Tik Tok: Users are not interested in “people to follow”, how to promote social interaction?

One of the focuses of Douyin 's future product and operational directions is to establish a stronger "following" relationship, strengthen interaction, promote social interaction , and help Douyin's content producers monetize.

1. A problem with Douyin

Tik Tok is an amazing product and is growing very fast, but as a social platform, it has a relatively weak “follow system” and “social interaction” problem.

I think there is this problem, on the one hand because I saw Mr. Pan Luan said that " Kuaishou 's average VV and average number of followers are both higher than Douyin", and on the other hand, it is a phenomenon I have seen:

Nowadays, more Tik Tok users browse the “recommendation stream” than the “follow stream” . Many people will follow others but never click on the “Follow” tab .

The direct source of this is my observation of the Douyin users around me (including myself) - open Douyin, ignore the "99+" on the follow tab, and see that many Douyin users with 200,000 or 300,000 followers often have works that have less than 100 likes.

2. Causes of the problem

There are two reasons that lead to this problem:

1. On TikTok, the efficiency of content “algorithm distribution is much higher than social relationship distribution.”
2. Douyin's original product design and operation tendencies ignored "individual characteristics" and "interaction".

1. Let’s first look at the issue of “ content distribution efficiency”

It has been proven that ordinary (mass) users do not have the ability to follow a set of accounts that they "really like". The method of obtaining content through "subscription" can only be achieved by highly educated people.

Unless it is an institutional account, it is impossible for ordinary people to output high-quality content in a continuous and stable manner.

After users discover an account in the recommendation stream and follow it, they find that it is “not as good as it was at first” and the signal-to-noise ratio in the follow-up stream is too low. In my own experience, the things in the follow tab are not as good as those in the recommendations.

2. Let’s talk about the original product design and operation tendencies

• The earliest attention stream was listed in the "Home" tab alongside "Recommendations" and "Same City", and later became a separate tab;
• In the early product design, whether it was recommendation or attention, you could directly watch new videos by swiping down, without any further operation (compared to Kuaishou videos, swiping down to the comment area);
• Early seed users were the "coolest people," and most of the earliest videos were of the "technical flow" camera movements and hand gesture dances. These videos only made people feel awesome, but they didn't make people want to interact.
• Toutiao's early operations ignored individual characteristics. It used the "challenge" (topic) mechanism to educate users, but it also led to everyone using only a few sets of audio to shoot similar things, and the content was too templated. Such "topics" occupied a tab in the early days, which resulted in "sameness" and made users lose the desire to pay attention;
• In terms of algorithm mechanism, audio should account for a large proportion, and coupled with the early operational tendency, it will give users a "gust of wind" feeling - for a while it is always seaweed dance, and for a while it is everywhere "there is a little brother in front, let's go and flirt with him"

It’s not that this kind of product design and operation method is bad - it gives users a good immersive viewing experience and effectively educates users. It’s just that there is no such thing as having the best of both worlds. We can only try to strike a balance between "better consumer experience" and "promoting social interaction."

3. Impact of this issue

So, what is the impact of such a problem?

In a nutshell, it will have an impact on community stickiness and the content ecology of the community:

1. If you don’t follow or interact, you don’t need to carefully maintain an account system;
2. Weak attention relationships and interactions will make it difficult for creators to gain profits from fans, and the creators will be more concerned about how to get recommended rather than making better works that fans will like. For example, there will be meaningless content such as "TikTok has a new bug, you can't click the like button at this position";
3. Fans are worthless, and in the long run there will be problems on the supply side.

4. How to solve this problem

Tik Tok is created by a really great team. They discovered the problem very early and tried to solve it.

1. What has the TikTok team done?

1. Fold the "Challenge" function - weaken the impact of manual operations on content orientation and achieve diversification;
2. Set up a separate "Follow" tab;
3. Give more priority to displaying content from “people you follow” in the recommendation algorithm;
4. The story function is visible only to fans, deleted after 24 hours, has no complex special effects, and facilitates lightweight interactions;
5. The comment area will prioritize comments from people you follow;
6. The format of the Follow tab has been greatly changed. In addition to optimizing the viewing experience, it is also designed to promote more interactions. The like and comment buttons have been changed to a " friend circle-like " style, and highly liked comments are displayed directly (without clicking).
7. Some other things I didn't notice...

2. I think there are some things that can be done

The following are irresponsible suggestions:

1. Add more things that give content producers (anchors) personal characteristics and attributes, such as tags, achievements, etc. For example - "fitness expert", "technical expert, etc."
2. Increase “exclusive” interaction methods between anchors and fans. This story counts as one, but it seems that no one is using it now, and there is no noise (this is my observation, if the data is different, just treat it as nonsense). Other options include "only fans can comment" and "fans can send gifts to each other";
3. Design a fan growth system. Interactive actions against anchors can increase fan value. This fan value can be reflected in priority display in the comment area, obtaining exclusive medals, etc. In terms of operations, it can also be given priority to get tickets for fan meetings;
4. Help anchors better monetize their fans - that is, make fans as valuable as possible, and then "motivate anchors to guide fans to interact with them." What I’m doing now is shopping guide, but I don’t know why I can’t see the “shopping guide button” on the focus interface. This includes fan recommendations (similar to Weibo fan headlines), and ( game ) performance advertising can also be cooperated with anchors.

5. Final Thoughts

I think the core of this issue is still short video products - the efficiency of algorithm distribution is higher than that of social relationship distribution .

As algorithms become increasingly intelligent and mass product content is being pushed to the grassroots, this is somewhat irreversible, so this problem has no solution to a certain extent and can only be alleviated .

The author of this article @何雷 is compiled and published by (Qinggua Media). Please indicate the author information and source when reprinting!

Product promotion services: APP promotion services, advertising platform, Longyou Games