Bitcoin Adoption on the S-Curve

If bitcoin adoption behaves like other innovations, it follows an S-curve: slow at first, then steep, flat at the end. As of November 2026 the best available estimates put bitcoin owners at around 4 to 4.5% of the world's population. In the model of sociologist Everett Rogers, that places bitcoin between the innovators and the early adopters, roughly where the internet stood around the year 2000. The Power Law takes a different view: it assumes growth that slows a little every year but never hits a ceiling.

What the S-curve describes

New things don't spread evenly. A few curious people try them early, more follow because they see others using them, and at some point almost everyone who wants it has it. Plot the share of users against time and you get a reclining S: flat start, steep middle, flat end.

The classic example is hybrid corn in Iowa. In 1941 the sociologists Bryce Ryan and Neal Gross asked farmers when they had switched to the new seed. From 1928 it took several years before a handful of farmers used it, then the share rose steeply, and by the early 1940s it was standard almost everywhere. The same shape later turned up for refrigerators, colour televisions, mobile phones and the internet.

The S-curve is not a law of nature. It is an observation that fits many products well and some not at all, for example when a product disappears before it reaches the majority.

Rogers' five groups: from innovators to laggards

Everett Rogers summed up the research in 1962 in his book Diffusion of Innovations. His idea: if you record for each person when they adopt an innovation, you usually get a bell curve. The S-curve is simply the cumulative bell. Rogers cut the bell at its standard deviations into five groups:

The percentages are a convention taken from the normal distribution, not measured shares of any particular market. They are still useful because they show how different the people are who carry an innovation in each phase.

In 1991 the consultant Geoffrey Moore added an idea in Crossing the Chasm that is often quoted for bitcoin: there is a gap between the early adopters and the early majority. Early adopters buy a vision; the early majority wants a finished, convenient solution that people like them already use. Many products get stuck at exactly this point.

How many people own bitcoin? The estimates

Nobody can count bitcoin owners. The blockchain shows addresses, not people. One person often has many addresses, and a single exchange address can hold coins for millions of customers. Someone holding an ETF share in a brokerage account appears in no address at all. Every figure is therefore an estimate built from exchange data, surveys and assumptions. Why active addresses are no substitute for users is explained in detail in the article on Metcalfe's law.

The estimates that can be sourced (as of November 2026):

Three things stand out. First, the worldwide estimates are of the same order, about 4 to 4.5% of roughly 8.2 billion people. Second, adoption is clearly higher in rich countries: in Switzerland and the US, almost one in five holds crypto or bitcoin. Third, the surveys measure different things, sometimes bitcoin, sometimes all crypto assets, sometimes people, sometimes households. The numbers are therefore not directly comparable.

The Chainalysis Global Crypto Adoption Index is also widely quoted. It does not measure a share of the population, though; it ranks 151 countries by transaction activity. In 2025 India and the US came first. For the question of where bitcoin stands on the S-curve, it only hints at where usage is growing.

Bitcoin compared with the internet and mobile phones

The obvious question: if bitcoin stands at around 4%, when was the internet there? The International Telecommunication Union (ITU) has published the share of the world's population using the internet for decades.

At 4%, bitcoin would be roughly where the internet stood in the late 1990s. The internet crossed into the early majority (16%) around 2005 and reached half the world around 2018 to 2019.

Mobile phones were faster still. According to the ITU there were 738 million mobile subscriptions worldwide in 2000 and more than 7 billion in 2015, almost as many as people. In rich countries mobile was saturated early; later growth came from Africa and Asia.

The comparison limps at one important point. The internet and the phone spread because they were immediately useful in daily life and kept getting cheaper. For most holders, bitcoin is primarily a store of value. You don't need one every day, and its use only shows over years. The curve may well be slower for that reason. It may just as well be much faster in countries with high inflation than in Switzerland. How different it can be is shown by El Salvador, where a law tried to force its use and actual uptake still stayed small.

