A Typical Strategic Planning Meeting… At First

A few weeks ago, I was sitting in on a strategic planning session with one of our clients at CEO Coaching International. Strategic planning is one of the key ways we work with CEOs and their Exec teams, and the process follows a fairly structured rhythm. The leadership team completes pre-work that documents the previous quarter’s (or year’s) performance, identifies what worked, calls out the misses, and captures what lessons were learned. The Coach aggregates those inputs and works with the CEO to shape the agenda for the planning offsite.

Most strategy sessions follow a familiar sequence that balances reflection, goal-setting, and strategic debate. The team begins by reviewing financial performance and operational metrics from the previous period, focusing on the drivers that explain the company’s results. From there, the conversation shifts toward defining the next set of quarterly (or annual) goals and the input metrics that will signal whether the company is making progress. The leadership team then turns to broader strategic questions, such as:

• Where should the company focus its energy?

• How fast should it grow?

• How should we allocate capital and headcount?

Eventually, teams get to the most difficult part of planning: determining which initiatives to fund and which to pause or kill. That is where the run really begins!

A new player at the table

I have participated in or led sessions like this for the past 15 years as an executive, CEO, or Board Member, so the rhythm is very familiar. But this meeting had a new player at the table. In this case, the CEO Coach helped the leadership team build what they called a “Strategic Partner” GPT that (who?) was effectively sitting alongside the executives during the conversation. The team had trained it on their operating plan, recent board decks, financial models, and internal operating metrics. In other words, the GPT had access to the same information the leadership team normally relies on to prepare for planning discussions.

The experience reminded me a little of having “Data” from Star Trek sitting in the room, listening to the conversation, and ready to answer questions. Except in this case, the role was not science fiction, and the cost structure was radically different. In many strategic planning sessions, this analytical role is filled by FP&A leaders, strategy teams, or, sometimes, expensive management consultants, with a team of analysts preparing the models behind the scenes. The leadership team debates the question, and then the analysts or consultants run the numbers. The process usually takes hours or days to complete, producing a new deck that requires yet another meeting to discuss. It is the epitome of “soul-sucking time-wasting.”

What was different in this session was that the “analyst” was operating in real time. As the conversation unfolded, the team began using the GPT to summarize historical performance, surface operating trends, and test strategic scenarios DURING the discussion. At one point, the conversation turned to headcount planning, and the CEO asked how three different hiring paths would affect growth, margins, and the company’s Rule of 40 trajectory over the next two years. The GPT produced a clear analysis within seconds, something that normally would have required hours of spreadsheet work, followed by several meetings to review and pressure-test the assumptions.

The realization

What struck me was not simply the speed of the answer, although that was impressive. The deeper realization was that the presence of a “Strategic Partner” GPT at the table begins to change how leadership teams should operate. If an AI system can synthesize operating data instantly, the question becomes how much effort organizations should spend preparing information for leadership meetings in the first place. If strategic scenarios can be modeled during the conversation itself, the implication is that teams should spend far less time producing slides and far more time debating the actual decision. As someone who has spent thousands of hours preparing for these types of planning sessions and asking teams to do the same, I find the potential shift in efficiency, effectiveness, and even job satisfaction difficult to overstate.

The Deeper Question

That experience led me to a deeper question about what AI-powered augmentation might mean for how companies are run. If the cost of analysis is collapsing, CEOs should begin asking how their companies’ operating systems should evolve in response. Traditionally, a company’s operating system governs how information flows, how decisions are made, and how resources are allocated across the organization.

Why does this matter?

Because a wrong decision can impact your growth trajectory and, by extension, your valuation, which in turn affects your ability to secure funding for future growth. For example, research from SaaS Capital shows that the median private SaaS company has been growing roughly 30%/ year. If a company is growing at 30% per year, a single resource-allocation mistake can erase a significant portion of its future growth potential. For example, if you choose to hire too aggressively in sales/marketing, overinvest in an unproven product initiative that does not expand the addressable market, or underfund customer success when retention is the real growth lever, momentum can start to stall. The challenge is not alignment around goals. The challenge is allocating capital and talent in ways that compound growth over time

Why Do We Even Have Operating Systems?

Most CEOs think about operating systems in terms of methodology. The conversation usually focuses on whether the organization should run EOS, OKRs, Make Big Happen, or an internally developed business rhythm (ROB) process. Those choices certainly matter, but over time, I have come to believe that the deeper value of an operating system lies not in the methodology itself. The real value lies in the discipline it creates around tradeoffs and capital allocation. When an operating system works well, it forces leadership teams to decide where resources will be placed and which initiatives will not receive investment.

The challenge becomes clearer as companies scale into the $10M to $200M ARR stage. At that point, most organizations already have alignment on strategy and goals, yet sustaining growth becomes more difficult. The reason is that success at this stage depends less on clarity and more on consistent resource allocation decisions across the organization. Teams propose new initiatives, product roadmaps expand, and hiring plans accelerate across multiple functions. Without a disciplined operating system to force trade-offs, the company slowly drifts toward reactive decision-making rather than deliberate growth.

Four Layers of Operating Systems

In my experience, most leadership operating systems fall into four broad categories, each designed to address a different coordination problem inside a growing company.

The first category: strategic alignment systems ensure that everyone in the organization is literally on the same page regarding company priorities. The original version of this approach was Management by Objectives (aka MBOs), introduced by Peter Drucker. Andy Grove later evolved this construct into the Objectives and Key Results (OKRs) framework, popularized by Google. The key value of alignment systems is that they translate strategy into measurable outcomes that every team can understand. Without this layer, individuals work hard, but it can seem that teams are moving in slightly different directions, or in the worst case, working at odds because teams interpret priorities differently. I wrote an article on how to right-size one of these systems (Salesforce’s V2MOM) for a start-up’s context.

If the first category emphasizes alignment, the second category focuses on operational execution. Systems such as EOS by Gino Wickman and Chief Executive Operating System by Joel Trammell establish a structured cadence of quarterly priorities, weekly leadership meetings, and accountability mechanisms that help organizations translate plans into consistent action. For companies transitioning from the founder-led stage, where the founder can be involved in every decision, to a leadership team model (~$1M-$5M), these systems can feel transformational. The value of execution systems is that they impose discipline on how the organization operates, allowing teams to spend less time reacting to problems and more time executing against clear priorities.

A third category focuses on growth and capital allocation. Frameworks such as Mark Moses’ Make Big Happen System, which we use at CEO Coaching International, force leadership teams to make deliberate investment decisions across people, capital, and strategic initiatives. The central insight behind these systems is that durable growth does not come from pursuing every opportunity but from concentrating resources where the company has the highest probability of compounding advantage. These systems create structured conversations about trade-offs, helping leadership teams decide which initiatives to invest in and which ones to divest or defer.

A fourth category is beginning to emerge as artificial intelligence changes the economics of analysis. As mentioned above, leadership teams have historically relied on FP&A groups, strategy teams, or expensive consulting engagements to synthesize information and test strategic scenarios. That analytical layer sat outside the company’s operating system and typically delivered insights hours or days after a discussion. AI is beginning to collapse the cost of that analytical work, allowing leadership teams to explore scenarios, test assumptions, and synthesize operating data during the conversation itself.

This creates what might be called an intelligence layer inside the leadership operating system. Instead of waiting for analysis to arrive after the meeting, the leadership team can interact with a system that surfaces insights in real time and helps orchestrate the flow of information between humans and machines. The implication is not simply faster decision-making. Over time, it changes how leadership teams structure meetings, evaluate trade-offs, and coordinate strategic decisions. For the past twenty years, most SaaS leadership teams have embraced the first three layers. The fourth layer is just beginning to appear

The Productivity Frontier is Rising

One way to understand the magnitude of this shift is through a simple “growth efficiency” metric: revenue per employee. For most of the SaaS era, the best companies operated within a fairly consistent range. Benchmark studies by SaaS Capital and High Alpha show that typical SaaS companies generate roughly $125,000 to $250,000 in revenue per employee, while strong performers often reach $300,000 to $400,000. A handful of highly efficient public SaaS companies can leverage their scale to achieve $ 500,000+ per employee. Those tiers have traditionally reflected the natural productivity limits of the SaaS operating model (i.e., the “productivity frontier). As I have always said, SaaS is a people business. This is because 70-80% of your costs are people-related. For example, you need engineering teams to build, deploy, and maintain the product; marketing teams to generate demand; sales teams to acquire customers (unless you have a product-led growth [PLG] model); and customer success teams to delight customers, ensure retention, and drive expansion. Thus, even with great execution, the traditional SaaS model requires significant human labor to generate and support revenue.

