Early Adopters

Rogers' diffusion curve: people split by adoption speed into five segments — each wanting something completely different

New things are never accepted uniformly. Rogers split people by order of adoption into five segments along a bell curve: innovators (~2.5%, chasing novelty itself), early adopters (~13.5%, wanting vision and first-mover edge), early majority (~34%, wanting proven, practical value), late majority (~34%, following only under peer pressure), and laggards (~16%, adopting passively).

Non-trivial: (1) these five are not personality labels but positions relative to a specific innovation. The same person is an early adopter of AI tools and possibly a laggard in fashion — don't tag people permanently. (2) Each segment adopts for reasons and needs evidence that are completely different: innovators want novelty; early adopters bet on vision and competitive edge and tolerate roughness; the early majority trusts only practical results that "others have used, with word-of-mouth." (3) So the pitch that wins one segment repels the next — "be the first to try it" tempts early adopters but alarms the early majority (nobody's used it = risk). (4) Early adopters are the opinion leaders bridging the few and the many; the social capital in their hands is the hinge of whether diffusion carries onward.

Practice: to spread anything new, find that small group of early adopters first, rather than arguing with the most stubborn laggards. Concentrate limited resources on them and let them vouch for you to the majority. Push with the adoption curve, not against it.

↑ The Chasm Innovators Early Adopters Early Majority Late Majority Laggards ← adoption time → (population distribution)
The bell curve of adopters: between early adopters and the early majority lies the "chasm" of Card 2
Classic example

The spread of hybrid corn seed across rural America: the first to switch were a few risk-tolerant farmers with slack to experiment; neighbors followed only after seeing their real yield gains. The seed's technical advantage was there all along — what truly set the diffusion pace was "who believes first, and who watches whom" in the social network.

Scenario · BigCat

(1) AI super-individual: those who fold LLMs and agent workflows into daily work earliest gain a near-uncontested compounding window — the early-adopter dividend is essentially the time lag before others catch up. (2) Rolling out a new tool on a team: don't start with the most resistant person; let the tinkerer with credibility produce results first, then let the results speak. Ignite the opinion leaders, not the whole roster.


English Prompt
I'm trying to spread [a new product / tool / practice] but it's an uphill battle. Use the diffusion-of-innovations adoption curve to analyze: 1. Which segment am I mostly speaking to (innovators / early adopters / early majority)? Does my pitch match the evidence they need? 2. Who are the likely early adopters and opinion leaders in this circle, and what's their real motive to adopt? 3. Give me a sequence to "win early adopters first, then leverage word-of-mouth to move the majority."

Crossing the Chasm

Moore: a deep chasm splits early adopters from the early majority — most new products die there

Moore found that the diffusion curve is not a smooth ramp — between early adopters and the early majority yawns a chasm. Countless acclaimed products die precisely at the step from "loved by a few visionaries" to "broadly adopted by the mainstream."

Non-trivial: (1) the chasm exists because the two sides want exactly opposite things: early adopters want a revolutionary leap and tolerate imperfection for the sake of being ahead; the early majority (pragmatists) wants incremental improvement, an out-of-the-box complete solution, and — proof that other pragmatists already use it. That's the deadlock: the majority waits for the majority. (2) Early hype is a dangerous false signal. Visionaries' love doesn't predict mainstream success, yet easily fools you into scaling too soon and burning resources. (3) The crossing move is counterintuitive — not broader but narrower: the "bowling-alley" strategy concentrates all firepower on one narrow niche, becoming the undisputed number one in that small pond. Word-of-mouth in that pond spreads among pragmatists, then knocks over adjacent markets like bowling pins.

Practice: when a product is adored by the tech crowd but can't open the mainstream market, don't raise the volume — first seize one painful, narrow beachhead completely. Better a big fish in a small pond than a forgotten small fish in a big one.

Classic example

Tech history is full of products lavished with praise by the press and geeks yet never crossing the chasm: the demo dazzles, visionaries are obsessed, but ordinary users find no "must-have" reason and it stalls in a niche. The higher the early hype, the easier it is to misjudge the mainstream's real needs.

Scenario · BigCat

(1) Rolling out an AI agent in a company: the demo wows management and tech early-tasters, but it stalls the moment a pragmatic business team must depend on it daily — they want stability, completeness, and peer precedent, not flash. (2) Startup products are the same: don't mistake seed users' obsession for PMF (product-market fit); become indispensable in one extremely narrow scenario first, then expand. Cross the chasm by focus, not by volume.


English Prompt
My [product / solution / initiative] is loved by a small circle but stuck before the larger crowd. Diagnose it with the chasm model: 1. Are my current fans early adopters (wanting vision) or the pragmatic majority (wanting a complete solution + peer precedent)? Where do their wants diverge? 2. How much of my early traction is a "false signal" tempting me to scale too soon? 3. Help me pick one narrow, painful beachhead niche and design a path to dominate it first, then expand outward.

Tipping Point Threshold

Everyone holds a "how many must go first before I do" threshold — ignition depends on whether the threshold chain links up

Why does the same idea sometimes blaze overnight and sometimes pass in silence? The threshold model answers: everyone holds a threshold — "what fraction around me must adopt before I follow." Ignition depends not on how persuasive the idea is, but on whether these thresholds link end-to-end into a chain.

