In June 2023, Reddit announced that it would begin charging third-party developers for API access. The decision, framed by leadership as a straightforward business move, was meant to capture value that had long been flowing freely out of the platform.
What followed was thousands of subreddits going dark in coordinated protest.
Volunteer moderators, people who had spent years, in some cases the better part of a decade, managing communities without compensation, walked out. The backlash was so swift and so large that observers who weren’t paying close attention to the platform’s internal culture could be forgiven for finding it disproportionate.
Volunteer moderators, people who had spent years, in some cases the better part of a decade, managing communities without compensation, walked out. The backlash was so swift and so large that observers who weren’t paying close attention to the platform’s internal culture could be forgiven for finding it disproportionate.
It’s important to note that Reddit was charging developers, not users. The price was debatable, but it wasn’t outrageous.
So why did so many people react as though something had been taken from them?
While pricing and communication were part of the backlash, they don’t fully explain the intensity of the response. The moderators weren’t being charged directly, but they relied on third-party apps built on the free API to do their moderation work: tools for blocking spam, flagging abuse, and managing communities on mobile. When Reddit priced those apps out of existence, the moderators lost the tools they depended on to do their unpaid jobs.
The protest was both practical and moral.
They couldn’t function without those tools, and they felt Reddit had broken its end of an unwritten agreement with the people who had sustained it for free.
The moderators who left weren’t doing a cost-benefit analysis. They were responding to a shift in the terms of a relationship, one that had never been formalized, never been written down, yet governed their behavior for years.
Understanding why that shift felt like a betrayal, and why it produced the response it did, requires looking at research that most community builders never seek out.
It also requires being honest about a problem that runs much deeper than pricing strategy.
The psychological contract.
Denise Rousseau, an organizational psychologist at Carnegie Mellon, spent decades studying what she called the psychological contract. The unwritten set of expectations people develop about what they owe an organization, and what the organization owes them, built from what people have come to expect based on how things have always worked.¹
Her foundational research made something clear that sounds simple but consistently gets underestimated:
People treat unwritten expectations as real agreements.¹
These psychological contracts are specific, deeply held beliefs about how the relationship works. When an organization acts in a way that breaks them, people don’t feel mildly inconvenienced, they feel betrayed.¹
The anger, the disengagement, the withdrawal; these responses show up with a high degree of reliability, even when nothing was ever formally promised.¹
Reddit’s moderators had one of these contracts with the platform and its terms were clear from years of behavior on both sides.
Moderators gave their time and expertise, Reddit provided the platform and the audience, and money never entered the picture.
It should come as no surprise that this isn’t a Reddit-specific problem.
This happens in communities of every size whenever a builder introduces money into a relationship that formed without it. Most of the time, there’s no protest and no headlines, just a slow erosion of the thing that made the community worth being in. The reason it keeps happening is that most builders don’t have language for what they’re disrupting.
There are four ways humans relate to each other, and only one involves money.
To understand why introducing a monetary layer into something that was free can be so disruptive, it helps to understand what kind of relationship most free communities are actually built on.
Alan Fiske, an anthropologist at UCLA, spent years looking at how people across many different cultures structure their relationships. His conclusion was that every human relationship, regardless of culture, runs on one of four basic models.²
- In communal sharing, people treat everyone in the group as equal and give freely without keeping score. The group itself is what matters, and generosity is just the default way of operating.²
- In authority ranking, relationships follow a clear chain of command. People higher up are deferred to, and everyone accepts that asymmetry.²
- In equality matching, people track what they give and what they get, and try to stay balanced over time.²
- In market pricing, everything comes down to a ratio: what am I getting for what I’m putting in, and does that trade make sense?²
Most relationships draw on more than one of these models at once. A family primarily runs on communal sharing, but shifts to equality matching when splitting an inheritance. A workplace runs on authority ranking most of the time, but market pricing activates the moment bonuses come up. The models aren’t rigid boxes and are highly context-dependent, and the same two people can move between them depending on the situation.²
In any given moment, one model tends to be dominant, and that shapes how people interpret everything that happens within the relationship.²
This is part of what makes introducing money so disruptive.
It doesn’t just add something new. It shifts which model people are operating from and directly maps onto how motivation works.
Communal sharing runs on intrinsic motivation.
People show up because of who they are and what they care about, not because of what they’re getting back.²
Market pricing shifts behavior toward extrinsic motivation.
People pay to receive something, and when what they’re receiving no longer feels worth the cost, they stop paying.²
Most communities that start free settle into communal sharing by default. Members show up without being asked, tell their friends, and tolerate imperfections they would never accept if they were paying. That behavior is a direct result of the relational model in place. Change the model, intentionally or not, and the behavior changes with it.²
When money enters a relationship, the rules change.
James Heyman and Dan Ariely tested this in a series of experiments called Effort for Payment: A Tale of Two Markets, published in Psychological Science in 2004.³
When no money is involved, people give generously and don’t track what they get back. When money enters the picture, people start calculating. They expect to get something back that matches what they contributed.
However, the most important finding wasn’t about pure social or pure monetary relationships. It was about what happens when you mix the two.