AI-native start-ups are turning the SaaS operating model on its head, as reflected in their revenue-to-employee efficiency. In fact, some early AI companies are exceeding $1 million in revenue per employee, and a few analyses suggest that top performers may eventually reach $2 million or more as automation expands across product development, customer support, analytics, and go-to-market operations. A couple of examples include:

• Midjourney (~$4.1M revenue per employee)

• Cursor / Anysphere (~$3.3M revenue per employee)

• OpenAI (~$2.8M revenue per employee)

Yes, these are extreme examples, but they are directionally important data points. Over time, these figures will almost certainly normalize as those companies scale and hire more people, but the clear signal is impossible to ignore. If the productive capacity of a software company rises materially, the leadership team must change the operating system that governs how information flows, how resources are allocated, and how decisions are made.

Strategic inflection points

This dynamic illustrates what Andy Grove described as a strategic inflection point. Strategic inflection points occur when the underlying economics of an industry change, and a company must adapt its operating and business models to remain competitive. The classic example was the shift from “on-prem” software deployment to “cloud-delivered” applications in the early 2000s, which ushered in the SaaS revolution. Software companies redesigned their distribution models and, more importantly, their pricing structures to target OPEX rather than CAPEX. The result was a sea change in how the business model and economics worked for vendors and buyers. And there were clear winners and losers from this age.

The recent revolution of generative (and agentic) AI has created another such inflection point. The technology is not simply making companies faster; it is changing expectations for the productivity frontier. What used to be good enough (~$200k/employee) is starting to look downright inefficient, especially as the cost structure is also changing. For example, Sam Jacobs wrote a great article on “agentic math” that captures the new reality of gross margins for AI companies. To net it out: for pure software companies, compute used to be effectively free, and, thus, you would expect a SaaS company’s Gross Margin to be >85%. For AI-native companies, the current reality is GM is 40% because the cost of compute has risen dramatically. The only way to offset this hit to GM is to increase efficiency through leverage, which can be measured by revenue/ employee. Given this new reality, every SaaS CEO must reevaluate how they run their businesses and whether they need to rethink their operating systems to align with the new market’s economic realities.

The CEO’s role in an intelligence-augmented company

The implication for CEOs is that their role needs to shift from managing information to orchestrating intelligence. As companies grow, CEOs naturally spend less time executing work themselves and more time designing the systems through which work happens. Leadership operating systems such as OKRs, EOS, or Make Big Happen were created to address that scaling challenge by helping CEOs align teams, coordinate decisions, and execute consistently as the company grows. These systems assumed that collecting data was difficult, analysis required time, and leadership teams needed significant preparation before strategic discussions could occur. As a result, companies built elaborate processes to gather information from multiple systems, assemble presentations, and distribute analysis through structured meetings.

Artificial intelligence changes the economics of that model by collapsing the cost and time required to synthesize information. CEOs are no longer primarily responsible for ensuring that information flows upward through the organization. Instead, they must design the environment in which human judgment and machine intelligence interact productively in near real time. The upside is that, in this new reality, meetings can evolve from information briefings into decision forums where leaders debate trade-offs rather than review slides. What it means to prepare for a planning session also changes. Now, rather than focusing on assembling reports, leaders can focus on framing the strategic questions that matter most.

In this model, human expertise does not disappear, but its role changes. Analysts and operators remain critical participants in the decision process, yet their contribution increasingly centers on interpretation and judgment rather than data preparation. AI systems surface insights, model scenarios, and synthesize information immediately, while leadership teams focus on evaluating implications and choosing a course of action. The CEO’s role, then, is to orchestrate the interaction among people, data, and intelligent systems (i.e., agents) so that the organization can move faster while maintaining execution discipline aligned with strategic priorities.

From Management Systems to Orchestration Systems

Seen in this light, AI introduces a new layer into the company’s architecture. The operating systems that defined the SaaS era were designed primarily to coordinate human activity. Alignment frameworks ensured that teams pursued common priorities, execution systems imposed operational discipline, and capital allocation frameworks forced leaders to make trade-offs about where to invest resources. These three layers solved the core coordination problems of scaling organizations during a period when analysis was expensive, and information moved slowly through the company.

The emerging AI layer changes how those systems function by embedding intelligence directly into the leadership workflow. Instead of waiting for analysis to arrive after a meeting, leadership teams can now explore scenarios, test assumptions, and synthesize operating data during the conversation itself. Analytical copilots and specialized agents become participants in the decision process, surfacing insights and helping leaders evaluate strategic alternatives in real time. The result is not simply faster decisions but a different operating rhythm in which insight emerges continuously rather than episodically through prepared reports.

Just as cloud infrastructure transformed how software was built and distributed, intelligence systems are beginning to transform how leadership teams process information and coordinate decisions. Companies that recognize this shift will start to redesign their operating systems to deliberately incorporate these capabilities TODAY! Those that do not will gradually accumulate “operations debt” (e.g., more meetings, more reporting, and more coordination overhead in an attempt to manage increasing complexity). Over time, the difference between these approaches will compound. CEOs who learn to orchestrate intelligence allocate capital more effectively, adapt strategy more quickly, and sustain momentum as they scale. And they will leave their competition behind.

A few months ago, I was sitting in a board meeting at a growth-stage SaaS company when one of the directors asked about the company’s AI strategy. . . again. From what other CEOs tell me, this has been the recurring question in nearly every board meeting over the past 18 to 24 months. It is no longer a thought experiment; it is an expectation. Many VCs are shifting their entire investment thesis toward funding AI-first companies. And most are pushing their current “non-AI-first” portfolio companies to become AI-first as quickly as possible

In this case, the CEO answered well. He walked through how the company was using AI to augment people’s capabilities. For example, Engineering was adopting the latest tools to write higher-quality code faster; Customer Success was experimenting with summarization and internal knowledge assistants; and Marketing was leveraging large language models (LLMs) to produce personalized content at scale. At this meeting, I was struck by the fact that all of these activities fall in the realm of “AI augmentation.” I.e., the company was using AI to accelerate existing workflows and improve output quality. But it begs the question: “Is faster output the same thing as an AI-first strategy?”

What no one in that room asked were questions such as:

• “If AI meaningfully reduces the cost of intelligence, should we still be organizing the company around static dashboards and monthly reporting cycles?”

• “Should forecasting still be a quarterly ritual, or should it become a continuous ‘learning system’ that reconciles sales activity, product usage, and billing data in real time?

Fast forward to this past month: Mark Moses, CEO of CEO Coaching International, challenged all partners to be deliberate in helping clients think about their AI strategy. We are now using four questions to frame these conversations:

1. How is AI impacting your industry?

2. How is AI impacting your business?

3. What job can an agentic AI solution reinvent in the next six months?

4. How is AI impacting your job as CEO?

The first two questions are increasingly straightforward, though still important to clarify. McKinsey’s 2025 State of AI report shows that 65% of organizations now report regular use of generative AI in at least one function, nearly double the figure from the year prior. Adoption is broad and accelerating. Yet nearly two-thirds of respondents say their organizations have not begun scaling AI across the enterprise. In other words, experimentation is common; however, companies are still struggling with integration.

The third question, around agentic AI, is generating a lot of interest and, to be honest, a fair amount of confusion today. People are asking questions such as:

• What does agentic even mean?

• What roles can an agent play?

• How do I think about building out functions that have human+agent teammates?

According to McKinsey, roughly 62% of organizations are at least experimenting with AI agents. There is far less clarity around what to redesign and what to leave alone.

The fourth question is the one I find most interesting for CEOs to think about more deeply. At the consumer level, Pew Research reports that roughly half of U.S. adults have used a LLM (e.g., ChatGPT) at least once. But regular, structured use is far lower. Which brings us back to the boardroom. When directors ask about your AI strategy, they are partly asking about product features and cost leverage. But they are also asking something more fundamental: “Have you, as the CEO, changed how you think and behave?”

This essay is about that shift. Because the way you adopt AI as an individual will shape the ceiling of how your company adopts it.

My own journey through augmentation

When ChatGPT came to market, like most people, I started experimenting with it. I used it to draft emails, help with research, and summarize findings for articles. I spent time playing with prompts to make “chats” more efficient. In May of last year, I wrote an essay that summarized where I was at that point. At that time, I was intrigued by the ideas of cognition and authorship. Specifically, how do you know what is “your idea” versus what was “augmented intelligence”?

At Maxio, I was encouraging the organization to rethink workflows. At the same time, we had more than $20 billion in billing and invoicing data across our customer base. The strategic question was not whether AI could write better emails. It was whether that data, properly modeled, could reposition us as an insight engine rather than just a system of record. We wanted to be AI-first, but we also had 2000+ customers we were currently supporting and needed to keep improving the current platform, addressing tech debt, etc.

What I did not fully appreciate at the time was that my own maturity with AI was still largely at the augmentation stage. I was using it to make myself more productive. I was not yet using it to redesign how I thought about systems.

What’s the difference?

Augmentation thinking asks, “How can AI help me do this task faster or better?” Systems thinking asks, “If the cost of intelligence collapses, how should I redesign the workflow entirely?” The first improves efficiency; The second changes architecture. And this “systems thinking” is what I needed to embrace to help lead the company forward. I left Maxio in March 2025, but have continued my own AI journey.

