ESSAY
December 15th, 2024
How To Think About Large Groups of People
How To Think About Large Groups of People
SETH HOLLIS
SETH HOLLIS
The fundamental challenge of building for scale has never been technological—it's been epistemological. How do we know what we claim to know about large groups of humans? What justifies our confidence when speaking about the preferences, behaviors, and motivations of thousands or millions of people we've never met? This question lies at the heart of every business endeavor that aims beyond individual craftsmanship into mass production or mass service, yet it remains surprisingly unexamined by most organizations. We develop sophisticated technical infrastructures for serving large populations while relying on dangerously simplistic models for understanding them. The result is a peculiar form of confidence—executives and entrepreneurs who can explain in precise detail how their distribution systems work but can only gesture vaguely when asked how they know what their customers actually want or why they behave as they do. This gap between operational sophistication and interpretive crudeness explains why so many technically excellent products fail to resonate, why so many carefully planned campaigns miss their mark, and why organizations remain perpetually surprised by how the very people they've been studying for years actually respond to their offerings. The problem isn't insufficient data—modern organizations drown in customer information—but rather in how we conceptualize the relationship between individual complexity and collective patterns.
The fundamental challenge of building for scale has never been technological—it's been epistemological. How do we know what we claim to know about large groups of humans? What justifies our confidence when speaking about the preferences, behaviors, and motivations of thousands or millions of people we've never met? This question lies at the heart of every business endeavor that aims beyond individual craftsmanship into mass production or mass service, yet it remains surprisingly unexamined by most organizations. We develop sophisticated technical infrastructures for serving large populations while relying on dangerously simplistic models for understanding them. The result is a peculiar form of confidence—executives and entrepreneurs who can explain in precise detail how their distribution systems work but can only gesture vaguely when asked how they know what their customers actually want or why they behave as they do. This gap between operational sophistication and interpretive crudeness explains why so many technically excellent products fail to resonate, why so many carefully planned campaigns miss their mark, and why organizations remain perpetually surprised by how the very people they've been studying for years actually respond to their offerings. The problem isn't insufficient data—modern organizations drown in customer information—but rather in how we conceptualize the relationship between individual complexity and collective patterns.
The default approach to understanding large groups—segmentation, averaging, and persona creation—provides the comfort of apparent comprehension while often obscuring the very insights necessary for meaningful connection. We reduce millions to archetypes, complex motivational landscapes to bullet points, and the messy reality of human decisions to clean customer journeys. This reduction feels necessary; the alternative—acknowledging the true complexity of large-scale human behavior—seems paralyzing. If every person is unique and every decision context-dependent, how can we possibly create anything that works at scale? This fear drives us toward oversimplification, toward the false clarity of statements like "our customer wants..." as if tens of thousands of individuals share a singular desire expressible in a single sentence. The more meaningful approach begins with a conceptual shift: large groups don't think, feel, or want anything. Only individuals do these things. What large groups exhibit are patterns—statistical regularities in behavior that emerge not despite individual complexity but because of it. The patterns are real and actionable, but the motivational narratives we construct to explain them are often fictional conveniences. The skilled interpreter of human behavior at scale recognizes this distinction and works within it, using patterns to identify opportunities while resisting the temptation to overlay simplistic psychological explanations that feel satisfying but lead to misguided decisions.
The default approach to understanding large groups—segmentation, averaging, and persona creation—provides the comfort of apparent comprehension while often obscuring the very insights necessary for meaningful connection. We reduce millions to archetypes, complex motivational landscapes to bullet points, and the messy reality of human decisions to clean customer journeys. This reduction feels necessary; the alternative—acknowledging the true complexity of large-scale human behavior—seems paralyzing. If every person is unique and every decision context-dependent, how can we possibly create anything that works at scale? This fear drives us toward oversimplification, toward the false clarity of statements like "our customer wants..." as if tens of thousands of individuals share a singular desire expressible in a single sentence. The more meaningful approach begins with a conceptual shift: large groups don't think, feel, or want anything. Only individuals do these things. What large groups exhibit are patterns—statistical regularities in behavior that emerge not despite individual complexity but because of it. The patterns are real and actionable, but the motivational narratives we construct to explain them are often fictional conveniences. The skilled interpreter of human behavior at scale recognizes this distinction and works within it, using patterns to identify opportunities while resisting the temptation to overlay simplistic psychological explanations that feel satisfying but lead to misguided decisions.
