fired after 12-18 months when growth doesn’t materialize
- Boards scrambling to find someone — anyone — who can navigate the AI transition
Daniel described it as needing to “go back to our roots” and rebuild “a truly customer-centric organization.” But here’s the hard truth: Most founders won’t want to come back. They left for a reason. UiPath forced Daniel’s hand because the alternative was worse.
So you’ll have companies stuck in no-man’s land: The founder doesn’t want to come back. The successor isn’t working out. And the window to pivot to AI-native is closing fast.
The Harder Question: Is It Getting Easier or Harder?
“This is the most stressed I’ve ever been. And that’s actually a good sign.” — Aaron Levie, CEO Box
Here’s where I’m genuinely uncertain: Is being a public B2B CEO about to get easier or harder?
The “easier” case: AI should make operations more efficient. You can run bigger companies with smaller teams. Customer success can be automated. Pricing becomes clearer. Maybe the job gets easier because the tools get dramatically better.
The “harder” case: You’re competing with AI-native companies that move 10x faster, have better economics, and can iterate daily instead of quarterly. Every product decision has AI implications. Every hire needs to be AI-literate. The pace of change isn’t slowing down — it’s accelerating.
I’m betting on “harder” for at least the next 3-5 years. Which means more transitions. Almost every public B2B CEO I’ve talked to says it’s much harder. Exciting, dramatic, stressful. And harder.
What We Can Learn From Dev
Dev did this right. A few things stand out:
- He was honest with himself: When he couldn’t commit to five more years, he didn’t fake it. He didn’t try to white-knuckle through it. He told the board, “I don’t know if I can do this,” and they started succession planning.
- He picked a strong successor: CJ Desai isn’t a generic hire. He’s someone who’s been in the arena, understands AI deployment at scale (critical for MongoDB’s future), and has proven he can execute. The board didn’t settle.
- He’s staying involved: Dev remains on the board, will advise CJ, and can help with the transition. It’s a thoughtful handoff.
- He timed it well: MongoDB is beating numbers. Atlas (their cloud database) is growing. They’re positioned for AI workloads. He’s leaving from a position of strength, not weakness.
The Bottom Line: Welcome to SaaS 3.0
Dev stepping down isn’t just one CEO transition. It’s a signal that the era of 10-15 year SaaS CEO tenures — which defined the 2010s — is hitting an inflection point.
SaaS 1.0 was Salesforce, NetSuite, Workday. Pioneers who proved you could sell software as a service. Marc Benioff is the last man standing from that era, and he’s reinvented himself as an AI-first CEO.
SaaS 2.0 was the 2011-2022 golden age. Slack, Zoom, Twilio, HubSpot, Asana, Dropbox, Box, MongoDB, Snowflake. These companies perfected the playbook: PLG, consumption pricing, land-and-expand, usage-based revenue. They scaled to billions in ARR. They made their founders and early employees generationally wealthy. They dominated the 2010s and exploded from and 2020-2021.
SaaS 3.0 is what comes next. AI-native companies. Vertical AI agents. Autonomous workflows. Models that can code, design, sell, and support. Companies that can scale to $100M ARR with 20 people instead of 500.
The CEOs who built SaaS 2.0 are looking at SaaS 3.0 and making a rational calculation:
Do I want to spend the next 5 years tearing down what I built and rebuilding it from scratch for an AI-first world? Or do I want to hand the keys to someone who was born into this?
Most are choosing the latter.
The job changed. AI changed it. And a generation of CEOs who built incredible companies in the cloud era are looking at the AI era and realizing: Maybe it’s time to pass the torch.
The next CEO of MongoDB isn’t going to be managing a database company that competes on features and pricing. They’re going to be managing an AI infrastructure company that competes on speed, intelligence, and ecosystem effects. That’s a different job. It requires different skills. Different energy. Different obsessions.
Dev recognized that. He made the right call.
We’ll look back on 2024-2025 as the inflection point. Frank in February 2024. Jeff Lawson in January 2024. Dustin in March 2025. Brian in 2021 after his accident. Dev in November 2025. Daniel forced back in May 2024 when his handoff failed.
These are successful CEOs who won SaaS 2.0 and are stepping aside for the leaders who will win SaaS 3.0.
The real questions are:
- How many of the remaining SaaS 2.0 CEOs are having this same conversation with their boards right now?
- How many CEO transitions will fail in the next 2 years, forcing founders to return (like Daniel at UiPath)?
- Who are the CEOs that will define SaaS 3.0? (Spoiler: They’re probably 28 years old and haven’t gone public yet.)
The era of the 15+ year Classic SaaS CEO is over. What comes next will be messier, with more failed handoffs, more boomerang founders, and more boards realizing that finding someone who can navigate the AI transformation is much harder than they thought.