A Personal AI Maturity Table

To make this concrete, I started thinking about AI adoption not just as a company journey, but as a personal one. I believed that, before you can architect AI into an organization, you have to understand how it shows up in your own cognitive workflow (i.e. how you think). That led me to sketch what I am calling a Personal AI Maturity Matrix. There are three primary dimensions that determine where you sit on the Matrix.

• Reach: How consistently are you using AI in your own work? Is it occasional experimentation, or is it part of your daily rhythm?

• Depth: How embedded is AI in your recurring cognitive workflows, such as strategy formulation, hiring decisions, capital allocation, and board preparation? Is it a peripheral tool, or is it shaping how you structure problems?

• Delegation: How much intellectual authority are you granting it? Is it editing your words, stress testing your logic, modeling scenarios, synthesizing data, or shaping first-draft frameworks that you then refine?

When you look at these three dimensions together, patterns start to emerge. I then built out five stages that reflect different ways of thinking and behaving along the AI journey. To ground this in something more than anecdote, I asked ChatGPT to synthesize adoption data from Pew, McKinsey, and the Stanford HAI AI Index to approximate how these levels might distribute across the U.S. population. To be clear, these are directional estimates, not census-level precision, but they align with what public research shows. Currently, there is high awareness, relatively lower integration, & very low systems-level redesign.

Interestingly, if you step back, the pattern roughly maps to Geoff Moore’s technology adoption curve from Crossing the Chasm. Today, most people fall into the late majority or laggard categories in their use of AI. Very few people are operating at the frontier.

Personal AI Maturity Table

• Level 0 | Avoidance
  Estimated U.S. adult population: 40 to 50 percent
  This is where most people still are. They are aware of AI. They have opinions about it. They may even feel threatened by it. But they are not using it in any meaningful way. Reach is effectively zero. Depth and delegation are nonexistent. From Moore’s lens, these are the laggards, i.e., those who do not see a compelling reason to change their behavior because they don’t know how to start, don’t think there is real value, or are just afraid.

• Level 1 | Curiosity
  Estimated population: 20 to 25 percent
  These individuals have experimented with AI, trying ChatGPT or Claude once or twice. They may have asked them to draft a vacation itinerary or summarize an article. But their usage is sporadic. At this level, reach is inconsistent, and depth is shallow. Delegation is limited to simple tasks like drafting or summarizing. This group is interested, but not yet committed. In Moore’s framework, this aligns loosely with the late majority who are observing but not yet rethinking their behavior.

• Level 2 | Utility
  Estimated population: 15 to 20 percent
  At this level, people start to use AI regularly. They use it for drafting emails, preparing outlines, summarizing research, or debugging code. Reach is high at the individual level. Depth, however, remains task-based rather than workflow-based. Delegation extends to research and light analysis, but their thinking about AI and its impact on their daily lives remains largely unchanged. This is where I was in May of last year. I was faster. I was more efficient. But I had not yet redesigned how I approached problems.

• Level 3 | Integrated Workflows
  Estimated population: 5 to 8 percent
  At this level, people are embedding into recurring workflows. It is part of how they prepare for board meetings, how they frame strategic tradeoffs, how they evaluate hiring decisions, or how they structure complex writing. On a personal level, it could include how they think about training and nutrition on a daily basis. Reach is consistent. Depth increases because AI is now shaping recurring processes rather than isolated tasks. Delegation expands to include scenario modeling, counterargument generation, and synthesis across multiple inputs. This is where I believe I sit today.

• Level 4 | Systems Thinker
  Estimated population: 1 to 3 percent
  This is the pivot. At Level 4, AI begins to influence how you design systems, not just how you execute tasks. You start asking: if the “cost” of intelligence has collapsed, what workflows should I redesign? Delegation is bounded but strategic. You define explicit roles for AI in planning, review, and analysis. This is where augmentation turns into architecture. This is where I am intentionally trying to move over the next several months.

• Level 5 | Orchestrator
  Estimated population: <1 percent
  At this level, an individual intentionally designs systems that integrate AI across domains, with governance, metrics, and defined human override. AI is not just a tool; it is an integrated layer in how value is created. Ownership and accountability are explicit. This is rare territory today, but it will not remain so for long.

There are three macro observations from this exercise:

1. If these estimates feel skewed toward the lower levels, that is consistent with the research. McKinsey data shows widespread experimentation, but far fewer organizations report material workflow redesign or measurable business impact. Personal maturity mirrors that pattern.

2. There is no shame in being at Level 1 or Level 2. That is where most people are. This technology cycle is still early, and experimentation is rational.

3. The only way to move up the table is through practice. AI maturity is not theoretical. It is behavioral. You have to use it, reflect on it, and intentionally redesign your own workflows before you can credibly redesign your company’s.

And that last point is where this stops being an intellectual exercise and starts becoming a leadership issue. Because in a growth-stage SaaS company, the CEO’s cognitive ceiling often becomes the company’s strategic ceiling.

The Impact of the CEO’s Personal Adoption of AI

What has become increasingly clear to me is that for CEOs in particular, learning how to use AI well is no longer optional. Not because it is trendy, but because it is becoming foundational to how decisions are made and systems are designed.

If you, as the CEO, are operating primarily at Level 2, your company will likely adopt AI at Level 2. You will encourage productivity gains. You will sponsor pilots. You will celebrate efficiency wins. But you will unconsciously limit the scope of what the company can become, because you are still thinking in terms of augmentation.

If, on the other hand, you are operating at Level 4, you begin asking a different class of questions. Instead of asking, “How can we use AI to do this task better?” you begin asking, “If the cost of intelligence has collapsed, how should we redesign this workflow entirely?” You begin asking architectural questions such as:

• Revenue Model: If AI improves our ability to predict customer behavior, should our pricing, packaging, or expansion model evolve to reflect that new insight advantage?

• Capital Allocation: If intelligence becomes cheaper and faster, how should that change where we deploy human capital versus automation across product, go-to-market, and customer success?

• Org Design: If AI meaningfully reduces the cost of analysis and coordination, which roles in our organization are still designed for information routing rather than decision-making?

That shift in your personal experience allows you to reframe both the threat and the opportunity that AI presents for how your company competes and how you build it.

Where do you sit?

Do I expect CEOs to print out the Personal AI Maturity Table and bring it into their next board meeting? No.

That is not the point. The point of this essay is to encourage folks to ask: “what is my actual relationship to AI?" and “how can I get better?” If you want to diagnose yourself, do not ask how often you use it as “frequency” is not the indicator of success. The real indicator of success is if your thinking has changed. So ask yourself, based on your use of AI today:

• Has your decision-making speed meaningfully improved?

• Are you seeing second-order consequences more clearly because you are modeling scenarios and stress-testing assumptions more rigorously?

• Have you redesigned at least one major leadership workflow because AI changed the economics of intelligence?

• Are you spending more time on architecture and less time producing artifacts?

The answers to these questions can help reveal whether you are still augmenting output or have begun to operate at the systems level. In my own journey over the past ten months, that has been the real inflection. I interact with AI constantly, but the meaningful shift was not volume. It was perspective. I am now using it to structure problems differently, to surface blind spots earlier, and to redesign workflows I once accepted as fixed.

The unusual opportunity in the AI tech revolution wave is that the barrier to raising your own maturity is low. You do not need a transformation budget to evolve from augmentation to systems thinking. You need to be intentional in your practice and disciplined in your reflection. In my own journey over the past ten months, that shift has been the real inflection. As I wrote about in the previous essay, I am talking to some form of AI all day/every day. But now I am using it to literally think differently

So, when the board asks for your AI strategy, they are asking about product and margin. The more important question is whether you have one for yourself.

In mid 2024, I was sitting with Brian Austgen, our head of Demand at Maxio, and we were discussing a troubling trend in our pipeline analytics. Paid search was still generating leads at a reasonable cost, but our organic traffic had declined significantly. He concluded that it was due to the rise of AI summaries (what became known as AEO), powered by LLMs. While we had pretty sophisticated attribution models in place, this was a new dynamic, which led to a conversation about shifting spend from demand to brand, and then to a more existential question: what does ‘brand spend’ mean in an AI world?

I was reminded of this conversation while reading the 2026 Brand vs Demand Benchmarks Report from Benchmarkit.ai. This report surveyed 168 CMOs of B2B tech companies and found that, on average, they spend ~70% of their dollars on demand gen and only ~25% on brand building. Yet when asked what the ideal allocation should be, these marketers said it should be ~50% demand and ~40% brand.

Why the disconnect?

You fund what you can defend. 73% of them agreed that brand spend makes demand generation more efficient over time; however, only 28% say they could directly tie brand investment to pipeline. Thus, in a digital-dominant marketing landscape, demand gen has been more defendable, at least until the new world of AI marketing we are all facing today.

