Dunbar's Number: How to Manage More Than 150 Relationships

Quarter after quarter, the same thing happens. A handful of important relationships quietly cool. Someone gets promoted and you miss it. A warm contact goes quiet. A promise to reconnect next month turns into silence that lasts a year. It is not negligence. It is math.

Anthropologist Robin Dunbar's research on the relationship between primate neocortex size and social group size produced a finding that has held up across decades of scrutiny: humans can maintain roughly 150 stable relationships at once. Dunbar's number is not a hard ceiling, but it is a practical constraint, created by the cognitive load of maintaining relationships and the time it takes to keep them warm.

The problem is that most professional contexts require far more than 150 active relationships. Founders managing investors, advisors, and candidates simultaneously. BD teams running conversations with dozens of partners. Recruiters keeping a bench of strong candidates warm between active searches. The Dunbar limit explains why these professionals constantly feel like they are losing ground on their network, no matter how much effort they put in. Hustle is not the variable. The system is.

This article explains what Dunbar's number actually means, why it shows up so clearly in professional relationship management, and how a systematic approach to contact management can extend what any individual can maintain.

What Dunbar's number actually means

Robin Dunbar's original 1992 paper in the Journal of Human Evolution established the link between neocortex size and typical group size across primate species. When extrapolated to humans, the model predicted a natural group size of approximately 150 stable relationships. The finding has been replicated and extended since.

The number is not a single wall but a series of concentric layers. Dunbar's research identifies roughly five intimate contacts, 15 close ones, 50 good relationships, and 150 meaningful ones. The layers reflect the different levels of maintenance required: intimate relationships require regular, high-quality contact; the outer layers can be maintained on lower frequency. In Dunbar's 2024 review in Annals of Human Biology, he reaffirmed that these layered limits remain a robust empirical pattern across diverse populations and contexts.

Large-scale digital data supports the model. Gonçalves, Perra, and Vespignani's 2011 analysis of Twitter conversations, published in PLOS ONE, found that people maintain a relatively constant number of active, reciprocal ties (roughly 100 to 200) even on high-velocity platforms. Digital tools create more access to people; they do not remove the cognitive cost of maintaining real relationships.

There is legitimate debate about the precision of the number. Lindenfors's 2021 paper in Biology Letters argues for significant individual variability and critiques the statistical methodology behind the 150 estimate. Dunbar's response has been that layered network limits remain a robust pattern even if the precise figure varies. The practical takeaway, as Maria Konnikova summarised in the New Yorker, is to treat 150 as a design constraint, not a fixed rule. The real bottlenecks are attention and time, and those do not expand no matter how connected the tools you use.

The concept of weak ties is relevant here. Sociologist Mark Granovetter's research established that weak ties, the outer layers of your network, are often more valuable for novel information and new opportunities than strong ties in your close circle. The people you know less well are plugged into different worlds than you are. Dunbar's number describes how many of these ties you can realistically maintain before relationships start to decay. The goal is not to eliminate the outer layers but to maintain enough signal with them that the relationship stays alive.

Who runs into the Dunbar limit

The constraint is universal, but it becomes most visible in professional contexts where maintaining a broad, active network is not optional, it is the job.

Founders and operators managing a network of investors, advisors, candidates, and partners are routinely above 150 active professional relationships before the company reaches Series A. Each relationship requires some level of maintenance: an update, a follow-up, a congratulations when something changes. At scale, this is not manageable on memory alone.

Sales and business development teams whose deals start with relationship capital, not cold outreach. A BD motion typically involves keeping dozens of partner conversations warm across different stages and timelines. The cost of letting one cool at the wrong moment is a missed deal.

Recruiters keeping a bench of strong candidates between active searches. Recruiting from relationships means maintaining contact with people who are not actively looking, over months or years, until the right role opens. That requires a systematic approach to staying in touch without the cadence becoming purely transactional.

Consultants and advisors whose professional network is their primary asset. Every referral, every new engagement, every introduction begins somewhere in the relationship network. The quality and breadth of that network determines the quality and breadth of opportunities.

Partnership and community managers who are professionally responsible for the health of a large, varied network. These roles involve tracking relationship status across dozens or hundreds of contacts simultaneously, often across a team, which makes the individual Dunbar limit less relevant than the team's collective capacity.

Why hustle alone does not extend it

The typical response to feeling behind on relationship management is to work harder: more emails, more messages, more outreach. The problem is that attention and time are the actual bottlenecks, and hustle does not expand either.

What hustle produces, in practice, is generic outreach. A "hope you're doing well" message sent to 20 people at once is not relationship maintenance. It is noise. The people who matter can tell the difference between a message written for them and a message written for everyone. Generic outreach at scale erodes the relationship rather than maintaining it.

