Glossary
C
Community Mobilization
Community mobilization refers to the structured activation of interest graph–driven communities through repeatable systems such as challenges, affiliate programs, gifting campaigns, and in-product shareables.
Unlike organic community behavior, mobilization is brand-led and strategically engineered to trigger user participation that aligns with platform-native behaviors. Mobilization transforms individual engagement into distributed behavior, increasing visibility, building cultural relevance, and converting community interest into algorithmic momentum.
Collaborative Filtering
Collaborative filtering is an algorithmic method used within interest graph systems to group users and content based on shared behavioral patterns rather than direct preferences.
Instead of recommending content based solely on what a user has already engaged with, the system identifies other users with similar interest profiles and surfaces content those users found engaging. This approach allows platforms to infer latent interests, diversify recommendation pools, and increase the likelihood of content discovery beyond a user’s known engagement history.
D
Discovery Layer
A new structural tier that sits above awareness, consideration, and conversion.
It represents the algorithms, recommendation systems, and community dynamics that determine what enters cultural visibility in the first place.
F
Formats
A format is a repeatable content structure that organizes how information or behavior is presented in a platform-native way.
Examples include GRWM videos, “What I Eat in a Day,” or tutorial step-throughs. Formats reduce creative friction, guide audience expectations, and help the algorithm classify and surface content.
E
Earned Social Media
Earned social media refers to brand visibility generated through organic creator participation, where multiple individuals voluntarily create or share content that features, references, or builds on a brand.
It functions as the social-first counterpart to earned media in public relations, which measures unpaid press or editorial coverage. In interest graph ecosystems, earned social media reflects cultural uptake, not distribution strategy. Its strength is measured not by how often a brand posts, but by how often others do.
I
Interest Graph
The interest graph is a recommendation model that organizes users and content based on behavioral alignment rather than social connection.
It is built by tracking patterns in what users watch, like, save, or share—and is used by platforms to cluster content and communities based on interests, not relationships.
Interest Graph Platforms
Interest graph platforms are digital environments—such as TikTok, YouTube Shorts, and Instagram Reels—that surface content based on user behavior rather than social following.
These platforms are optimized for discovery, not distribution. They reward repeatable behaviors, prioritize content structure over creator identity, and are where most culture-led product discovery now occurs.
Interest Graph–Driven Communities
IGDCs are user clusters formed by shared content behaviors, not social relationships, and are surfaced by platform algorithms based on observed engagement patterns.
These communities form based on shared interests rather than social relationships, and they operate around repeated structures such as formats, rituals, trends, and memes. IGDCs are central to how discovery and cultural organization now occur on platforms like TikTok, Instagram Reels, and YouTube Shorts.
L
Latent Interests
Latent interests are inferred user preferences that have not been explicitly expressed through engagement, but are predicted based on behavioral correlations within the interest graph.
Platforms like TikTok and Douyin use collaborative filtering and tag-level association to identify potential interests based on what similar users engage with. These inferred interests are tested through content exposure; if a user engages positively, the latent interest may be promoted to a known interest. Latent interests enable the platform to expand content recommendations, accelerate discovery, and surface emerging affinities before the user actively seeks them out.
M
Memes
A meme is a symbolic, culturally-recognized content unit that spreads through remixing, reference, and repetition.
Memes can be visual, audio, or textual and often serve as shorthand for community beliefs, in-jokes, or stances. In IGDCs, memes are an organizing mechanism for shared language and social bonding.
R
Referential Content
Referential content describes the dominant content behavior on TikTok and similar interest graph platforms, where most videos build on pre-existing formats, trends, or cultural patterns—often unintentionally.
Rather than inventing from scratch, users typically recreate, adapt, or echo recognizable content structures already circulating in the ecosystem. This behavior is shaped by algorithmic incentives and cultural norms, making referentiality a default mode of creation rather than a strategic choice.
Rituals
A ritual is a recurring behavioral pattern or themed content moment that occurs at regular intervals within a community.
Examples include “Sunday Reset,” “Weigh-In Wednesday,” or “Monthly Favorites.” Rituals reinforce community identity, create predictable posting cycles, and give users a low-friction way to participate.
S
Self-Organized Movements
Self-organized movements are cultural behaviors that emerge and spread through intentional community participation, often originating outside of platforms but accelerating within interest graph ecosystems.
These movements are not algorithmically surfaced but socially constructed—driven by shared identity, rituals, or offline behavior. Examples include the rise of run clubs, themed concert outfits, or participatory cultural moments like Cowboy Carter styling. Once they enter short-form platforms, these movements often become structured through formats, trends, and referential content—making them legible to the algorithm and enabling scale. While the origin may be ambiguous, their traction in IGDCs is shaped by how well they align with existing participatory structures.
Serialized Content
Serialized content is a content strategy where creators deliver related pieces in a repeated or structured sequence to build familiarity, anticipation, or retention.
This structure increases watch time, encourages return behavior, and aligns with the platform’s interest-based recommendation logic. Serialized content is a core mechanic for growth within IGDCs.
Episodic
Episodic content is a type of serialized content where each installment follows a consistent format or structure, but does not rely on a chronological storyline.
Examples include Keith Lee’s food reviews, “What’s in My Bag?” videos, or weekly GRWM series. Each piece can stand alone but works better in sequence due to audience familiarity with the format.
Sequential
Sequential content is a type of serialized content that unfolds across multiple, ordered parts and typically depends on the viewer watching in a specific sequence.
Examples include “Storytime Part 1, 2, 3,” or Reesa Teesa’s “Who TF Did I Marry?” series. This format capitalizes on curiosity and narrative tension to encourage binge behavior.
Short-Form Video
Short-form video refers to mobile-first, vertical video content typically under 90 seconds in length.
It is the default format on platforms like TikTok, Instagram Reels, and YouTube Shorts, where content spreads through interest-based recommendations. Short-form video accelerates trend cycles and prioritizes engagement signals like watch time, format adherence, and replication behavior.
Social-First Theory
Social-First Theory holds that cultural relevance, product discovery, and early-stage growth are now driven by behavior inside interest graph–driven ecosystems.
Under this theory, brands must prioritize participation within user-led communities—by aligning with platform-native structures—to be visible, credible, and scalable in modern digital environments.
T
Talk Value
Talk value refers to the human-driven amplification of content through social interactions such as tagging friends, sharing posts, and commenting in ways that invite visibility beyond algorithmic recommendation.
It reflects the degree to which content travels because people want others to see it, not just because platforms surface it. High talk value increases the likelihood that content will be passed from user to user—intentionally—through direct conversation, commentary, or call-outs. This metric is key to understanding how culture circulates organically within IGDCs and complements algorithmic visibility with socially driven distribution.
Trends
A trend is a time-bound content wave marked by high engagement around a specific structure, audio, or behavior.
Trends emerge when participation converges around a theme, and they often scale rapidly. They serve as growth opportunities for creators and brands to temporarily ride shared attention.