The Rise of Measurability

When I joined Avenue A in 2000, digital advertising was still a small fraction of global ad spend. Television was still where most of the dollars went, and, if you can believe it, print still mattered. At that time, internet display advertising was funded from experimental dollars, and people were trying to figure out if it really worked. Online, creativity mattered a lot less than on TV. I remember someone joked that to improve a display ad’s performance, you should “Make it pink and make it blink.” The key was measurability. We could track impressions, clicks, conversions, and cost per acquisition. We believed we had solved the problem that John Wanamaker, the famous department store owner and early marketer, decried in the 19th century: “Half the money I spend on advertising is wasted; the trouble is I don’t know which half.”

Wanamaker was talking about print. He could see his sales increase, but he could not attribute it to specific ads. Digital advertising promised to fix this, and we went on a tear for 26 years. Internet advertising grew from low single digits of global ad spend in the early 2000s to become the dominant channel, eventually surpassing television in most markets. The rise was not just about targeting. It was about accountability.

That shift also changed who won the budget debate. In the television era, the creative director held influence. Think of the show Mad Men, which captures the glory days of TV advertising. Brand campaigns shaped preference and identity. In the digital era, the analyst (aka the performance marketer) became the hero. Clicks, leads, and ROAS drove allocation decisions. Because bottom-of-funnel actions were easier to measure, digital marketers increasingly optimized for them. Everyone still believed that Brand programs mattered. However, they just became harder to defend. And that has been the tension for 20 years+.

Search Engine Land recently reported in Nov 2025 that when Google AI Overviews appear, organic click-through rates (CTR) can fall by more than 60 % and paid CTRs drop as much as 68% for some searches. To be fair, “organic” CTRs declines are most pronounced on informational queries, where AI Overviews appear. The data shows that the impact on “commercial” queries is more mixed. In any case, AI-generated summaries reduce the need to click through to underlying sites. At the same time, multiple industry analyses show the continued rise of zero-click searches, where users find answers without ever visiting a website. To that point, in an Oct 2025 article, “New Front Door to the Internet: Winning in the Age of AI,” McKinsey noted that ~50% of consumers are already using AI-powered search tools as part of their research and decision making process. I am a data point of one, but I now do 90% of my “searching” with Google Answers and ChatGPT. I never click on the paid links. If I want to buy a product, I go directly to Amazon, REI, or another retailer.

This is a fundamental disruption in how CEOs and CMOs need to think about distribution. One way to frame the new way of thinking could be:

• Search Engine Optimization (SEO) is about ranking.

• Answer Engine Optimization (AEO) is about answering.

• Generative Engine Optimization (GEO) is about being included.

Traditionally, for SEO, you optimized your website content to ensure high rankings on Google. This led marketing orgs to invest significant time and money into content production. I have talked to many CMOs who have teams of people focused on writing SEO optimized articles, embedding keywords, and ensuring the HTML code was correct. In the classic “ten blue links” world, visibility meant ensuring you were on page one of the search results. An entire industry of Search Engine Marketing Agencies grew up to help companies optimize their SEM and SEO spend. This took years.

With the rise of AEO, teams needed to start to create content so that AI systems could extract and present it directly as an answer. But this change has happened in less than 3 years, and few people really know how to do this today. Brian would say that the best marketers have figured out that optimizing AEO is similar to SEO in terms of trying to capture “long-tail” demand. The two key differences is that you have a more constrained budget, it is more complicated and Google is not making it easy to understand the rules of the game.

GEO is even harder because now you have multiple engines, each with its own algorithm. The objective for GEO optimization is to be cited or summarized inside an LLM-generated response. The challenge is that these engines are taking in much more input to determine your “authority.” Thus, to be successful, you have to optimize your brand’s presence across:

• Earned media

• Reviews

• Forums

• Wikipedia

• Industry publications

• Analyst reports

In the AI-mediated world, visibility means inclusion in the answer set. Here is a quick check to see if your company is doing this well.

1. Go to ChatGPT and prompt it to evaluate your market, the key competitors, the relevant trends, and how it is evolving.

2. See if your brand and/or products are listed and if your content is cited.

If you aren’t, look at who is. They are doing it right.

The underlying business dynamic is changing.

Discovery is now increasingly mediated by models that synthesize information before a buyer ever reaches your site. Thus, “influence” has moved upstream of the click. As a result, attribution models built for the SEO era are now less reliable because fewer clicks occur.

How should CEOs and CMOs think about this new reality?

In the digital era, you have a lot more direct control over your demand gen spend and results. You would increase paid spend and watch the leads rise. Yes, at some point, you would hit the efficient frontier, but you often had a lot of room to spare before things became inefficient. This is why 55% of marketers in the 2026 Brand versus Demand Benchmark Report said they would protect demand, compared with only 11% who said they would protect their brand spend when faced with cuts.

At the same time, if you continue to optimize solely for last-click performance metrics while organic discovery shifts to AI summaries, you will see traffic decline, panic, and over-rotate toward spending more on paid search. The impact will be rising CAC, slowing pipeline velocity, and eventually pressure on growth multiples because efficiency erodes. I just wrote an article about concerning trends in CAC Payback. This dynamic is one of the causes.

In the AI era, you must treat authority and clarity as strategic assets. Your brand must be structured in a way that models can recognize, trust, and include. This requires even more cross-functional alignment across marketing, product marketing, content, PR, and even product documentation. To that point, user-generated content platforms e.g., Reddit) are increasingly influencing answer sets. How many of you have a Reddit activation/content strategy?

Brand versus Demand in the AI world

The question that will make budget conversations more complex over the next two years is this: when you fund AEO and GEO initiatives, are those brand dollars or demand dollars? If your company appears in an AI-generated shortlist when a buyer asks for the best platforms in your category, you are clearly influencing “in-market” demand. Yet, the reason you appear is the accumulated “authority” you have built over time. This new reality blurs the traditional split. I don’t think this means AI advertising is a new category, as there are still only two core economic objectives behind marketing: (a) increase preference and (b) capture intent.

What is actually happening is a collapsing of the artificial separation between brand and demand we saw in the “Search”-dominant era. In the digital-dominant era of 2020-2026, brand programs largely drove awareness while demand programs captured intent once it appeared. You could treat them as sequential and somewhat distinct. In an AI-mediated environment, that separation weakens. Brands now need to focus on building authority so that you are included in the answer set. Inclusion in the answer set directly shapes in-market demand capture. In other words, brand authority becomes a prerequisite for demand capture. That is the shift.

What Changes is the Mechanism

I would argue that AI optimization is not a new objective; it is a new distribution layer. Just as television, search, and social each reconfigured how preference and intent were connected, AI now sits between buyer and brand as the next distribution layer. The economic objectives remain the same: increase preference and capture intent, but the mechanism that links them has changed in a way that forces CEOs to rethink how the dollars are allocated.

As a CEO or COM, you should not wait until your traffic collapses before adjusting. As I mentioned in the other article, this is where second-order metrics can help, e.g., rising CAC despite stable spend. Other things you could look at include shorter website sessions, as answers are consumed off-site. You could poll your sales teams to see how many people are arriving with a short list of competitors and evaluation criteria, obviously generated by AI tools.

What does not change is the same question that Wanamaker had: “Where do we spend to maximize sales?” The answer has always been some combination of building preference and capturing intent.

The new reality facing today’s CEOs and CMOs is that if AI reduces attributable clicks, then your board is going to pressure you to spend less money on marketing. The CEOs who win this board battle will recognize that the power center has shifted again, from the creative director to the analyst to the agent, and they will have to completely rethink their marketing strategy, team composition (human+agents), and tactics before the agents start deciding which brands matter.

Managing the Cost of Growth While Building the Next Engine

Over the past couple of weeks, I have been thinking about what is required to scale a company from $100M to $500M. One part of the catalyst has been my conversations with CEOs who are operating squarely in this zone in my new role as a CEO coach. But part of the catalyst has been my reflection on my own experience in this zone after writing this article on “Strategic Inflection” points.

When I ran Rocket Fuel, we were a ~$400M public company. The issues we wrestled with were very different from the ones I faced earlier in my career. And when we sold Percolate, then a ~$30M company, to Seismic, I had a front-row seat to how a scaled company thinks about cadence, capital allocation, and managing multiple go-to-market motions at once across products, segments, and regions.

So I have been thinking about what lessons CEOs should learn and apply as they face this inflection point. It is actually hard to define what “best in class” companies and CEOs do at this stage, given the few data points. According to Crunchbase, of the 4600 public software companies, only 46 fall into the $100-$500M range. However, the good folks at Bessemer have done the hard work of looking across private and public companies to develop a set of benchmarks, which they aggregated in their 2024 report: “From Start-up to Centaurs: The Founders’ Roadmap to $100M.”

At first glance, companies between $100M and $500M look relatively healthy. And many of the headline metrics do not change dramatically as companies cross from $50M-$100M into the $100M+ range. Gross margins remain consistent. Net revenue retention tightens slightly but stays strong. One metric, however, changes the CEO’s job: CAC payback. In the $50M–$100M range, the average CAC payback is roughly 21 months. At $100M+, it is closer to 30 months. This is significant.

But what does this mean?