What makes this approach both powerful and difficult is that it requires simultaneously holding two seemingly contradictory truths: that humans are far too complex to be reduced to simple models, and that their behavior in aggregate follows discoverable patterns that can be meaningfully engaged. The organizations that navigate this paradox successfully develop what might be called epistemological humility—they make decisions based on observed patterns while maintaining awareness of how much remains unknown about the individuals creating those patterns. They recognize that their models are provisional tools rather than accurate representations, useful for generating hypotheses but always requiring verification through actual human response. They understand that segmentation is a retrospective analytical convenience, not a pre-existing reality to be discovered. Most importantly, they resist the temptation to confuse statistical regularities with psychological uniformity, recognizing that similar behaviors can emerge from radically different internal states and motivations. This humility doesn't paralyze decision-making—quite the opposite. It creates space for continuous discovery rather than presumed expertise, for genuine curiosity rather than confirming pre-existing beliefs. Organizations with this orientation build feedback mechanisms that test not just whether people use their products but whether their understanding of why people use them is accurate.
What makes this approach both powerful and difficult is that it requires simultaneously holding two seemingly contradictory truths: that humans are far too complex to be reduced to simple models, and that their behavior in aggregate follows discoverable patterns that can be meaningfully engaged. The organizations that navigate this paradox successfully develop what might be called epistemological humility—they make decisions based on observed patterns while maintaining awareness of how much remains unknown about the individuals creating those patterns. They recognize that their models are provisional tools rather than accurate representations, useful for generating hypotheses but always requiring verification through actual human response. They understand that segmentation is a retrospective analytical convenience, not a pre-existing reality to be discovered. Most importantly, they resist the temptation to confuse statistical regularities with psychological uniformity, recognizing that similar behaviors can emerge from radically different internal states and motivations. This humility doesn't paralyze decision-making—quite the opposite. It creates space for continuous discovery rather than presumed expertise, for genuine curiosity rather than confirming pre-existing beliefs. Organizations with this orientation build feedback mechanisms that test not just whether people use their products but whether their understanding of why people use them is accurate.
The most successful approaches to scale recognize that the goal isn't to understand millions of people—an impossibility—but to create offerings flexible enough to be meaningful to millions of individuals for their own reasons. This subtle but profound shift moves from "understanding the customer" to "creating possibilities that diverse customers can find meaningful in diverse ways." Consider the difference between a product designed for a narrowly defined persona versus one designed around a capability that serves a broadly shared need while accommodating individual variation in how that need manifests. The former assumes homogeneity where none exists; the latter acknowledges heterogeneity while recognizing the patterns within it. This perspective explains why the most enduring scaled products often provide platforms for personalization rather than prescriptive solutions, why they create spaces for individual meaning-making rather than imposing singular narratives. It explains why attempts to appeal to everyone through genericness typically appeal to no one, while products with strong points of view find passionate audiences that engage for their own varied reasons. The paradox resolves: to succeed at scale requires not understanding everyone but creating something specific enough to matter differently to different people.