When a relationship has both social and monetary elements, it doesn’t settle somewhere in the middle. It tends to behave like the monetary one.³
Money doesn’t just add something new on top of what already exists. It often changes which rules the whole relationship runs on.
Heyman and Ariely found that people sometimes put in more effort when they receive nothing at all than when they receive a small payment.³ The introduction of even a low payment was enough to shift the relationship from social to monetary, and with that shift came less generosity, less discretionary effort, and less intrinsic motivation to participate.³
This is why donation models can coexist with communal sharing in ways that subscription models often can’t.
A donation is an expression of belonging.
A subscription is a transaction.
The direction of the money matters as much as its presence.³
Reddit was an example of this. The moderators who left weren’t looking at a compensation offer and deciding it wasn’t enough. They had been showing up for years because they belonged to something. When Reddit introduced money into that relationship, it didn’t just change the economics; it changed what kind of relationship it was.
Unsurprisingly, Reddit didn’t reverse the API pricing changes. The protest lasted through the summer, but by August 2023 the last major holdouts had relented.
Reddit won.
CEO Steve Huffman called the protesting moderators “landed gentry” and threatened to replace those who kept their communities dark.
Reddit followed through, removing moderation teams from some subreddits.
The third-party apps that moderators had depended on, including Apollo, shut down permanently.
Many moderators stayed because they had no real leverage, not because the trust was restored.
The relationship wasn’t just disrupted. It was permanently changed.
Charging isn’t the problem. When you charge is.
Communities cost resources to run. Time, infrastructure, effort — none of it is free. Expecting community builders to absorb that indefinitely, on goodwill alone, isn’t realistic nor fair.
There is nothing wrong with charging for something you built.
The problem is introducing a price into a relationship that was never designed for one.
Research on organizational imprinting found that the conditions a community starts under leave a lasting mark, long after those original conditions have changed.⁴ The norms and expectations that form early on don’t just describe the community in its early days; they become the standard everything else gets measured against.⁴
Members don’t just experience those founding conditions; they internalize them, bring others in around them, and build their psychological contracts on top of them, so that when something changes, what people are reacting to is the gap between what they expected and what’s happening now.⁴
That gap can feel very large even when the actual change seems small.
Long game >
So what does this mean in practice?
The science explains what happens but losing the gap between knowing it and actually building for it comes down to three decisions worth considering before your first member joins.
The first is what relational model you are building on.
Are you building a communal sharing community — open, contribution-driven, free — or one where market pricing is part of the structure from the start? Both can work. But you need to choose deliberately, because your members will choose for you if you don’t. They will arrive, form expectations, build psychological contracts around whatever they find, and those contracts will be very hard to renegotiate later.¹²
The second is whether you plan to charge, and when.
If the answer is eventually, make that visible from the start. A detailed pricing page isn’t necessary, but honest communication about what you are building and where you envision this going, is.
Members who join knowing a paid tier is coming form a different psychological contract than members who join assuming the relationship is permanently free.¹ Founding conditions shape long-term expectations. What you signal at the start is what people build their participation around, and what they later evaluate for fairness.⁴
The third is what you are charging for.
Avoid putting a price on something that was free.
If you want (or need) to charge for something, build a paid offering separately so you’re adding something new, not changing the terms of something that already exists.³
It’s important to note that none of this requires you to have every single answer before you launch. However it does require being honest with yourself and with your members about what you are building and what your community will eventually need to sustain itself.
Balancing altruism and architecture.
There’s a version of this argument against monetizing communities, and our counter argument is that it’s really just asking builders to be generous forever with no regard to the human energy that it takes to run and scale communities.
That argument doesn’t hold up against the reality of what it takes to build and sustain something, and it’s not what this research is saying.
Charging isn’t the problem. The problem is the timing:
If you plan to build something sustainable—and most serious builders eventually need revenue to do it—you have to design for that early. You have to do it before relational norms set in, before psychological contracts form, and before the founding imprint takes hold.
Introducing market pricing into an established social market is consistently harder than it looks, and the longer you wait, the harder it gets.³
Reddit didn’t have a pricing problem in 2023.
It had an architectural problem that started years earlier, when it built everything on communal labor under communal norms and never planned for the moment it would need to introduce market terms.
By the time that moment came, the psychological contracts were too established, the relational model too entrenched, the imprint too deep to change without the kind of response it got.¹²⁴
The community didn’t only reject the price. It rejected the shift in what kind of relationship Reddit thought it was in with the people who had built it.
The builders who avoid that problem aren’t the ones who avoid revenue. They’re the ones who understand early enough that the relational model and the revenue model have to be designed together because one will always shape the other, whether you plan for it or not.
Sources:
¹ Rousseau, D.M. (1989). Psychological and Implied Contracts in Organizations. Employee Responsibilities and Rights Journal, 2(2), 121–139.
² Fiske, A.P. (1992). The Four Elementary Forms of Sociality: Framework for a Unified Theory of Social Relations. Psychological Review, 99(4), 689–723.
³ Heyman, J. & Ariely, D. (2004). Effort for Payment: A Tale of Two Markets. Psychological Science, 15(11), 787–793.
⁴ Stinchcombe, A.L. (1965). Social Structure and Organizations. In J.G. March (Ed.), Handbook of Organizations. Rand McNally.