Building an influencer community is like founding a new country. Early TikTok attracted creators the way nations draw migrants with economic policies: First, it rewards a few users to demonstrate upward mobility and rapid successes, then spreads opportunities to grow a broad middle class.
Video: Musical ly's Alex Zhu on Igniting Viral Growth and Building a User Community 2016
How can an early social media platform solve the classic marketplace chicken-and-egg problem? Viewers want great content, but new creators need proof their effort will pay off, especially when considering a move to an unknown app.
Early TikTok viewed building an influencer community as building a new country and economy from the ground up. You want to grow the population by attracting people to migrate to your country.
Established platforms like Instagram and YouTube are like Europe, where social hierarchies are entrenched. The average citizen of Europe has almost zero opportunity to move upward in the social class. On the other hand, TikTok positioned itself like America, a new land promising opportunity and upward mobility.
Their strategy is a planned economy with a two-phase plan.
Phase one: Centralize fame.
In this new land, you must build a centralized economy in the early days.
The algorithm funneled traffic to a small percentage of people in your land. You ensure they successfully build an audience and wealth, making them role models for the country. You effectively create the American dream. For example, overnight successes like Baby Ariel served as living ads.
People in Europe (Instagram and YouTube) will start to realize that this "normal" person went to America (TikTok) and became super rich. Maybe I can do the same? Fear of missing out will lead to many people migrating to your country.
Phase two: Decentralize reach.
Having an American dream is good, but people will wake up eventually. Therefore, you have to decentralize the economy once the supply of hopeful newcomers grows.
Ranking rules shifted toward short-term engagement signals and random exploratory boosts. This let many clips land on the For You feed, giving most users a hit of validation. A middle class of semi-popular accounts kept uploading, which kept viewers scrolling and data flowing back into better recommendations.
Using this lens to examine the initial algorithm for all the new social media platforms is interesting. Do you think Threads use the same strategy?