The S-curve in numbers: the logistic model and the Bass model

Mathematically, the S-curve is usually described with the logistic function. It has three dials: the ceiling (how many people take part in the end), the speed, and the moment when half is reached.

The Bass model, published by marketing researcher Frank Bass in 1969, additionally separates two drivers: people who come on their own (innovation, through advertising or their own curiosity) and people who come because others already have it (imitation). A review of many products by Sultan, Farley and Lehmann (1990) found an average innovation coefficient of 0.03 and an imitation coefficient of 0.38. The imitation effect is typically more than ten times as strong. For bitcoin, where people mostly get in through friends, colleagues or family, that fits well.

This calculation is a thought experiment, not a forecast. It assumes that in the end almost everyone holds bitcoin and that bitcoin spreads as fast as the internet did. Both are open questions.

S-curve and Power Law: two different shapes of growth

The Bitcoin Power Law describes the price as a power of the time since the genesis block. The article on Metcalfe's law shows how to connect this to users: if value grows with the square of the users, the number of users has to grow roughly with time to the power of 2.8 for the site's formula to work. We won't repeat that bridge here. The question here is a different one: does that kind of user growth fit an S-curve?

The difference lies in how the annual growth rate changes.

• On the S-curve, the number of users grows at an almost fixed percentage at first, that is exponentially. The rate only falls noticeably once a large part of the ceiling is reached, and then drops towards zero.
• Under the Power Law, the rate falls evenly from the start, because each new year is a smaller fraction of the time that has passed. It never reaches zero, though, and there is no ceiling.

For decades the two lines stay surprisingly close, as long as the S-curve is given the internet's pace. They only part when adoption nears saturation: the S-curve comes to a standstill, the Power Law keeps growing, if more slowly. With a different speed or a lower ceiling, for example if only half of all people ever hold bitcoin, the lines separate much earlier.

A reality check: Crypto.com estimates that the number of bitcoin owners grew 8.3% in 2025, from 337 to 365 million. That is less than the roughly 17% the Power Law requires via Metcalfe. And in 2026 the price did sit well below the trend, as the article on criticism of the Power Law shows. It is no proof, because the price also depends on how much each owner holds, and companies, funds and states appear in no head count.

Which adoption matters for the price?

With bitcoin, "adoption" is several curves at once, and each is at a different point.

People. About 4% of the world, between 10 and 20% in rich countries. Most with small amounts.

Companies. Listed companies with bitcoin on the balance sheet are a small group that holds large amounts. Who they are and how much they hold is shown in the overview of treasury companies.

Funds. Spot ETFs have existed in the US since January 2024. People can hold bitcoin through them without ever opening a wallet. For the S-curve, this means some of the new owners show up only in surveys, not on the blockchain.

States. A few states hold bitcoin from seizures or buy it deliberately. That is the earliest phase of the curve, measured in innovator percentages.

What matters for the price is not how many people own a fraction of a bitcoin, but how much capital in total wants to hold bitcoin. A curve of heads can flatten while the curve of capital still rises steeply, for example when pension funds follow private individuals. That is one reason why a saturated user count need not immediately end price growth.

What the S-curve means for you as a long-term holder

Three conclusions can be drawn from the model without trusting it too much.

The steepest phase probably still lies ahead. If bitcoin follows an S-curve and stands at 4%, the fastest growth in people comes when the early majority gets in. That is no promise about the price: the Power Law still expects percentage price gains to shrink every year.

The curve can stall. Moore's chasm is real. Many products never reach the early majority. For bitcoin, a lot depends on whether custody, inheritance and taxes become simple enough for ordinary people.

Saturation is the real test of the Power Law. A model that never reaches a ceiling must at some point be slowed by a network with a finite number of people. Once adoption approaches saturation at the latest, the curve will flatten below the Power Law line. How the Power Law deals with its own limits and what it gives for the coming years is shown in the bitcoin forecast based on the Power Law; you can work through individual years in the forecast calculator.