Mutual, Follow, and Broadcast Ties: Three Ways Platforms Structure Social Networks

Three relationship structures, three different questions

A social platform can display several kinds of relationships that look similar on screen but represent different network structures. A friendship connection, a follow button, and a subscription to a broadcast account all indicate some form of connection. However, they do not answer the same question. Friendship usually indicates mutual recognition, following records a direction of attention, and broadcasting describes how information reaches an audience.

For social network analysis, the important distinction is what each relationship permits and what it records. A network edge is not simply a decorative line between two accounts. Its direction, reciprocity, and meaning determine which questions the graph can answer. Treating every platform connection as an interchangeable friendship risks producing misleading conclusions about influence, community, and social cohesion.

Consider three accounts: A, B, and C. A and B are mutual friends, B follows C, and C publishes information that reaches both A and B. These relationships form three different structures even though all three accounts participate in the same platform environment.

• Mutual: both participants recognize the connection.
• Follow: one account directs attention toward another.
• Broadcast: one publisher distributes information to an audience.

Mutual ties and undirected network graphs

A mutual relationship exists when two accounts establish a connection that requires recognition or acceptance from both sides. A conventional friendship network is a familiar example. If A and B are friends, the relationship is usually represented as an undirected edge. The line between them has no arrow because the connection is treated as reciprocal.

In graph notation, an undirected relationship between A and B can be written as A-B. The graph does not need two separate edges to represent the same friendship. This makes it useful for studying connected groups, network density, and the number of relationships surrounding an account.

Suppose a fictional study group has five members and four mutual friendship connections. Its undirected graph contains five nodes and four edges. The degree of a node is the number of connections touching it. A member with three connections has a degree of three, indicating that three other members are connected to that person.

Mutual connections are useful for examining social cohesion, but reciprocity should not be confused with emotional closeness. Two people can be connected through a platform's friendship mechanism without being close friends. The graph describes the platform-defined relationship, while claims about trust, intimacy, or support require additional evidence.

Follow relationships and directed ties

A follow relationship has a direction. If A follows B, the connection begins at A and points toward B. B does not necessarily follow A. This is a directed tie, represented by an arrow from A to B. In a directed graph, A to B and B to A are separate possible relationships.

The distinction matters because incoming and outgoing connections describe different forms of network position. Out-degree counts the number of accounts an individual follows. In-degree counts the number of accounts that follow that individual. An account with an out-degree of 40 follows 40 accounts, while an account with an in-degree of 40 has 40 followers.

For example, imagine a researcher studying a fictional community of six accounts. One account follows all five others, but none of the five follows it. Its out-degree is five and its in-degree is zero. Replacing those directed edges with undirected connections would erase this asymmetry and incorrectly suggest mutual relationships.

A follow graph can therefore help analyze attention, information-seeking, and potential exposure pathways. However, a follower count alone does not establish how frequently someone sees content, whether they read it, or whether they respond to it. Platform distribution rules and individual behavior remain separate parts of the explanation.

Broadcast networks and audience relationships

Broadcast structures organize communication around a publisher or source and the people who receive its content. A newspaper account, public information channel, or creator profile can distribute messages to many followers without requiring a reciprocal connection with each audience member. The resulting structure often resembles a hub with numerous outward-facing or inward-facing relationships, depending on the direction used in the data model.

For example, a fictional public information account publishes updates followed by 100 accounts. A directed follow graph records 100 incoming edges to the publisher. The publisher can distribute a single message to that audience, but the graph does not show that the 100 followers communicate with one another. Their shared connection is to the source, not necessarily to each other.

Broadcast communication is therefore different from a densely connected community. A large audience can coexist with very few relationships among audience members. This distinction is useful when studying public communication, creator economies, institutional accounts, and information distribution.

The term parasocial relationship describes a perceived interpersonal relationship in which an audience member experiences a sense of connection with a media figure or character without an equivalent reciprocal relationship. A follower may feel familiarity with a creator who does not personally know that follower. This psychological experience should not be inferred from a follow edge alone. The network records a platform relationship; parasocial attachment requires evidence about audience perceptions.

One diagram for comparing the three structures

A small fictional diagram can make the differences visible. Start with three nodes labeled A, B, and C. Draw a plain line between A and B to represent a mutual connection. Draw an arrow from B to C to represent a follow relationship. Finally, represent C as a publisher with arrows pointing from C toward a separate group of audience nodes to illustrate message distribution.

For a reproducible teaching example, use five nodes: A, B, C, D, and E. Record A-B as a mutual friendship, B-C as a directed follow, and C-D and C-E as broadcast distribution relationships. Keep the relationship type in a separate edge attribute. This prevents the drawing from suggesting that all four connections have identical social meanings.

A simple data dictionary can include source, target, and relationship_type. The source identifies where a directed edge begins, the target identifies where it ends, and relationship_type records whether the connection is mutual, follow, or broadcast. An undirected friendship can be stored as one normalized pair, provided the graph-building instructions explicitly treat it as undirected.

The diagram can be redrawn from a CSV edge list using Python and a graph library such as NetworkX. The essential modeling decision comes before the visualization: friendship edges should be undirected, while follow and broadcast edges should retain their direction.

• A-B, mutual, undirected.
• B-C, follow, directed from B to C.
• C-D, broadcast, directed from C to D.
• C-E, broadcast, directed from C to E.

Choosing the correct representation for analysis

The appropriate network model depends on the research question. An investigation of reciprocal friendship should preserve mutuality. An analysis of attention or account discovery should retain follow direction. A study of public communication should distinguish publishers from audience members and identify the relationships through which messages are distributed.

A platform may combine these structures or use terminology that differs from conventional network sociology. Some products allow one-way following alongside mutual friendship, while others use subscriptions, memberships, or other connection mechanisms. Researchers should consult the relevant platform documentation and inspect the available relationship fields before assigning graph meanings.

The next practical step is to create a three-column edge list containing source, target, and relationship_type. Add five fictional accounts, assign each connection its correct direction and reciprocity, and draw the resulting graph with separate visual conventions for mutual, follow, and broadcast ties.

Questions readers ask