The "Just-Right Buzz" in Community — The 70% Rule

The trap of “let’s make it more lively”

Every community operator has felt it: “I want more posts,” “I want members to be more active,” “we need to liven things up.”

This impulse is natural. But there are situations where engagement campaigns backfire. Under certain conditions, pushing for more content and encouraging members to post more can actually accelerate departures.

Why does this happen? The answer lies in a fact: every community has an optimal level of buzz density.

Three theories that all point to “70%”

Three independent theoretical frameworks converge on the same optimal density range of roughly 70%. Let us walk through each.

Foundation 1: Queuing theory — Kingman’s formula

Queuing theory is a mathematical framework applied to systems where demand and processing capacity meet — service counters, servers, roads. One of the most well-known results in this field, Kingman’s approximation, states:

As utilization rate $\rho$ approaches 1.0 (100%), wait time grows sharply — theoretically, without bound.

A restaurant analogy makes this intuitive. A restaurant at 50% capacity: you get a seat immediately. At 80%: a short wait but manageable. At 98%: wait time to get a single seat spikes dramatically.

A community’s “information processing capacity” follows the same structure. Here we introduce the concept of $v_{max}$ (maximum throughput): the maximum number of messages a member can comfortably follow per channel per day. The ratio of actual message flow to this limit is utilization rate $\rho$. As utilization approaches $v_{max}$, the stress from unread content increases exponentially.

The theoretical safe operating ceiling is roughly 70–80% utilization.

Foundation 2: Flow theory — Csikszentmihalyi’s optimal experience

Psychologist Mihaly Csikszentmihalyi’s Flow theory describes the conditions under which people enter their most fulfilling state of immersion.

Flow emerges in the zone where skill level and task difficulty are matched. Too easy: boredom. Too hard: anxiety and exhaustion. Translated to community:

• Too little information (sparse): “Nothing’s happening, nothing to say” — a state of boredom
• Too much information (crowded): “I can’t keep up, I’m tired” — a state of anxiety and exhaustion
• Just the right amount: “There’s enough going on that I can join in” — a flow state

The flow state — between boredom and exhaustion — is the participant’s felt sense of “just-right buzz.” Within the flow theory framework, this balance point corresponds to roughly 70–80% of cognitive load capacity.

Foundation 3: The community temperature bell curve

When community “temperature” is expressed as a function of spatial density $\rho$, it traces a bell-shaped curve. Each variable is defined as follows:

• $T(\rho)$: community temperature at density $\rho$ (participant engagement intensity and immersion)
• $T_{max}$: peak temperature (maximum energy when the community is at its liveliest)
• $\rho_{opt}$: optimal spatial density (the density at which temperature peaks)
• $\sigma$: tolerance width (how far $\rho$ can deviate from $\rho_{opt}$ before temperature noticeably drops)

$$T(\rho) = T_{max} \cdot \exp\left(-\frac{(\rho - \rho_{opt})^2}{2\sigma^2}\right)$$

What this model shows is that temperature peaks ($T_{max}$) only when $\rho$ matches $\rho_{opt}$, and drops in either direction from that point.

Theoretical and empirical analysis places this $\rho_{opt}$ at approximately 0.6–0.8 — consistent with the queuing theory and flow theory findings above.

The convergence of three independent frameworks on the same range is what gives the “70% rule” its explanatory power.

Why 100% breaks down and 50% is boring

It might seem intuitive that a full house means a lively community. The reality is the opposite.

Why 100% (full capacity) breaks down

As Kingman’s formula shows, utilization near 100% causes wait times — unread messages, unanswered posts — to pile up rapidly. Members repeatedly encounter “more than I can read” every time they open the community, and gradually stop opening it at all.

Conversation threads also get buried more frequently, breaking context. Replies to specific members go unnoticed, questions get left without answers, and relationship-building becomes difficult. What looks like high activity may be a state where no one can actually have a conversation.

Why 50% (half capacity) is boring

At the other extreme, low density creates a different problem. Post something and no one responds for hours; a single reply eventually trickles in. Members start to wonder: “Is there any point in talking here?”

In flow theory terms, this is “the task is too easy — boredom.” Without stimulation, attention drifts away. A community that is too sparse quietly disappears from members’ awareness.

The optimal density shifts with community “type”

The 70% figure is a general starting point. The optimal density shifts up or down depending on a community’s purpose, culture, and member profile.

Purpose-driven communities ($\rho_{opt} \approx 0.8$–$0.9$)

DAOs (decentralized autonomous organizations), open-source software communities, and learning communities — groups that share a concrete problem or goal — fall here.

Members of these communities have a higher tolerance for tracking information, and active discussion actively improves output quality. Even at slightly higher density, the motivation “I don’t want to miss an important discussion” keeps members engaged. As a result, the optimal density ceiling is somewhat higher: $\rho_{opt} \approx 0.8$–$0.9$ still functions healthily.