Effectively, it signals that acquiring the next dollar of ARR requires more capital and more time. If margins and retention remain intact while CAC payback extends, the issue is most likely not product-market fit: It is capital efficiency. At this size, capital gets tied up longer, and it is harder to pivot. The business can continue to grow, but growth becomes more expensive to finance. This essay focuses on what CEOs should consider as they face the “Efficient Scale” inflection point.

$100M to $500M | Efficient Scale

The transition from $100M to $500M is not primarily about adding complexity. Most companies have already layered in new products, segments, or regions by this stage. Scaling efficiently means managing the rising cost of growth while deliberately building the next “growth engine.” When CAC payback lengthens, the CEO can’t afford to fund multiple initiatives haphazardly. Capital allocation becomes tighter; decision rights matter more; tradeoffs can no longer be implicit. Perhaps most importantly, the executive team can no longer just solve for their function. At this stage, the CEO’s job shifts from launching more initiatives to underwriting fewer, larger bets with clear ownership and defined outcomes.

What success looks like from $100M to $500M

The research shows that companies that successfully navigate the $100M to $500M transition embrace three best practices.

1. Protect and Extend the Core Before You Invest in Adjacencies

Chris Zook, in his book “Profit from the Core,” states that enduring growth comes from expanding outward from a position of strength (aka “your core”), not from abandoning your core in search of the next cool thing. This is because at at $100M, your core growth engine is working. You most likely have a defined ICP, a repeatable GTM motion, and some form of structural advantage. But because the company feels “established,” there is a temptation to treat the core as solved and shift attention and dollars to the next thing. By contrast, the most successful CEOs at this stage go all-in on their core. They spend the time to define the core explicitly. They quantify their differentiation. And they invest to deepen the core’s advantage by building moats and walls.

At the same time, the most successful CEOs don’t have blinders on. They are deliberate about finding the next $100M business. At this stage, then, the strategic question becomes: “What adjacent potential growth engine leverages our core capabilities and can realistically become $100M on its own?”

This requires a different way of thinking about investment. Specifically, it requires dedicated ownership, milestone-based funding, and explicit “kill” criteria that people will actually use to kill a project if the engine starts to stall.

The pressure on the CEO and executive team intensifies as the risks of making a mistake are high, and you probably can’t make many mistakes and keep your job. The risk then becomes incrementalism. You explore segments, products, and geographies without forcing any of them to graduate into a scaled engine. The organization feels innovative, but you start to see CAC payback stretch. The result is not a collapse. It is drift.

2. Design for Coherence as Complexity Rises

By $200M, the issue is usually one of alignment. One of my favorite books is Patrick Lencioni’s “The 5 Dysfunctions of a Team.” He has lots of useful concepts, but the one that resonates most at this stage is his “first team” concept. In essence, the executive team must operate as leaders of the business and not as a collection of functional leaders. At this scale, it becomes easier to optimize for your specific function’s needs amid so much complexity. For example, sales pushes for more flexibility to close deals so they can hit the number. Product expands its roadmap commitments because it now has to support multiple customer segments across multiple regions. Finance tightens cost controls to improve EBITDA in accordance with investors (most likely PE) and/or public company shareholders’ desire for the company to generate (and distribute) cash. Each executive’s decision is individually rational. Collectively, they can create a coordination tax. And the death knell is leadership meetings that report on activity versus focusing on the single most important objective and working through hard trade-offs to realize the company’s ambition.

I first experienced this when I was at RocketFuel, working with very senior executives. We were executing a major turnaround, and I needed each executive to lean in, figure out what was best for the business, and make appropriate sacrifices. Most importantly, they needed to be able to represent these tradeoffs to their organizations.

This is why, at this stage, the CEO’s job requires them to see organizational design as a strategic lever. What does this mean? A couple of examples include:

• Clarifying decision rights and the decision-making process.

• Making enterprise tradeoffs explicit in the All Hands.

• Aligning incentives around shared outcomes, specifically an executive scorecard.

• Protecting a small number of cross-functional priorities.

If you do not deliberately design for coherence, the organization’s complexity will start to drag down your growth potential. Thus, the cost of growth, as represented by CAC payback, increases not because the market disappeared, but because organizational “friction” is mucking up the system.

3. Make Capital Allocation Explicit and Earn the Right to Invest

Warren Buffett has long argued that capital allocation is the CEO’s most important job. In early-stage SaaS companies, this can feel abstract because if growth is solid, capital should be relatively abundant and investors relatively forgiving. At $100M, the calculus changes. Every dollar invested in a new segment, product, geography, or AI initiative has an opportunity cost.

If your CAC payback is now 30+ months, you are, in effect, underwriting a longer risk window. This is where McKinsey’s Three Horizons framework becomes informative.

Horizon 1 is about protecting and extending the core (point #1 above). For most companies between $100M and $500M, Horizon 1 products account for the bulk of revenue over the next 12 to 24 months. They fund everything else, so you want to make sure youare devoting enough time and energy to these products.

The challenge at this stage is either complacency or starvation. Complacency shows up as incremental roadmap work, while competitors innovate faster than you and start winning more deals on the strength/breadth of their product. Starvation shows up when leadership diverts too much attention and capital into adjacencies, assuming the core is stable enough.

The CEO’s question for Horizon 1 is straightforward: “What are the two biggest threats to our core engine, and are we investing enough to defend and deepen it?” If CAC payback is starting to extend, you cannot afford a weak core. Embracing Horizon 1 discipline earns you the right to invest in the next two horizons.

Horizon 2 is about scaling adjacencies that can become material. As CEO, you need to define a small number of “close” adjacencies that leverage your core strengths. That could be a new product in the same ICP, a new segment with the same capabilities, or a tuck-in acquisition that accelerates something you were already building. The key is to be intentional. As CEO, you are underwriting the possibility that this productor business can become meaningful within 2 to 5 years.

The CEO’s question for Horizon 2 is: “If this works, how big can it be, and what must be true for it to get there?” If you cannot answer that clearly, you are funding activity, not building another real “growth engine.”

The biggest potential mistake a CEO can make when thinking about Horizon 1 versus Horizon 2 is running the whole company like Horizon 1, then telling the board you have a growth strategy because you are “exploring” new markets or “doing AI.” The trick is to invest in Horizon 2 initiatives as if they were real businesses, with ownership, milestones, and kill criteria, while still protecting Horizon 1 discipline.

Horizon 3 is the hardest to protect. This horizon is where you invest to prevent disruption. By definition, Horizon 3 investments have a long-term payback, involve significant uncertainty, and are easy to cut when quarterly pressure rises. To point #2 above, this is where you need the executive team to act like a “first team” and make the necessary trade-offs to fund Horizon 3 investments, even when this may impact their team’s ability to succeed in Horizon 1.

Think about the companies that saw the AI wave forming in 2020 and invested in data infrastructure and product re-architecture before customers demanded it. By 2024, they were positioned to lead. The ones that waited are now reacting from a position of weakness.

The CEO’s question for Horizon 3 is: “ What shift could make our current advantage irrelevant within five years, and are we investing early enough to shape that shift?” When capital discipline is clear, the organization needs to understand why some initiatives scale and others stop. That clarity alone improves coherence and reduces friction, thereby improving CAC payback.

The $100-$500M Failure Mode

In my other article, I talked about “dominant failure modes” as a way to call out the one thing to watch out for at each stage. In the $100-$500M stage, the failure mode is not chaos; It is incrementalism. At this size, the company will usually keep growing, as it is likely a category leader, has an established customer base, and has built out appropriate processes and systems to support “at-scale” operations. But this growth tends to be about layering more on top of the same operating model. For example, you may add a new segment, a new product, or a new region. Maybe you introduce a new partner channel. And, at this size, you probably have bought at least one company to add “inorganic” growth. When you are inside this company, it feels like it is progressing because there is always something happening.

Meanwhile, CAC payback is extending because every new bet adds friction before it generates revenue. The organization compensates by holding more meetings, granting more exceptions, and escalating more to the top. The result is a company that looks active, but the growth is not compounded efficiently. The real impact is that your “next growth engine” never manifests. Over time, you end up optimizing the core until it slows down or you start to lose market share to more nimble/ smaller competitors. It becomes a death spiral because your options start to narrow, and the cost of change rises even higher.

Summary: The Real Shift

As mentioned above, the $100M to $500M transition is not about “handling complexity.” Most companies at this stage are already complex. The shift is economic. Growth costs more, as indicated by an increasing CAC Payback. The CEO’s job changes accordingly. It becomes less about inspiration and more about architecture. I would suggest there are 5 key pivots to embrace at this stage.

• First, your strategy must move from a priority list to a portfolio. At $100M, a company’s strategy can often seem like a list of annual initiatives. At this inflection point, the CEO needs to realize that they are managing a portfolio of bets with different payback windows and risk profiles. If the CEO does not force that discipline, everything will get a little funding, and nothing will get enough.

• Second, capital allocation must become explicit. CAC payback is not just a sales metric. It is a capital efficiency metric. Core expansion, new product, new segment, international, partner leverage, M&A, and AI initiatives all compete for capital. The CEO’s role is to define how those bets are compared and what must be true for them to continue. The shift is from “what feels important” to “what we are underwriting.”

