Glenn Gow, The Scaling Executive Coach and host of The Scaling Executive Podcast, sits down with Carla Larin, CEO of MaxRTE, to examine how scaling CEOs remove themselves from the decision-making chain — and what has to be true about your team before that becomes possible.
This episode is for CEOs who know they are slowing their company down and want a structural fix, not a mindset pep talk.
Key Takeaways
- CEOs who feel like the bottleneck usually are — and the fix requires two moves in sequence: first, stop requiring decisions to flow through you; second, hire leaders capable of making those decisions without you.
- Choosing between SMB, mid-market, and enterprise is not a toe-in-the-water decision. Enterprise sales cycles run a year or longer, so the commitment to a segment must be real before you make it — and the data that justifies the decision takes years of customer conversations and churn patterns to accumulate.
- Larin’s approach at MaxRTE separates two AI questions most companies conflate: internally, it is a full-throttle productivity play; in the product, it targets validated pain points only — not features fitted to buyer budget categories or procurement buzzwords.
- Building a sales function with no sales background, as Larin did at MaxRTE, requires failing fast, resisting inbound noise from adjacent verticals, and accepting that a repeatable process takes years — not quarters — to develop.
- Proprietary data is the most durable competitive moat in AI-adjacent markets. If your product depends on data connections that took decades to build, no AI solution can replicate that overnight.
How CEOs Remove Themselves as the Bottleneck to Company Growth
When Carla Larin reflects on the periods when she was the bottleneck at MaxRTE, she does not describe a confidence problem. She describes a structural one.
“Not everyone should go through the CEO,” she says. “I think the other part of that too is, again, back to hiring. Everything goes back to people and team.”
The diagnosis Larin applies is two-part. First, the CEO must stop requiring decisions to flow through them — which means actively coaching themselves to hand off authority and accepting that decisions made by others will not always mirror the ones they would have made. Second, the right leaders must be in the seat before that handoff becomes safe. Both conditions must be true. Empowering the wrong people does not solve the bottleneck; it creates a different kind of problem.
At MaxRTE, Larin built the structural scaffolding that makes genuine autonomy possible. Equity compensation tied directly to enterprise value impact means that the people making decisions have a financial reason to make good ones. Quarterly two-way performance reviews — not annual, and explicitly bidirectional — create the feedback loop that keeps leaders developing without requiring the CEO to catch every gap personally. The CEO receives feedback in those sessions too, which signals to the organization that accountability runs in both directions.
“I think a great sign of a leader, at least at MaxRTE, is someone who loves to receive feedback, is always looking to learn and improve,” Larin says. “Sometimes just sharing that feedback more regularly can help people reach that next level in their career development a little sooner.”
The bottleneck problem is, at its core, a hiring problem that has been delayed. CEOs who build companies where every decision still runs through them have usually not yet hired leaders they fully trust — or have not yet given those leaders the authority and incentives to act independently. The structural fix is not delegation. It is building the team and the culture where delegation becomes safe.
How CEOs Choose Between SMB, Mid-Market, and Enterprise — Without Guessing
For a CEO who has not yet committed to a market segment, the temptation is to test all three simultaneously. Larin’s experience at MaxRTE argues against it.
“It’s not a decision that you can dip your toe into and say, let me try,” she says. “Selling an enterprise contract takes a year, at least.”
The time cost of enterprise sales means that a half-committed experiment produces no usable signal. You cannot learn whether enterprise is the right segment from one or two short cycles because the cycle itself is too long and the sample too small.
What produces a real answer is patient accumulation of customer data over time. Larin spent years talking to prospective and current customers, tracking where churn clustered and where satisfaction was highest, and watching the picture clarify gradually. MaxRTE’s go-to-market experiments included both SMB and enterprise motion before the data made the answer obvious.
“We found that our superpower actually was more on the enterprise side,” Larin says, “where absolutely it takes a long time to get everybody in the boat and row in the same direction.”
