Ellison and Resnick: Replicating TikTok’s algorithm is harder than it looks

As ByteDance prepares to sell TikTok’s U.S. operations,replicating the recommendation algorithm — the secret behind its famed ‘For You’ page — may prove difficult. 

For Business Insider, University of Michigan School of Information professors Nicole Ellison and Paul Resnick, experts in social media and recommendation systems, say duplicating TikTok’s content recommendation engine won’t be simple. 

Ellison, who has coauthored two papers on TikTok’s algorithm, says even referring to the system as a single “algorithm” oversimplifies its complexity. She says such systems are often governed by "very complicated computational formulas that look at many, many, many data points about an individual, including their past behavior in a particular online space" to predict what type of content they will engage with or enjoy.

Resnick adds that the people behind the code are just as important as the code itself. Even if ByteDance offered "the whole code base and said, 'Yep, it's yours,' if it didn't come with any people who were involved in creating that code base, it would be very, very hard to make good use of it,” he says. 

RELATED

Read “TikTok insiders and creators worry its powerful algorithm could lose its magic after a sale” in Business Insider. 

Learn more about Nicole Ellison and Paul Resnick’s research on social media, algorithms and online engagement by visiting their UMSI faculty profiles.