• Third, you need to build a decision-making system that forces tradeoffs. At this scale, companies rarely fail due to a lack of effort. They stall because trade-offs are not clearly resolved. Functional leaders make rational local decisions that create enterprise friction. The CEO must clarify decision ownership, escalation paths, and the few priorities that truly matter across the enterprise.

• Fourth, focus requires subtraction. Between $150M and $250M, there are many viable paths. The key for the CEO is to be clear about what is the appropriate sequencing and timing. Even more important is naming what will not be done. This reassures people that the CEO is willing to make the hard calls and say no.

• Finally, CAC payback should be treated as a constraint, not a lagging metric.
  Top-line growth can mask stress. Second-order metrics do not. CAC payback, retention by segment, and sales efficiency trends often move before the P&L tells you something is wrong. Specifically, when CAC payback starts to extend, it usually signals ICP blur, conversion friction, or weak expansion relative to acquisition cost. Those are fixable problems. But they cannot be outspent indefinitely.

Ultimately, the CEO’s role in the $100- $500M range is not to be everywhere. It is to design a system that makes tradeoffs repeatable, capital allocation disciplined, and Horizon 2 real, while ensuring Horizon 1 remains strong enough to fund the journey. Between $100M and $500M, the companies that win are not the ones that add the most initiatives. They are the ones who manage the cost of growth while deliberately building the next engine.

Why Strategy Must Change as You Scale

Last year, I wrote an article called Embarking on the CEO’s Odyssey: Mastering the Art of Strategy. In that article, I used sailing as a metaphor. Essentially, if “execution” is about trimming the sails, then “strategy” is about setting the course. At the time, I was focused on helping CEOs answer a foundational question:

“Do you know the game you are playing, and can you win it?”

I introduced the strategic choice framework from Playing to Win that I learned at Salesforce, and which I have used at every company since.

As I have reflected on my own journey and started working with other CEOs of different-sized companies, I have started to think more about how the nature of strategy itself changes as you scale. I dug into this during a recent podcast with CEO Coacing Intenational, where I am now working.

As we talked about strategy, execution, and results, I realized that many of the hardest CEO struggles do not come from a lack of strategic thinking, but from using the wrong kind of strategy for the company’s current scale. What works at $5M can is not sufficient for a company that is $25M. Similarly, what you would do for a $100M company can feel onerous if you are $40M.

The trick is understanding what the core set of questions you need to answer at these inflection points is before you get there. The challenge is that most CEOs are not taught how to recognize these inflection points and lack a framework for how to adjust their strategy in response. Below, I provide a way to think about what you need to do as a CEO as you ramp from $1M to $100M+.

I would argue that there are 4 revenue thresholds that act as proxies for different strategic inflection points. In my experience, it is approximately at these stages that things start to break or the work feels extra hard because there is a mismatch between how the company is being run and what the next stage actually demands. Specifically, at each inflection point, three things shift:

• The underlying business dynamics,

• The role of the CEO, and

• The dominant failure mode

The purpose of the framework below is not to prescribe everything a CEO should do at a given size, but to surface the primary strategic question that must be answered, the role the CEO must play to answer it, and the failure mode that emerges when yesterday’s strategy is applied to tomorrow’s scale.

$1M ARR | Validation

Clarity Is Strategy

For a SaaS company, hitting $1M in ARR is a significant milestone. It is a validation of what calls Problem/Market fit. The two questions to answer are:

• “Do we have a real problem, with a real buyer, who is willing to spend real money?”

• “Is this replicable?” I.e., is there a swath or buyers (matching an ICP) that we can execute a repeatable motion with?

Knowing where to play and how to win matters. But at this scale, the goal with strategy is not a strategic framework. It is really about thinking about how to institute a set of constraints to ensure focus. Thus, the CEO’s job at this stage includes things such as:

• Narrowing the ICP until the signal is unmistakable

• Commiting to a primary use case

• Optimizing for learning velocity over growth rate

• Distinguishing traction from noise

The most common failure mode at this stage is confusing activity with validation. Especially as a first-time founder/CEO, booking real revenue, getting positive customer feedback, and/or having enthusiastic investors can create a false sense of confidence. In response, many CEOs start to add headcount, primarily Sales and Marketing to soon. They may start to broaden the product roadmap before they have truly validated problem/market fit. The result is that the organization scales complexity before it has scaled learning.

The impact?

You end up burning a lot of cash. and may be making “one-way” product decisions too early. This is how Seed or Series A capital gets consumed before the business has demonstrated that it can win in a repeatable way.

$10M ARR | Repeatability

From Individual Wins to Systems That Scale

$10M is the next major milestone that few startups actually achieve. In fact, according to Chart Mogul, only 13% of start-ups reach this milestone even after 10 years. If you are one of the few who make it, the good news is that the company is working. You have real customers who are paying you real money. And hopefully, you have a large # of customers renewing with you. The challenge at this stage is that success is driven mostly by individual heroics, direct founder involvement, or one-off deals.

At this stage, the key strategic question changes from “do we know the problem we are trying to solve” to “why do we win, and can we reproduce it without extraordinary effort?” This is also the stage where many companies plateau. It is not a demand problem. It is a systems challenge. This is where the “where to play, how to win” logic from my earlier article becomes critical. The CEO must now:

• Articulate a clear and consistent win narrative

• Enforce ICP and deal discipline

• Decide what kind of go-to-market company this actually is

• Design systems that make winning repeatable

The failure mode at this stage is the institutionalization of inconsistency. Early success often comes from adaptability, heroics, and deal-by-deal problem solving. For example, every rep sells a deal differently. You see this manifest in non-standard pricing and individual customer promises, which impact the product roadmap or cause major headaches for the implementation and customer success teams. You see this when the CEO or a few trusted leaders have to stay deeply involved to “make it work.” You see this when systems are implemented to manage growth rather than to reinforce strategy.

Some companies at this stage will argue that flexibility is a competitive differentiator. And I don’t disagree. This is how small companies compete with the big ones. However, over time, when this type of inconsistency becomes structural, the company will struggle to scale results, no matter how hard people work. This is what Kieth Pham is pointing to in this article. He says that only .4% of start-ups make it past that point. Yes, you read that right, 99.6% of start-ups are unable to scale beyond the founder.

$50M ARR | Orchestration

When Coordination Becomes the CEO’s Primary Job

If you have reached the $50M revenue threshold, you are in rarefied company. According to SaaStock only .04% of all SaaS companis are able to reach this milestone.

Why?

Because now you are dealing with compounding complexity. You have most likely moved from ICP to multiple ICPs. You probably have multiple products and may be working on selling a platform versus a point solution. You likely have different GTM motions (e.g., Product-led Growth [PLG] and Sales-led Growth [SLG]). You may have introduced partners as another channel, or perhaps you have executed your first M&A deal to get new capabilities or add more customers or segments. None of these initiatives are wrong. But you have to get really clear on the sequencing. At this stage, the strategic question becomes: “How do we align multiple growth vectors into one coherent GTM engine?”

This is why I prefer to describe this stage as “orchestration” rather than “focus” or anything else. The CEO is no longer just choosing a direction. They are integrating leaders, systems, and tradeoffs across product, go-to-market, and capital. His/her job is not about:

• Hiring and then aligning strong leaders who have competing priorities

• Managing tradeoffs explicitly, not implicitly

• Protecting organizational coherence as complexity rises

• Deciding where incremental investment actually produces returns using business cases and ROI analysis

The most common failure mode at this stage is allowing complexity to outrun coordination. As the company adds products, segments, GTM motions, and leadership layers, executives add meetings instead of strengthening business operating rhythms. In your weekly leadership meeting, executives spend time reporting on activity rather than being laser-focused on the metrics that matter and then teeing up the decisions on the highest-priority trade-offs.

Functional leaders optimize their own areas instead of operating as what Patrick Lencioni describes in his book The 5 Dysfunctions of a Team as a “first team” aligned on a small set of shared priorities. The organization stays busy and moves quickly, but it does not move together, and sustained growth becomes harder to achieve.

$100M ARR | Durability

Capital Allocation, Leadership Depth, and the Next Act

At $100M and beyond, the company no longer needs to prove it can grow. Instead, it needs to prove that it can endure. At this stage, the strategic question shifts to: “How do we compound advantage over time?”

Here, strategy becomes inseparable from capital allocation, developing leadership depth, and building credibility with a different set of investors as you prepare for an IPO or strategic exit. The CEO is no longer able to know all the details, nor should they. They need to be really good at asking good questions and then being the final say on tradeoff deadlocks and capital allocation. Their work is mostly about:

• Allocating capital across competing growth initiatives

• Building leadership depth and succession

• Reinforcing the core economic engine

• Ensuring the company delivers on the results for the next 12 months (horizon 1) while also investing in “next act”-type initiatives that pay off in the next 24-36 months (horizon 2 and 3)

The failure mode at this stage is optimizing near-term performance at the expense of long-term strength. CEOs and leadership teams feel the pressure to respond to investor expectations (especially if public), board pressure, or market conditions by prioritizing predictable, short-term results over investments that sustain advantage. They can defer leadership development, underinvest in succession, and starve future investments because that is a “next year” problem. They risk making capital allocation decisions that favor incremental optimization of the current engine rather than deliberate bets on what comes next.

