How much reading does Zhihu have to harvest?

Zhihu is an online Q&A community, which belongs to Beijing Wise Man Tianxia Technology Co., Ltd., launched on 2011/and was founded by Kelly Y Zhou.

Zhihu has gathered the most creative people in the fields of Internet technology, commerce, film and television, fashion and culture in China. It has become a comprehensive, all-category, knowledge-sharing community with key influence in many fields and an original content platform where creators gather, and established a community-driven content realization business model.

Based on the question-and-answer business, it covers a series of products and services such as "question-and-answer" community, brand-new member service system "salt-selected members", organization number and popular list. , and established multimedia forms including graphics, text, audio and video. The mission of the brand is to "let people better share knowledge, experience and opinions and find their own answers".

Basic rules of video revenue:

(1), video revenue is mainly related to the amount of play, followed by factors such as likes and comments.

(2) The video cannot violate the rules, otherwise the income of the corresponding video will be deducted; Violation is nothing more than advertising, discrimination, brushing, infringement and so on.

Video traffic source:

On the PC side, video does not support search, which means that there is no video content in the interface after searching for keywords.

On the mobile side, only the video column can be selected for keyword search, but the search traffic of this column is not large.

Zhihu's video traffic mainly depends on platform recommendation, and the recommendation rules are basically similar to Tik Tok and Aauto.

First, the content matches the user, and the corresponding video is distributed to the corresponding user.

When the video is released, you need to choose the field and add related topics, that is, label the video.

As for users, they will be labeled accordingly when they see more content in the usual search.

After matching, it is the first wave of recommendations, not many, usually hundreds; Then look at the factors such as broadcast rate, like, collection, etc., among which the broadcast rate is relatively high, and then see whether to continue recommending according to these data.