Tom Chavez has built and sold two data companies — Wrapped to Microsoft and Crux to Salesforce — generating a 17x return for investors across both exits. Glenn Gow, The Scaling Executive Coach and host of The Scaling Executive Podcast — with 25 years as a CEO and 5 years in venture capital — sat down with Chavez, now co-founder of SuperSet, a startup studio that builds data-driven companies from scratch, to extract what most CEOs get wrong about data, go-to-market, and AI, and what the CEOs who actually scale do differently.
The central argument Chavez makes is specific: data is not something to be processed. It is the main event. CEOs who still treat data as a supporting function — a thing the database handles while the business runs — are competing with the wrong mental model. In the age of AI, virtuosity with data is the basis of competition.
This episode is for CEOs of established companies — manufacturing, distribution, professional services, or software — who know AI matters but are unsure how to act on it without losing discipline or wasting resources.
Key Takeaways
- Tom Chavez, who generated a 17x investor return across two data company exits, holds that CEOs treating data as infrastructure rather than as the central competitive asset are standing on an embarrassment of riches they cannot use.
- Chavez holds that the single biggest scaling failure for technical founders is not a product problem — it is a go-to-market problem they never apply systems thinking to.
- Ego — specifically a certainty-to-knowledge ratio that is “hopelessly out of control” — is what prevents CEOs from scaling personally, according to Chavez.
- The right AI question is not “which model should I use?” It is “what problem am I solving, and which parts of it can be automated, augmented, or accelerated?”
- CEOs who are not yet running AI experiments are not being prudent — they are losing the organizational learning their competitors are accumulating right now.
Data Is the Dog Wagging the Tail — Not Along for the Ride
The framing most CEOs carry about data comes from an earlier era. In the Oracle era, the database was infrastructure. The business process — supply chain, finance, customer records — was the central concern. Data existed to serve the process.
Chavez draws a sharp contrast to where we are now:
“Data is now the dog wagging the tail. Before, data was just along for the ride. So making that transition in your head and understanding that capturing, organizing, orchestrating, wrangling, moving, activating data — that’s God’s work. That’s the main thing that we do, particularly again in the age of AI.”
This is not a philosophical point. It has a direct operational consequence. CEOs who have not made this mental shift are competing at a structural disadvantage, because the companies ahead of them are building AI systems on top of data infrastructure their competitors do not have.
The mistake cuts in both directions. Chavez sees CEOs who do not recognize the data they already own as an asset. He also sees CEOs who overestimate how unique their data actually is:
“I do see a number of people who are standing on an embarrassment of riches with data that they could do incredible things with, and they’re not fully aware. I also see people making the opposite mistake where they think their data is more valuable or precious than it actually is.”
The competitive reality Chavez names is this: every company is running on the same frontier AI models. The basis of competition is not which LLM you use. It is how well you feed it.
“We don’t know what’s going on inside them, and we actually don’t care. It’s virtuosity with data that feeds those models. That, from my vantage point, is the central basis of competition now. Leave to the frontier models the science of AI. We and others get to compete in the engineering of AI through data.”
Technical Founders Scale Their Products — Then Stall on Go-to-Market
Chavez reflects on his own early failure honestly. When he founded Wrapped — his first company, which he later sold to Microsoft — he was a freshly minted Stanford PhD in engineering economic systems. He understood the product deeply. He did not treat go-to-market with the same rigor.
The pattern he describes is not unique to him. It is one of the most common traps Glenn Gow sees in the CEOs he coaches: technically brilliant founders who go where they are comfortable rather than where the company actually needs them.
“The product is their macaroni and cheese. So we tell a lot of the founders we work with who are technically amazing — and we want that, it’s a necessary condition — but then they go and start thinking, well, that’s the sufficient condition. I’ve built a kick-ass product, now it’s just gonna sell itself.”
The shift Chavez made — and that he now installs in the founders SuperSet backs — is to treat go-to-market as a systems engineering problem. Not a people problem. Not a culture problem. An engineering problem with inputs, throughput, and measurable outputs.
“The go-to-market of a company is its own systems engineering challenge. And it’s at least as important as whatever systems engineering you do in the product.”
