Imagine you work at a 100‑person company. You come in on Monday, and during your first meeting, you notice that Mary, a woman with whom you frequently interact, has begun using a new piece of software. You recall that the company had pushed for employees to start using the software several months ago, but organizational enthusiasm seemed to have dissipated quickly. You shrug and chalk it up to chance; maybe Mary’s manager is getting pressure from her boss to get people to use it or something?

After lunch, you decide to go to visit your friend in accounting before heading back to your desk. As you arrive, you happen to notice that they, too, are using the software right before you enter their office. “Hmm, that’s a bit peculiar,” you think to yourself, but again you brush it off as a random coincidence.

But as you’re leaving for the day, you stop by a colleague’s cubicle so they can show you the restaurant that they had been talking about during an earlier meeting. They lean over their desk and as they search their screen for the right tab you see an open window with what you recognize as the logo of the same software.

You leave the office wondering why everyone seems to suddenly be using this software. Is there renewed pressure from leadership? Was there a memo I missed? Or maybe it just is a really useful tool? Although you can’t pinpoint the exact reason, you walk in on Tuesday and open the new software to give it a shot. About 20 minutes later, a colleague comes in to ask a question about the client meeting later that day and they see you using it.

And this is precisely how new norms begin to spread.

Conventional wisdom is that you need a majority to bring about change; 51% of people have to be doing something new or embracing something different to get the entire group to ultimately shift. But this isn’t necessarily the case. The above example was about such change happening by chance, but we also have empirical evidence to suggest that, if you want to deliberately produce it, it still takes far less than a majority to trigger a formidable wave of change.

In 2018, Damon Centola and colleagues at the University of Pennsylvania published a study in Science that probed this very phenomenon: how large does a committed minority have to be to flip a group norm? They ran online experiments where participants would have brief one-on-one interactions with one another in a virtual environment (think of it as a “digital hallway,” where you’d bump into a “co-worker,” speak with them for a few moments, and then move on to interact with another coworker).

During the study, participants were asked to come up with a name for a new project, and these one-on-one interactions were an opportunity to “pitch” one another on the new name. Eventually, the group - through repeated interactions - would come to a consensus on a name they all agreed with. In other words, a “norm” would form.

But then came the twist. In later rounds, the researchers quietly seeded each group with a “committed minority”: confederates (i.e., supposed “participants” who, in reality, were working with the researchers) who would always use a new, different name for the project and, critically, refused to switch back to the name the group had originally converged upon, no matter the social pressure. Across a number of trials, the size of this minority varied between approximately 15% and 35% of the group.

What happened was remarkably consistent:

• Below roughly 25%, the minority’s impact was generally insignificant. Only about 6% of real participants temporarily adopted the new name, and most of those who did reverted to the old norm soon after. In other words, the system absorbed the influence the minority attempted to exert and returned to baseline.

• However, around 25%, dynamics seemed to flip. Conversions (i.e., adoptees of the new name) accelerated and then reinforced one another. In many runs, the new name quickly propagated to near‑total adoption in a short period of time. In one case, adding a single additional committed person abruptly pushed the group over the edge, from stalemate to a complete cascade of change.

In short, the tipping point did not require a majority; instead, it actually seemed to hover right around one quarter of the total group (far less than conventional wisdom would suggest).

There are two important distinctions to note here:

1. There’s a difference between an actual majority (i.e., >50% of the group) and what can feel like a majority. In the hypothetical example we began with, statistically, you only observed just 3% of the entire company using a new software. But because they happened to be three of only a handful of people you interacted with that day, it felt to you like “everyone” had started to make this change, and that convinced you to do so, as well.

2. Humans are not just sensitive to static norms (i.e., what everyone is or seems to be doing) but also dynamic norms (i.e., how norms seem to be changing over time). In other words, we observe not only what’s typical now, but also how “typical” is changing. When people see a behavior gaining traction, they adjust toward it, anticipating where the group will eventually land. When we perceive momentum to be shifting, we tend to try to shift in accordance with it. For example, research has shown that simply advertising how norms around vegan and vegetarian lifestyles were accelerating doubled the number of meatless orders at a cafe (e.g., Sparkman & Walton, 2017). Dynamic norms are powerful because when individuals sense momentum - even before a majority forms - they become more inclined to change.

Now, 25% isn’t some magic constant; it’s not like if you get to 25%, change will definitely follow. The actual percentage will move depending on how entrenched the current norm is and how costly deviation feels. Yet across conditions, the threshold to catalyze formidable change seems to stay well below the 51% mark that most expect.

Consequently, we can say that change is governed less by majority rule than by perceived momentum. Committed, visible minorities can precipitate powerful cascades of change, especially if they operate in small pockets where people begin to feel like “everyone” around them seems to be shifting.