The reason enterprise became the right fit was not just deal size. It was product fit. MaxRTE’s consultative sale — white-glove onboarding, workflows tailored to each client’s specific pain points, a product that adapts rather than installs — is better matched to buyers who want that kind of engagement. SMB buyers want speed and a standard offering. Enterprise buyers want a partner. MaxRTE’s product is built for the second kind of buyer, which made the segment choice, once Larin was willing to see it, inevitable.
The decision to commit to enterprise also forced clarity downstream. Different segment, different go-to-market motion, different pricing, different customer support model, different hiring profile on the sales team. “Over time, the picture becomes clear,” Larin says. For scaling CEOs who are still running parallel experiments across segments, the data that resolves the question is already accumulating — the commitment to look at it honestly is what is usually missing.
How CEOs Build a Sales Function When They Have No Sales Background
When Larin became CEO of MaxRTE, the company had almost no sales or marketing function. She had no background leading or building a sales team.
“It took a lot of talking to people who were a lot smarter than me, learning from others and iteration, like failing fast,” she says. “There were lots of experiments that we ran quite frankly.”
The first challenge was not building the function — it was resisting the noise. MaxRTE’s product is relevant across multiple healthcare verticals, and inbound interest came from many directions. The temptation to chase every opportunity — what Larin calls “shiny objects” — was constant. Each diversion looked like a revenue opportunity. Each one also delayed the work of building a repeatable process for the segment that actually fit.
It took Larin a few years to develop a sales motion that was genuinely repeatable and scalable. That timeline is not a failure. It is the realistic cost of building from zero in a complex market where the buyer is a hospital system with multiple decision makers and a year-long procurement cycle.
“The power of hiring amazing people is the biggest unlock,” she says. “Just even one good hire can make or break the year from a sales perspective, from a product perspective.”
For CEOs who are building a sales function in a category they did not come from, Larin’s path suggests a specific sequence: talk to more customers than feels necessary, resist the inbound noise until you have a repeatable motion for your core segment, hire one great person before you hire three adequate ones, and accept that the process takes longer than the board will want to wait.
How B2B SaaS CEOs Should Think About AI in Their Product Roadmap
Larin’s approach to AI at MaxRTE separates two questions that most companies conflate: what AI should do internally, and what AI should do in the product.
Internally, the answer is straightforward. “We are all full steam ahead, use every tool available to you to become more productive,” she says. Engineers, support reps, internal operations — every function gets access to every tool that makes them faster. There is no gate on internal AI adoption.
The product decision is more difficult. MaxRTE operates in a market where enterprise buyers now have dedicated AI budget lines, and where “AI-powered” has become a procurement category with its own purchase criteria. The temptation is to build AI features that fit the budget line rather than solve a real problem.
Larin’s framework is the opposite of that. “Our philosophy is really we’re trying to avoid doing that — really, really build for pain points.”
MaxRTE has three products. The company has identified one validated AI use case, within the prior authorizations product, where customer conversations confirmed a real pain point that AI can address. That is the one they are building. The other two products are not getting AI features because the pain point validation is not there yet.
The reason MaxRTE can afford to be selective is the proprietary data position that sits underneath all three products. MaxRTE holds real-time connections to hundreds of health plans, including all 50 state Medicaids — a data set built over decades of contracting that is not publicly available and cannot be replicated quickly. The accuracy of the member ID lookup that drives claim approval depends on that data. A general AI solution does not have access to it.
“That is lucky where we are positioned in a way that it would be very challenging for an AI solution to come up with the unique member ID that you might have for your insurance plan,” Larin says. “Because that needs to be exact, otherwise the claim will be denied.”
For B2B SaaS CEOs assessing AI strategy, MaxRTE’s position illustrates a durable principle: proprietary data that took years to accumulate is a moat that AI cannot cross quickly. The strategic question is not whether to build AI features. It is whether your AI features will be better than what a well-funded competitor can build on public data — and if the answer depends on proprietary data you already hold, that data should be the center of the roadmap, not an afterthought.