Third-place communities ($\rho_{opt} \approx 0.4$–$0.6$)

Hobby communities, fan communities, casual chat spaces — groups whose primary purpose is comfort, belonging, and light interaction — fall here.

For these communities, the experience of “I couldn’t keep up” is often fatal. Members are seeking the feeling of “I dropped in and it was fun,” and too much information makes the space feel heavy and demanding. A lower optimal density of $\rho_{opt} \approx 0.4$–$0.6$ is the healthy range.

Communities in between

Most real communities sit somewhere between these two archetypes. The useful question is: “What do members primarily come here for?” and “How heavyweight is the typical conversation?” Communities that mix light socializing and serious discussion can also benefit from setting different density targets for different channels.

Measuring your community’s density

Understanding the theory is one thing; measuring your own community is another. Here is a straightforward calculation.

The spatial density formula

$$\rho = \frac{\text{daily posts per channel}}{v_{max}}$$

$v_{max}$ is “the maximum number of messages a member can comfortably follow per channel per day.” A common rule of thumb is 30 posts/day/channel. This figure is grounded in Sweller’s (1988) cognitive load theory and Miller’s (1956) finding that working memory processes roughly 7±2 chunks at a time — translated to an empirical upper limit at roughly 1–2 posts per hour across a 2–3 hour daily usage window (Yamamoto 2026).

Example calculations

Community	Channels	Weekly posts	Daily/ch	$\rho$	Assessment
Discord, 100 members, 10 channels	10	700	10	0.33	Leaning sparse
Discord, 100 members, 10 channels	10	2,100	30	1.0	At the optimal ceiling
Slack, 50 members, 5 channels	5	3,500	100	3.3	Clearly crowded

Formula: $\rho$ = weekly posts ÷ 7 ÷ channels ÷ 30

Checking weekly whether $\rho$ falls in the 0.5–1.0 range gives a practical starting point for density management.

Operational decisions for staying near 70%

Knowing the optimal density is one thing; maintaining it is another. Here are the patterns where density drifts and the responses for each.

Sparse ($\rho < 0.5$): actions to take

Causes: too many channels, too few members, no trigger for posting.

Responses:

• Consolidate channels to concentrate flow (reverse over-fragmentation)
• Provide regular posting triggers (questions, prompts, topic drops from operators)
• Intentionally funnel traffic toward a hub channel

Adding new members first seems like the obvious fix, but if the community is sparse because of too many channels, adding people without consolidating channels will not improve density.

Crowded ($\rho > 1.0$): actions to take

Causes: too few channels, flow concentrated in one place, temporary burst from an event or release.

Responses:

• Use threads actively — the “pressure-release valve”: When traffic on a topic threatens to exceed $v_{max}$ in the main channel, redirect that discussion to a dedicated thread. The main channel becomes a “headline feed” while deep discussion branches off. Just as a pressure-relief valve in a pipe opens automatically when pressure rises, threads release the overflow before the main channel becomes a torrent no one can follow.
• Spread posting time across the day (morning and evening prompts to reduce midday peaks)
• For temporary bursts, “wait it out” is often the right judgment

Adding channels looks like the obvious solution but tends to cause sparse conditions over time. Treat channel creation as a last resort.

Revising the target density

As a community matures, member profiles and purpose tend to shift. A community that launched with purpose-driven intensity may gradually become a third-place community over years. The $\rho_{opt}$ target is not fixed — revisiting it every six to twelve months in line with how the community has evolved is ideal.

Summary

• Every community has an optimal spatial density ($\rho_{opt} \approx 0.6$–$0.8$); participant experience degrades if density falls too far above or below that range
• Three independent theories — queuing theory, flow theory, and the community temperature function — all converge on roughly 70%
• 100% (crowded) produces information overload, exhaustion, and broken conversation; 50% (sparse) produces boredom and quiet departure
• Optimal density shifts by community type: purpose-driven communities (0.8–0.9), third-place communities (0.4–0.6)
• The goal of community operations is not maximization but optimization — density is the lens that clarifies which direction to push

Related articles

• The Real Reason “Buzz” Disappears — Sparse and Crowded Are the Same Disease
• MAU Will Kill Your Community — Leading, Middle, and Lagging Indicators
• How to Choose and Operate Community KPIs

References

• Yamamoto, H. (2026). A Mathematical Model of the Online Public Sphere. §4.1, §5.2.
• Csikszentmihalyi, M. (1990). Flow: The Psychology of Optimal Experience. Harper & Row.
• Kingman, J. F. C. (1961). “The single server queue in heavy traffic.” Mathematical Proceedings of the Cambridge Philosophical Society, 57(4), 902–904.
• Miller, G. A. (1956). “The magical number seven, plus or minus two.” Psychological Review, 63(2), 81–97.
• Sweller, J. (1988). “Cognitive load during problem solving: Effects on learning.” Cognitive Science, 12(2), 257–285.