After selling Crux to Salesforce, Chavez watched the company’s go-to-market engine up close — and what he observed was a machine built around enablement. Salesforce grew from a startup to a $1 billion company in six years not on the back of technical breakthroughs but on the back of giving every sales rep the right content, the right insight, and the right asset for every customer conversation, at the right moment. Chavez saw this discipline operating at scale inside the company after the acquisition and brought it back as the standard SuperSet installs in every founder it backs.
The practical implication for CEOs: every minute a sales rep spends hunting for that content is lost money.
“Every minute that you shave off a rep’s day of just hunting and pecking in the dark, looking for the right next assets or insight to bring to a customer conversation — that’s lost time. That’s lost money.”
The CEO role in this is to stop protecting the product and start conducting the orchestra:
“The marketing organization with the right content is like the violins, bring them in at just the right moment. And then you’re doing all of the analytics on your outbounding and acquisition at the top of the funnel and you’re tuning that process every single day.”
CEOs Who Scale Personally Have a Controlled Certainty-to-Knowledge Ratio
When Glenn Gow asked Chavez what separates CEOs who scale personally from those who get stuck, Chavez compressed his answer to one word: ego.
“Ego is the enemy of progress. And I think that the CEOs who fail to scale — they let their ego get in the way. They become too certain. The best people, the best CEOs I’ve ever seen are like curious children learning every day.”
The specific diagnostic Chavez uses is the certainty-to-knowledge ratio. Most CEOs, in his observation, carry a ratio that is out of control — high certainty, and knowledge that has not kept up with how fast the environment is changing.
“I like to talk about the certainty-to-knowledge ratio. I see so many CEOs with a ratio there that’s hopelessly out of control. And especially in the days we live in now, like everything’s changing so fast. All the stuff I thought I knew — some of it’s useful, the patterns are useful, but the lexicon, the artifacts, all the tooling, everything’s different.”
The fix is not humility as a personality trait. It is daily learning as a discipline. What got a CEO to their current position will not get them to the next one. The operating environment, the tools, the competitive dynamics — all of it is shifting faster than any single person’s experience base can keep up with.
Glenn Gow, who spent 25 years as a CEO before founding The Scaling Executive Podcast, draws the same connection from his own experience: “When I was a CEO, I had a coach for 17 years. And that was because each time I met with my coach, I just learned a little something that became compound interest.”
The CEOs who scale personally treat learning as a compounding asset. The ones who stall treat their existing knowledge as sufficient.
CEOs of Non-AI Companies Apply a Problem-First Filter to Every AI Initiative
Chavez’s advice to CEOs of conventional companies — manufacturing, distribution, professional services, software — starts with the same place he starts everything: problem first.
“Don’t let all of the hype overtake you. It’s software. It’s still software on a screen. Stop talking about the model this and the model that, the data pipeline. Put the problem first. If you have clarity on what is a problem that you need to solve — put that first. And then, right after that, ask what pieces of that could be automated or augmented or accelerated through AI.”
The framing Chavez uses — automate, augment, accelerate — is the practical filter for any AI initiative. Chavez applies this filter inside SuperSet to every company the studio backs, keeping the focus on business outcomes rather than technology novelty.
The one failure mode Chavez flags with urgency: CEOs who are watching and waiting. Chavez is not sympathetic to this position.
“You gotta experiment tonight. I urge everybody to take shots. The ones who scare me the most are the ones who are just laying back.”
Chavez’s reasoning is not primarily financial. The ROI on an early AI experiment may be zero from a P&L perspective. The organizational learning is not zero. And neither is the signal it sends internally.
“The learning benefit for the organization — and also the symbolic power of teaching your people, no, no, no, we’re going to take shots. We’re not going to be nervous old ladies and lay back and just watch everybody else. Mindsets precede methods always.”
The CEO who is not experimenting is not just falling behind on the technology, Chavez argues. They are falling behind on the organizational capability to adopt it — a gap that compounds over time.