What Scaling CEOs Who Stop Being the Bottleneck Do Differently
| Principle | What it means in practice | Named evidence from this interview |
| Autonomy requires structure, not just trust | Empowering leaders to make decisions without the CEO only works when incentives, feedback, and accountability are built into the operating model — not assumed | Larin built equity compensation tied to enterprise value and quarterly two-way reviews at MaxRTE before handing off decision authority — structures she credits with enabling her to exit the daily decision flow as the company scaled |
| Segment commitment must precede segment investment | Choosing between SMB and enterprise is not a test-and-learn decision — the cycle length in enterprise means a half-committed experiment produces no signal | Larin spent years accumulating customer and churn data before committing MaxRTE’s go-to-market resources to the enterprise segment, at which point the fit became, in her words, clear |
| Sales function excellence is a hiring decision before it is a process decision | A repeatable sales motion cannot be built faster than the quality of the people building it | Larin built MaxRTE’s sales function from zero to a repeatable enterprise motion with no prior sales leadership background, crediting individual great hires as the variable that determined whether a given year broke forward or stalled |
| AI product strategy should be anchored in validated pain points, not buyer budget categories | Enterprise buyers have AI budget lines; building to fit those lines rather than to solve real problems produces features customers adopt for a quarter and abandon | MaxRTE targeted one prior authorizations AI use case validated by customer conversations, leaving the other two products unchanged until equivalent validation exists |
| Proprietary data is the moat AI cannot replicate quickly | A data set built over decades of contracts is not a commodity a well-funded competitor can recreate in a product cycle | MaxRTE’s real-time connections to hundreds of health plans, including all 50 state Medicaids, took decades to build and require exact member ID accuracy that no public AI data set can match |
Quotes from This Episode
- “Not everyone should go through the CEO.” — Carla Larin, CEO, MaxRTE
- “The power of hiring amazing people is the biggest unlock.” — Carla Larin, CEO, MaxRTE
- “It’s not a decision that you can dip your toe into and say, let me try.” — Carla Larin, CEO, MaxRTE
- “Our philosophy is really we’re trying to avoid doing that — really, really build for pain points.” — Carla Larin, CEO, MaxRTE
- “That is lucky where we are positioned in a way that it would be very challenging for an AI solution to come up with the unique member ID that you might have for your insurance plan. Because that needs to be exact, otherwise the claim will be denied.” — Carla Larin, CEO, MaxRTE
Frequently Asked Questions
How does a CEO stop being the bottleneck in a scaling company?
A CEO stops being the bottleneck through two structural moves, not one. First, they must stop requiring decisions to flow through them — which means actively handing off decision-making authority and resisting the pull to weigh in on every choice. Second, and more foundationally, they must hire leaders who are genuinely capable of making those decisions independently. Carla Larin, CEO of MaxRTE, ties this to compensation and culture: leaders whose equity is tied to enterprise value have financial alignment with good decisions, and a feedback culture built on regular two-way reviews keeps them developing without constant CEO involvement.
How should a scaling CEO decide between SMB, mid-market, and enterprise go-to-market strategy?
The decision requires patient accumulation of real customer data — not a short experiment. Larin spent years at MaxRTE tracking churn patterns, talking to prospective and current customers, and watching where satisfaction was consistently highest before committing to the enterprise segment. The critical constraint is that enterprise sales cycles run a year or longer, which means a half-committed test produces no usable signal. CEOs who try to straddle segments while gathering data are actually delaying the data that would resolve the question — because the resolution comes from full commitment and observation over time, not from parallel experiments.
How should a B2B SaaS CEO evaluate which AI features to build into their product roadmap?
The filter is validated pain points, not buyer budget categories. Enterprise buyers increasingly have dedicated AI budget lines, which creates pressure to build AI features that fit procurement criteria rather than solve real problems. Larin’s approach at MaxRTE was to identify one specific use case — within the prior authorizations product — where customer conversations confirmed the AI application would address a genuine pain point, and to leave the other two products unchanged until equivalent validation exists. For CEOs with proprietary data, the additional question is whether that data gives your AI features a structural advantage that a competitor building on public data cannot replicate.
CEOs Work with Glenn Gow to Scale Their Companies by Scaling Themselves First
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.