The other cost of hustle without a system is asymmetry: the people you happen to think of get contact; the people you do not happen to think of slip. The warmest, most strategically important contact in your outer circle can go quiet for two years not because you stopped valuing the relationship but because nothing surfaced them at the right moment.

The warmth-at-scale approach to relationship maintenance addresses this directly: a defined list of relationships that matter, a sense of which ones are active and which are drifting, and a lightweight cadence for reaching out before things go quiet. The system does the remembering; you do the relationship work.

How systems extend the Dunbar limit

The practical extension of Dunbar's number is not cognitive; it is operational. A well-designed relationship management system handles the maintenance overhead that cognitive limits cannot, freeing your attention for the relationships themselves.

Surface the right moment to reach out. The most powerful outreach moments are tied to real events: a contact changes roles, a company milestone is announced, someone you know moves into a position that is directly relevant to your work. These are not generic check-ins; they are specific, timely reasons to be in touch. Detecting them manually across a large professional network is not feasible. A system that monitors your contacts for job changes and title moves and surfaces them in your daily workflow removes the detection burden and converts outer-layer relationships into timely, specific conversations. In Rolodex, Title Alerts handle this automatically across your contact network.

Set a follow-up cadence without manual tracking. The outer layers of Dunbar's model require low-frequency but consistent maintenance: a quarterly check-in for important contacts, a twice-yearly note for friendly investors, an annual touch for alumni and advisors. The problem is not remembering that a cadence matters; it is remembering to execute it for every contact, every time, without letting anyone slip. A contact management system that supports scheduled follow-up reminders per contact, grouped by relationship type, removes the execution overhead. You define the cadence once; the system surfaces who needs attention each week. Rolodex's Keep in Touch feature is built exactly for this: set a cadence by segment, and the contacts who need attention surface at the right time without manual tracking.

Preserve relationship context across a large network. As the network grows past what memory can reliably hold, context starts slipping. Who introduced you to whom. What you discussed last time. What the contact was working on when you last spoke. Without a shared record, every return conversation requires rebuilding context from scratch, which is slow and signals to the other person that the relationship was not maintained. A personal CRM that logs interaction history, notes, and context per contact keeps that information accessible before every conversation. In Rolodex, the Activity Feed shows the full relationship history in order, so you can step back into any contact's context in under two minutes.

Share the relationship maintenance work across the team. This is the most underused lever. The Dunbar limit applies to individuals, not teams. A founding team of three, each with a Dunbar-scale professional network, collectively has access to 450 relationships. A BD team of five has access to 750. But those networks only function as a collective asset if the context is shared. When each person's contacts live in their own inbox and memory, the team functions at individual Dunbar capacity rather than team capacity. A shared relationship management system, where everyone's interaction history, introductions, and notes are visible to the whole team, is how small teams maintain the kind of network that previously required a much larger organisation. Rolodex is built for this: one shared view of the team's contacts, history, and relationship context, accessible to everyone.

A weekly practice for managing more than 150 relationships

The system-level changes above address the structural problem. The practice-level habit is what makes them work consistently.

A 30-minute weekly ritual covers the maintenance overhead for most professional networks. Start with five minutes on signals: check for any title changes or role moves among your contacts and send two or three specific, timely responses. These moments are your best re-engagement opportunities, and they take under a minute each when the signal has already surfaced.

Spend ten minutes on your follow-up cadence: clear the week's Keep in Touch queue and send a relevant one-liner to each contact due for a check-in. This is not about writing long emails. It is about maintaining signal in outer-layer relationships before they go quiet. A specific, two-sentence note that references something real is more effective than a thoughtful paragraph that arrives six months late.

Use ten minutes to advance any warm introduction requests currently in progress: reply to the introducer, propose meeting times, or thank them for a meeting that has already happened. A shared board of introduction requests with named owners and clear statuses keeps this from falling through the cracks in a team context. Then spend five minutes on one contact record: pull the current state of the relationship into a note so anyone on the team can step back into context cold.

Thirty minutes per week, run consistently over a quarter, produces a noticeably different network. More timely responses. Fewer missed moments. A broader ring of relationships that stay warm because the system maintained them, not because memory happened to surface them.

The takeaway

Dunbar's number is not a problem to solve. It is a design constraint to work with. Attention and time are finite. The relationships that matter most need consistent maintenance. A system that handles the overhead of timing, context, and team coordination extends what any individual, or any team, can sustain.

The path beyond the Dunbar limit is not more hustle. It is better infrastructure: a shared relationship management tool that surfaces the right moment, keeps the context intact, and lets a small team collectively maintain a network that would otherwise exceed what any individual could manage alone.

If your professional relationships quietly cool despite good intentions, the problem is the system, not the effort.

Start managing your professional network with Rolodex, free to start

Book a demo to see how Rolodex handles relationship management at scale