Over time, the company may continue to meet its numbers, but it does so by narrowing its strategic options and weakening its ability to adapt to major market shifts. For example, every SaaS company is now confronting this exact existential challenge the rise of AI. Thus, the company could continue to grow, but its durability erodes. The last few years have produced no shortage of modern examples, from Twilio and DocuSign to Peloton and Zillow, where companies sustained growth by optimizing their existing engines, only to discover too late that they had underinvested in adaptability, leadership depth, and strategic optionality when the environment changed. In many ways, this dynamic lies at the core of what people are now calling the “SaaSpocalypse,” which I touched on in this post.

The Evolution Most CEOs Miss

One thing is true across all of these inflection points: as a company grows, its strategy must evolve. Knowing the game you are playing still matters. But, probably more importantly, is knowing when the rules of the game have changed. One way to see these inflection points before they become crises is to look at second-order metrics, not just top-line growth. In a previous article, I wrote about Net Revenue Retention and ARR per employee as indicators of organizational inflection points. What struck me then, and continues to show up in my coaching work, is that these metrics often start to move before leaders acknowledge that something fundamental has changed.

At $10M, declining NRR often signals that the company has not yet made the leap from individual wins to repeatable systems. Sales may still be closing new deals, but expansion slows because customers experience inconsistency across onboarding, product, and support. At $50M, stagnating or declining ARR per employee often reflects coordination failure. Headcount grows faster than leverage as teams add capacity without aligning around a single growth engine. And at $100M, both metrics can look “fine” in the short term, even as the organization quietly sacrifices adaptability by optimizing for near-term performance. In each case, the data is not just reporting results. It is revealing that the strategy running the company no longer matches its scale.

This is why I believe strategic inflection points are not abstract concepts. They reveal themselves if you have a scorecard in place and are focused on tracking, reviewing, and engaging leading indicators. CEOs who learn to read these types of signals can get ahead of strategy evolution deliberately, rather than being forced to react after options have narrowed.

SaaS (i.e., software) businesses are “people” businesses. They do not face the inventory, manufacturing, or working-capital dynamics of traditional product companies, which means the majority of their cost structure and execution risk are tied to the people. To that point, most SaaS companies spend 70 to 80% of their total costs on people. This is why SaaS CEOs and investors spend so much time talking about efficiency. The issue is that these conversations often collapse two very different ideas into one.

• Cost per employee

• Revenue per employee, typically expressed as ARR per FTE

They are related, but they answer fundamentally different questions. And when leaders conflate them, they end up optimizing the wrong thing at the wrong time.

Cost per Employee Is About Containment

Cost per employee ($/FTE) is usually the first efficiency lens applied, especially by private equity firms and late-stage investors. There is a standard playbook to improve $/FTE:

• Tap offshoring companies or build offshore teams to arbitrage labor costs.

• Hire contractors and fractional talent versus full-time employees to save on additional costs (e.g., benefits) and enable quick ramp-up/ramp-down based on business performance.

• Initiate tool consolidation and process automation to optimize the way work is done.

Over the past 3 years, we have seen significant experimentation with AI augmentation. All of these levers matter since they can materially improve margins and extend the runway. But they all operate on the same axis. They are about pulling cost out of the system. And this axis has a floor, and there is an asymptote to how much you can wring out of the cost structure

There are also risks to the business that are sometimes underappreciated by those running the spreadsheet. For example, code quality can start to slip, and the customer experience can be negatively impacted. Leadership overhead increases as alignment becomes harder to maintain across a thinner, more distributed organization.

AI can help smooth that curve by reducing friction and absorbing work that used to require people. But AI does not remove the underlying constraint. In many ways, cost efficiency is ultimately a defensive discipline. It helps you avoid inefficiency, but it does not, by itself, create scale. That is where the second metric comes in.

ARR per Employee Is About Leverage

ARR per employee (ARR/FTE) tells a very different story than $/FTE. Especially in early and mid-stage SaaS companies, ARR/FTE is less about productivity and more about trajectory. It measures whether the revenue engine is compounding faster than the organization required to support it. In other words, it asks a structural question: is the company getting the maximum leverage from the organization without risking an implosion? This is why investors care so deeply about both the level and the trend.

• When ARR/FTE is rising, something structural is working.

• When it flattens or declines, growth is usually still dependent on linear headcount, heroics, or fragmented coordination.

AI only sharpens this distinction. As Ray Rike, CEO of Benchmarkit, recently highlighted in a post, the gap between traditional SaaS companies and AI-native companies is widening quickly, particularly in ARR/FTE. And this is where many companies get tripped up. They assume that efficiency problems show up only on the cost side. In practice, many companies fail not because they overspend, but because they outgrow the organizational model they are using to scale.

ARR per FTE as a Signal of Organizational Maturity

For years, many SaaS leaders used ~$200K of ARR/FTE as a meaningful milestone. It was never a hard rule. This threshold often represents product-market fit, a repeatable go-to-market motion, and early evidence that systems (and tools) are starting to support the load.

What has changed in the last few years is not the metric’s relevance, but its trajectory. Ray’s Benchmark data shows that ARR/FTE has been improving since 2022, as companies were forced to grow more efficiently and became more disciplined about headcount. AI is now accelerating this trend by compressing coordination costs across engineering, marketing, operations, and customer support. In fact, according to Ray’s data, the median “Implied ARR per FTE” in public SaaS is $401K and a mean of $487K (2x what used to be seen as industry leading).

• AppLovin: $3.6M Implied ARR per FTE

• Palantir: $1.07M Implied ARR per FTE

• DataDog: $545K Implied ARR per FTE

• SnowFlake: $538K Implied ARR per FTE

• HubSpot: $367K Implied ARR per FTE

And there are 50 AI-native companies on track to deliver $1M ARR/ FTE (5x what legacy SaaS is targeting.

This is where I see CEOs sometimes make a subtle but serious mistake. They treat ARR/FTE as a pure productivity metric. It is not. I would suggest that ARR/FTE is best understood as a lagging indicator of organizational design. When this number improves sustainably, it usually reflects deeper shifts that have already taken place:

• Decisions are no longer bottlenecked at the top

• Roles are clearly defined and clearly owned

• Managers act as operators rather than messengers

• Systems carry the load instead of individuals

In other words, the company has crossed an organizational inflection point. That idea may sound modern, but it is not.

Why Rome Matters in a Conversation About Scale

For thousands of years, large institutions have struggled with the same basic constraints modern companies face today.

• Communication does not scale linearly.

• Trust degrades with distance from HQ.

• Coordination costs rise as organizations spread across space and time.

If anything, these challenges are amplified in remote-first and hybrid environments, where shared context erodes even faster. The Roman army is a helpful metaphor because it solved a problem every scaling organization eventually faces. How do you increase output without increasing dependence on individual heroics? Rome’s answer was institutionalizing scale. Specifically

By the time Rome reached its peak, its army was deliberately designed around predictable unit sizes, clear command layers, and replaceable leadership. A Roman legion was not a mob of thousands. It was a carefully nested system of smaller units, each sized so that the army could operate across distance, time, and turnover without losing effectiveness. That is the core insight worth carrying forward.

Size as a Design Constraint

The Roman army implicitly acknowledged something modern leaders still struggle to accept. There are natural limits to how many people one person can effectively lead, how much context can be shared informally, and how far trust can stretch without structure. The Roman army was designed around those limits. Small units were sized for cohesion and visibility. Larger formations existed to coordinate those units, not replace them. Command scaled through layers, each with clear authority and responsibility.

Contubernium

The Contubernium was the smallest living and fighting unit (~8 soldiers) in the Roman Army. This was the basic trust unit. Soldiers trained, marched, and slept together. Accountability was immediate and personal.

• Insight for modern SaaS. This size (2 teams and a pizza) allow for a high-context team where peer pressure and shared work replace process.

Century (Centuria)

A century was sized (~80 soldiers) sized so that a single experienced leader could directly command and enforce discipline.

• Insight for modern SaaS companies: this is where the first true management layer where leadership replaces proximity.

Cohort (Cohors)

The cohort was the primary tactical unit (~500 soldiers [6 centuries]) that could maneuver independently while still operating within a larger system.

Insight for modern SaaS companies. This is where you start to introduce functional team or department teams who have clear ownership and measurable outcomes.

Legion (Legio)

The legion was the core operating unit of Roman power. It combined combat, logistics, engineering, and command into a self-sustaining system and includes ~4,800 to 5,500 soldiers (10 cohorts)

Insight but modern SaaS companies: This is where a CEO needs to think about an integrated system rather than a collection of teams.