The most successful approaches to scale recognize that the goal isn't to understand millions of people—an impossibility—but to create offerings flexible enough to be meaningful to millions of individuals for their own reasons. This subtle but profound shift moves from "understanding the customer" to "creating possibilities that diverse customers can find meaningful in diverse ways." Consider the difference between a product designed for a narrowly defined persona versus one designed around a capability that serves a broadly shared need while accommodating individual variation in how that need manifests. The former assumes homogeneity where none exists; the latter acknowledges heterogeneity while recognizing the patterns within it. This perspective explains why the most enduring scaled products often provide platforms for personalization rather than prescriptive solutions, why they create spaces for individual meaning-making rather than imposing singular narratives. It explains why attempts to appeal to everyone through genericness typically appeal to no one, while products with strong points of view find passionate audiences that engage for their own varied reasons. The paradox resolves: to succeed at scale requires not understanding everyone but creating something specific enough to matter differently to different people.
Perhaps the most valuable reframing comes in recognizing that patterns in large groups aren't primarily psychological but contextual—shaped less by who people inherently are and more by the situations they navigate. When thousands of people make similar choices, it's rarely because they share identical internal states but because they face similar constraints, opportunities, and environmental cues. This shift from dispositional to situational understanding transforms how organizations approach scale. Rather than trying to classify people into types, they identify common situational factors that shape behavior across otherwise diverse individuals. They ask not "what kind of person makes this choice" but "what situation leads diverse people toward this choice?" The tactical implications are significant: focusing less on who customers are demographically and more on the contexts in which they encounter offerings, designing less for imagined psychological profiles and more for actual decision environments, messaging less around identity alignment and more around situational relevance. The humility in this approach—acknowledging that we cannot truly know millions of minds but can understand the contexts they navigate—produces not paralysis but precision, not generic offerings but contextually relevant ones that resonate not because they "understand the customer" but because they address the realities people actually face.
Perhaps the most valuable reframing comes in recognizing that patterns in large groups aren't primarily psychological but contextual—shaped less by who people inherently are and more by the situations they navigate. When thousands of people make similar choices, it's rarely because they share identical internal states but because they face similar constraints, opportunities, and environmental cues. This shift from dispositional to situational understanding transforms how organizations approach scale. Rather than trying to classify people into types, they identify common situational factors that shape behavior across otherwise diverse individuals. They ask not "what kind of person makes this choice" but "what situation leads diverse people toward this choice?" The tactical implications are significant: focusing less on who customers are demographically and more on the contexts in which they encounter offerings, designing less for imagined psychological profiles and more for actual decision environments, messaging less around identity alignment and more around situational relevance. The humility in this approach—acknowledging that we cannot truly know millions of minds but can understand the contexts they navigate—produces not paralysis but precision, not generic offerings but contextually relevant ones that resonate not because they "understand the customer" but because they address the realities people actually face.
How To Think About Large Groups of People
ESSAY
ESSAY
December 15th, 2024
ESSAY
ESSAY
April 2nd, 2025
How To Think About Large Groups of People
The fundamental challenge of building for scale has never been technological—it's been epistemological. How do we know what we claim to know about large groups of humans? What justifies our confidence when speaking about the preferences, behaviors, and motivations of thousands or millions of people we've never met? This question lies at the heart of every business endeavor that aims beyond individual craftsmanship into mass production or mass service, yet it remains surprisingly unexamined by most organizations. We develop sophisticated technical infrastructures for serving large populations while relying on dangerously simplistic models for understanding them. The result is a peculiar form of confidence—executives and entrepreneurs who can explain in precise detail how their distribution systems work but can only gesture vaguely when asked how they know what their customers actually want or why they behave as they do. This gap between operational sophistication and interpretive crudeness explains why so many technically excellent products fail to resonate, why so many carefully planned campaigns miss their mark, and why organizations remain perpetually surprised by how the very people they've been studying for years actually respond to their offerings. The problem isn't insufficient data—modern organizations drown in customer information—but rather in how we conceptualize the relationship between individual complexity and collective patterns.
