Understand the algorithm behind Xiaohongshu recommendation engine

If a brand wants to gain traffic on the platform and realize the conversion gains, the first step is to understand the traffic distribution logic of the platform. This is because there is algorithm logic behind every action, for example, publishing a post (also known as ‘note’), writing a review, platform recommendation, user interaction, etc.

Xiaohongshu’s logic of traffic driving algorithm mainly has two core factors: content tag and social recommendation

Content tags mean that the content will be marked with a series of tags by the platform after it is published, and it will be recommended to those who may be interested, while the keywords, geographical location, and other information extracted from the posts are the main keywords of the tags. After the algorithm promotes the post to a certain number of users, the platform will score the posts quality according to the amount of interaction of the posts, and decide whether to continue recommending them to more users. This scoring system is called CES internally in Xiaohongshu.

The calculation of CES = number of likes * 1 point + number of favorites * 1 point + number of comments * 4 points + number of reposts * 4 points + number of followers * 8 points.

The quality of the content of the post will directly affect the amount of interaction of the post. The higher the score of the post, the higher the possibility of the post being continuously recommended, so as to obtain long-term and wide range of traffic. In addition to that, the closer the relationship between fans and celebrities, the higher the probability of the celebrities will appear in the “following” and “discover” sections of the fan’s homepage. When fans interact with internet celebrities frequently, the celebrity's posts will be further pushed to friends of fans, thus forming brand cultivation.

The brand needs to understand their users’ mindset, have e-commerce thinking, and refine the operation on Xiaohongshu. In addition to understanding the needs of the current product and competing products in the market, brands need to clarify their own goals, whether you want to transform or promote the brand, and do a good job in the early stage of marketing strategy. Down to the operation, the social media manager should create content by understanding their users combined with the characteristics of the influencers, and then fine-tune the strategy according to the data that is collected to achieve the best results.