The Organizational Implication for CEOs

This is where the metaphor becomes directly relevant. As SaaS companies grow, they hit the same invisible thresholds:

• A CEO can no longer be in every decision

• Informal alignment stops working

• Proximity is replaced by process

• Trust must be reinforced by clarity

When companies ignore these inflection points, they compensate with intensity. More meetings. More heroics. More escalation to the top. Rome avoided this trap by building a leadership system. Individual commanders mattered, but the system mattered more. Units could be recomposed, and leaders could be replaced. But the Roman Legions marched on to conquer the world.

ARR/ FTE is the modern proxy for that same transition. It signals when an organization has stopped relying on intensity and started relying on structure. The next question, of course, is what leaders are supposed to do with that signal.

Mapping ARR per FTE to Org Design Requirements

This is where ARR per FTE becomes operational rather than theoretical. At lower ARR per FTE levels, most companies are still building capacity. Hiring runs ahead of revenue, particularly in sales and marketing. Decision-making is centralized with the CEO. Communication and coordination feel smooth because everyone shares context, often literally in the same room.

As ARR per FTE rises, that model stops working.

Unless the organization evolves, growth slows or breaks.

Conceptually, the inflection points tend to follow a familiar pattern.

Lower ARR per FTE

• Founders make most decisions

• Context is shared informally

• Accountability is loose but effective

• Speed comes from proximity

• A lot gets done quickly

Mid-range ARR per FTE

• The first real management layer appears

• Functional ownership becomes necessary

• Processes begin to form

• The CEO must shift from doing to delegating

Higher ARR per FTE

• Managers manage managers

• Work is automated and codified

• Roles deepen and specialize

• Generalists give way to specialists

• Executives focus on prioritization, capital allocation, and context-setting rather than execution. At this stage, leadership is no longer about doing the work. It is about designing and maintaining the system that does the work.

When ARR per FTE rises faster than organizational maturity, the symptoms are predictable. Decisions slow. Accountability blurs. Culture becomes a values slide instead of a lived experience.

That friction is not a sign of failure. It is a sign that the organization is overdue for redesign.

AI Changes the Economics, Not the Constraints

AI forces leaders to revisit assumptions we have held for centuries about how organizations scale. AI can augment individual output; it can compress coordination costs; and, as many people fear, it can eliminate entire categories of work (i.e., jobs). The first-order impact will be improved cost efficiency per employee. Over time, it shows up as higher ARR per FTE.

Agentic systems push this logic even further. They raise the possibility that workflows, decisions, and execution loops can be partially or fully automated. This has fueled speculation about billion-dollar companies run by a small executive team. Some of that will prove true, and early examples already show extreme ARR per FTE outcomes.

But one constraint remains stubbornly intact. Organizations still hit inflection points driven by:

• Communication

• Coordination

• Clarity

• Leadership legitimacy

AI can amplify people. It cannot eliminate the need to create shared context, align on intent, make decisions, and enforce accountability. Those are human constraints, not technological ones. The mistake is assuming that because AI allows revenue to scale faster, organizational redesign can wait. History suggests the opposite. Rome’s lesson is not about armies. It is about respecting the constraints of human coordination. In fact, the faster output scales, the more deliberate leaders must be about structure.

The Core CEO Question

Which brings us to the real question. It is not how high ARR/FTE can go. It is whether the organization is evolving fast enough to support it. The companies that win will not be the ones that chase efficiency in isolation. They will be the ones who recognize that every efficiency gain creates a new organizational inflection point. That pattern has held for thousands of years. ARR/ FTE may rise dramatically in the AI era, but organizational inflection points do not disappear. They simply arrive sooner and with higher stakes.

Today, I launch my next chapter as a CEO coach, so I thought I would write an article about why and why now.

The CEO role is a lonely role

The CEO role places final accountability on one person, even though information is incomplete and unevenly distributed. As Ben Horowitz writes in his book The Hard Thing about Hard Things, CEOs are paid to make the decisions and be accountable for the outcomes. I learned this in my first CEO role at Rocket Fuel, which was a first-gen AI public company at the time. I stepped into the role after serving as CRO for six months. Overnight, my peers became direct reports, my manager became the board, and my world changed forever.

Monte Zweban, the chairman of our board, said something to me early on that proved true: “Nothing prepares you for being a CEO other than being a CEO.” As CEO, I was responsible for everything. I was expected to project confidence and even when I did not really know the right answer. I was learning the role in real time and having to make decisions with incomplete information. As a public company, many of these key decisions were visible and often irreversible “one-way door” decisions.

In the Navy, we used to say: “you can delegate authority, but not responsibility.” The CEO role works the same way. At the end of the day, the buck stops with you. This combination of increased accountability and exponential complexity creates pressure that other leadership roles do not.

The research is clear: My experience was not unique. It is structural.

• Research published by Harvard Business Review shows that CEO isolation is common and increases as leaders move up the organization. In one survey, roughly half of CEOs reported feeling isolated. A majority said it negatively affected their performance.

• Research from Stanford Graduate School of Business, led by David Larcker, reinforces this conclusion. Nearly two-thirds of CEOs said they needed leadership advice from outside their organization. Most said they lacked a safe internal source to get it.

Here is the kicker: Board members, as much as they might want to, can’t fill the gap. Nor can your executive team. Unfortunately, the CEO role inherently limits your ability to be fully candid with anyone. The result is predictable. CEOs make the hardest decisions with the least unfiltered input, and this is where isolation can lead to degraded performance for both the CEO and the business. First-time CEOs report an even higher impact. According to the research, many cite stress, self-doubt, and burnout. These conditions distort decision-making and slow learning.

The issue is not the CEO’s grit, determination, or intelligence. The issue is structural to the role.

What actually helps?

CEOs need an external perspective that is confidential, experienced, and non-political. Over time, I learned that effective CEOs build a support system, which I refer to as “building your tribe.” Your tribe includes four distinct roles.

• Mentor: A mentor provides pattern recognition. A mentor has operated at your altitude before. When I became CEO, I sought a mentor with public company and turnaround experience. I ended up working with Tony Zingale, a seasoned public-company CEO, who helped me see risks sooner and frame problems more clearly. His value came from his context and experience. Plus, he was just a great guy to talk to.

• Coach: A coach focuses on how you think, communicate, and lead under pressure. At RocketFuel, I worked with John Baird to better understand my personality, triggers, decision-making patterns, and boardroom presence. That partnership improved my effectiveness more than tactical advice would have.

• Peer groups. CEO Peer groups can provide perspective. Peer forums allow honest discussion without consequence. I participated in groups such as the Alliance of CEOs, Vistage, and Venwise. Other groups like EO and YPO target earlier-stage/founder-led companies. All of these groups can help reduce your sense of isolation by meeting monthly and creating a forum where you can share your personal wins/losses and priorities. I know many CEOs who find these cohorts extremely valuable and stay with them for years.

• A personal advisory board (PAB). Building out a PAB helps keep you grounded. The board should include people who know you well and will challenge you. In my own experience, my PAB helped me stay aligned with my goals at a critical career inflection point. I was going to accept a job with a company I thought was exciting, and they really wanted me. My PAB helped me see the challenges with the company and its context, and honestly helped me avoid a falling-knife company.

Each role solves a different problem. Together, they can help counteract isolation, providing valuable thought partnership when dealing with tough people, business, and/or personal decisions.

Why now?

Over the last decade, I served as CEO in several contexts: a ~$500M public company and take-private, a venture-backed pivot and strategic sale, and a private-equity-backed integration. Each situation was different, and each company had different challenges and opportunities. The common thread was that I always desired more help from those who have been there/done that.

Along the way, I began advising CEOs and executive teams as a board member and through CEOx.io. I found that the highest-value conversations and the ones I enjoyed most were not about tactics. They were ones where I talked with the CEO about their “why?” and then helped them think deeply about business critical decisions.

For the past 4 months, I have been training with CEO Coaching International. I chose CEOCI for three reasons.

• They work exclusively with CEOs.

• They use a disciplined execution system called Make Big Happen.

• Every partner is a former CEO.

This move is not a career pivot; it is a continuation of a career working with teams to overcome immense challenges. Now I am narrowing my focus to help CEOs and their leadership teams to scale smarter, operate better, and grow with confidence.

CEO as Athlete

The paradox of the CEO role is that it concentrates authority and uncertainty simultaneously. That combination creates intense pressure, which exposes a CEO’s limits in judgment, stamina, and focus.

The CEOs who scale well behave like elite athletes. They do not rely solely on grit. They train deliberately. They do the hard work consistently. They use coaches to sharpen technique, improve decision-making under pressure, and unlock more of their potential.

Peak performance is not accidental. It is the result of intention, commitment, and coaching.

Sources

• Harvard Business Review, It’s Time to Acknowledge CEO Loneliness

• Harvard Business Review, Work and the Loneliness Epidemic

• Stanford Graduate School of Business, David F. Larcker, Lonely at the Top? It Resonates with Most CEOs