Non-trivial: (1) you must first have a few near-zero-threshold people (innovators/early adopters) to light the first spark, or the chain can't start — this loops back to Card 1. (2) What matters is not the average but the threshold distribution. Even one person's threshold being a notch too high can break the chain there, and outcomes diverge wildly — the same crowd, the same idea, can either ignite or fizzle. This extreme sensitivity to tiny differences is the fingerprint of a complex system's critical point (echoing phase transitions and the butterfly effect). (3) It shares the same math as an epidemic's R-value and percolation: below critical mass, each adoption can't drive enough next ones and dies out; above critical mass, it turns into a self-sustaining cascade — positive feedback ignited (echoing cybernetics, D59).

Practice: to ignite, don't thin resources across a big crowd. Pile adoption density past critical mass inside one small network, let it catch fire locally first, then cascade outward. Momentum itself lowers bystanders' thresholds — "so many people are already using it" is the strongest nudge.

critical mass crossed → cascade below → dies out time →
Same start: below critical mass it decays and dies; past critical mass it turns into a self-sustaining cascade
Classic example

A standing ovation: a few in the front rows stand first, the back rows' thresholds get crossed one by one, and within seconds the whole house is up; but if too few stood at the start, they awkwardly sit back down. Rumors, fads, and viral hits follow the same logic — ignored at first, then suddenly everyone is talking about it once a critical point is passed.

Scenario · BigCat

(1) Rolling out a new norm in an organization: rather than mass-emailing a broad call to everyone, get one team to use it universally, break that small circle's critical mass first, then let momentum spill over. (2) Network-effect products inherently have a tipping point — too few users and everyone finds it useless and quits (a vicious cascade); past critical mass it becomes self-reinforcing. (3) The rise of social movements and open-source projects works the same way. Ignition comes from local density, not global breadth.


English Prompt
I want [a practice / product / idea] to truly tip in [a group / org / market], not surge then sink. Help me through the threshold and critical-mass lens: 1. Whose threshold is lowest (likely first adopters)? Whose is high enough to become a break in the chain? 2. Roughly what "critical mass" triggers a self-sustaining cascade, and how far am I from it? 3. Instead of spreading thin, which small circle should I concentrate density in to ignite first?

S-Curve Saturation

Slow start — explosive growth — saturation: the fated shape of nearly all diffusion, and how to jump to the next curve

Plot cumulative adopters and nearly all diffusion traces an S-shaped curve: a slow start, a steep explosion past the tipping point, then a plateau against a ceiling. It is the universal signature of bounded growth.

Non-trivial: (1) the S-curve = early positive feedback (contagion, word-of-mouth) tamed by late negative feedback (fewer non-adopters left) — exponential growth bent into an S by a population/resource ceiling (echoing the positive/negative-feedback balance of cybernetics, D59). (2) The most deceptive part is the middle: the steep explosion makes you believe it will rise forever, but at the inflection point growth has already begun to decelerate — still positive, yet the second derivative has turned negative. Treating the steep stretch as eternal and extrapolating linearly is a common cause of valuation bubbles and overexpansion. (3) Saturation is the signal forcing you to jump to the next S-curve: mature technology inevitably tops out, and the only escape is to launch a new curve (the second curve) while the old one hasn't saturated and you still have resources. Wait until it's flat to look for a way out, and there's often nothing left to deploy. (4) A technology's "performance vs effort" also follows an S: marginal returns plunge near maturity, which itself signals it's time to switch paradigms.

Practice: faced with any rapid growth, first ask "which segment of the S-curve am I on?" In the explosion phase, don't extrapolate the future linearly; near saturation, divert part of your profit and attention to incubate the next curve. True foresight is leaping to the next curve before the peak.

inflection (growth peaks) saturation plateau second curve time → (cumulative adoption)
The S-curve: at the inflection point growth is already slowing; the visionary leaps to the second curve before the old one saturates
Classic example

The penetration of TV, mobile phones, and the internet are all S-curves: lukewarm for years, sweeping in within a few, then approaching saturation. Bacterial growth in a petri dish is the same — as nutrients run out, exponential growth ultimately plateaus.

Scenario · BigCat

(1) Product user growth and the mastery of a skill both go through a "plateau" — not failure, but an S-curve saturating; either dig deeper for a breakthrough or jump to a new curve. (2) AI: is today's large-model scaling-by-stacking-compute-and-data approaching the inflection of some S-curve? If so, the next breakthrough may come from a paradigm leap rather than linear addition — a key lens for judging technology inflection points. (3) Career: positioning a second curve before topping out in your current track is far more graceful than turning around only after grinding yourself dry.


English Prompt
I'm watching/experiencing [some growth: product / skill / industry / technology] and want to judge where it sits on the S-curve and what comes next. Help me: 1. Using the S-curve, where are we: slow start, explosion, inflection, or near saturation? On what evidence? 2. If in the explosion phase, am I dangerously extrapolating linearly? Roughly where is the ceiling? 3. If near saturation, what could my "second curve" be, and how much should I invest now to incubate it?