The default approach to understanding large groups—segmentation, averaging, and persona creation—provides the comfort of apparent comprehension while often obscuring the very insights necessary for meaningful connection. We reduce millions to archetypes, complex motivational landscapes to bullet points, and the messy reality of human decisions to clean customer journeys. This reduction feels necessary; the alternative—acknowledging the true complexity of large-scale human behavior—seems paralyzing. If every person is unique and every decision context-dependent, how can we possibly create anything that works at scale? This fear drives us toward oversimplification, toward the false clarity of statements like "our customer wants..." as if tens of thousands of individuals share a singular desire expressible in a single sentence. The more meaningful approach begins with a conceptual shift: large groups don't think, feel, or want anything. Only individuals do these things. What large groups exhibit are patterns—statistical regularities in behavior that emerge not despite individual complexity but because of it. The patterns are real and actionable, but the motivational narratives we construct to explain them are often fictional conveniences. The skilled interpreter of human behavior at scale recognizes this distinction and works within it, using patterns to identify opportunities while resisting the temptation to overlay simplistic psychological explanations that feel satisfying but lead to misguided decisions.
What makes this approach both powerful and difficult is that it requires simultaneously holding two seemingly contradictory truths: that humans are far too complex to be reduced to simple models, and that their behavior in aggregate follows discoverable patterns that can be meaningfully engaged. The organizations that navigate this paradox successfully develop what might be called epistemological humility—they make decisions based on observed patterns while maintaining awareness of how much remains unknown about the individuals creating those patterns. They recognize that their models are provisional tools rather than accurate representations, useful for generating hypotheses but always requiring verification through actual human response. They understand that segmentation is a retrospective analytical convenience, not a pre-existing reality to be discovered. Most importantly, they resist the temptation to confuse statistical regularities with psychological uniformity, recognizing that similar behaviors can emerge from radically different internal states and motivations. This humility doesn't paralyze decision-making—quite the opposite. It creates space for continuous discovery rather than presumed expertise, for genuine curiosity rather than confirming pre-existing beliefs. Organizations with this orientation build feedback mechanisms that test not just whether people use their products but whether their understanding of why people use them is accurate.
The most successful approaches to scale recognize that the goal isn't to understand millions of people—an impossibility—but to create offerings flexible enough to be meaningful to millions of individuals for their own reasons. This subtle but profound shift moves from "understanding the customer" to "creating possibilities that diverse customers can find meaningful in diverse ways." Consider the difference between a product designed for a narrowly defined persona versus one designed around a capability that serves a broadly shared need while accommodating individual variation in how that need manifests. The former assumes homogeneity where none exists; the latter acknowledges heterogeneity while recognizing the patterns within it. This perspective explains why the most enduring scaled products often provide platforms for personalization rather than prescriptive solutions, why they create spaces for individual meaning-making rather than imposing singular narratives. It explains why attempts to appeal to everyone through genericness typically appeal to no one, while products with strong points of view find passionate audiences that engage for their own varied reasons. The paradox resolves: to succeed at scale requires not understanding everyone but creating something specific enough to matter differently to different people.
Perhaps the most valuable reframing comes in recognizing that patterns in large groups aren't primarily psychological but contextual—shaped less by who people inherently are and more by the situations they navigate. When thousands of people make similar choices, it's rarely because they share identical internal states but because they face similar constraints, opportunities, and environmental cues. This shift from dispositional to situational understanding transforms how organizations approach scale. Rather than trying to classify people into types, they identify common situational factors that shape behavior across otherwise diverse individuals. They ask not "what kind of person makes this choice" but "what situation leads diverse people toward this choice?" The tactical implications are significant: focusing less on who customers are demographically and more on the contexts in which they encounter offerings, designing less for imagined psychological profiles and more for actual decision environments, messaging less around identity alignment and more around situational relevance. The humility in this approach—acknowledging that we cannot truly know millions of minds but can understand the contexts they navigate—produces not paralysis but precision, not generic offerings but contextually relevant ones that resonate not because they "understand the customer" but because they address the realities people actually face.
SETH HOLLIS