The Framework: Three Questions CEOs Apply Before Any AI Investment
| Question | What It Forces | What Happens Without It |
| What is the problem I am solving? | Separates technology excitement from business need | CEOs who skip this question invest in AI capabilities with no defined problem to measure against — Chavez sees this repeatedly in companies that arrive at SuperSet AI Advisors after building tools they cannot evaluate |
| Can this problem be automated, augmented, or accelerated? | Connects AI capability to specific operational leverage | Without this filter, Chavez observes CEOs committing to what he calls “big redonkulous dream” builds — ambitions that collapse when a technical thought partner stress-tests the engineering reality |
| Do I have a technical thought partner to reality-check the build? | Ensures ambition matches what is actually buildable at reasonable cost and time | Companies that skip this step build PowerPoint-ready visions; Chavez created Superset AI Advisors specifically to provide this check for CEOs who lack it internally |
Chavez runs this discipline inside SuperSet and through Superset AI Advisors, an AI strategy and development thought-partnering entity he and his team created for companies wrestling with exactly these questions. The principle holds regardless of whether a company works with an outside advisor: problem before model, clarity before investment, reality check before commitment.
Quotes from This Episode
- “The product is their macaroni and cheese. So we tell a lot of the founders we work with who are technically amazing — and we want that, it’s a necessary condition — but then they go and start thinking, well, that’s the sufficient condition. I’ve built a kick-ass product, now it’s just gonna sell itself.” — Tom Chavez, Co-Founder, SuperSet
- “I like to talk about the certainty-to-knowledge ratio. I see so many CEOs with a ratio there that’s hopelessly out of control.” — Tom Chavez, Co-Founder, SuperSet
- “Every minute that you shave off a rep’s day of just hunting and pecking in the dark, looking for the right next assets or insight to bring to a customer conversation — that’s lost time. That’s lost money.” — Tom Chavez, Co-Founder, SuperSet
- “You gotta experiment tonight. I urge everybody to take shots. The ones who scare me the most are the ones who are just laying back.” — Tom Chavez, Co-Founder, SuperSet
- “Don’t let all of the hype overtake you. It’s software. It’s still software on a screen. Stop talking about the model this and the model that, the data pipeline. Put the problem first.” — Tom Chavez, Co-Founder, SuperSet
Frequently Asked Questions
How should a CEO who is not technically trained think about data as a competitive asset?
The shift is not technical — it is conceptual. Tom Chavez, co-founder of SuperSet and builder of two data companies sold to Microsoft and Salesforce for a combined 17x investor return, holds that CEOs must stop treating data as infrastructure their IT team manages and start treating it as the primary source of competitive advantage. Every company running AI systems today is running on the same frontier models — Anthropic, OpenAI, and others. The basis of competition is how well-fed those models are with organized, activated data that competitors do not have access to in the same form.
What is the biggest reason technically strong founders fail to scale their companies?
Chavez identifies the failure as a framing error: technical founders treat product excellence as both a necessary and sufficient condition for growth. It is only necessary. Go-to-market — pipeline generation, sales enablement, funnel analytics — is its own systems engineering challenge, at least as important as the product itself. CEOs who stay in the product because it is where they are comfortable leave the go-to-market motion underbuilt, underanalyzed, and dependent on reps operating without the tools and content they need to close deals efficiently.
How should a CEO start using AI in a company that is not an AI-first business?
Chavez recommends starting with the problem, not the technology. Name a specific operational or customer problem the business faces. Then ask which parts of solving that problem could be automated, augmented, or accelerated through AI. Do not begin with model selection or platform decisions. CEOs who start with the problem stay grounded in measurable outcomes. Those who start with the technology tend to build things they cannot evaluate and cannot scale. Running low-cost experiments with that discipline — tonight, not next quarter — is how organizations build both the skills and the culture to compete in an AI-driven environment.
CEOs Work with Glenn Gow to Scale Their Companies and Careers
Glenn Gow is The Scaling Executive Coach — he coaches ambitious executives into the CEO seat and CEOs into successful exits. With 25 years as a CEO and 5 years in venture capital, Glenn helps leaders scale their companies by scaling themselves first. If this conversation was useful, you can apply for executive coaching with Glenn Gow or apply to be a guest on The Scaling Executive Podcast.
