Home
Help Center
Xiaohongshu Traffic Knowledge: A Complete Guide from Mechanism to Practice

Xiaohongshu Traffic Knowledge: A Complete Guide from Mechanism to Practice

Help Center2026-04-25

Understanding Xiaohongshu's traffic logic is the first step to successful operations. While the algorithm may seem complex, grasping its core principles can effectively boost your content's exposure.

1. Traffic Sources: Dual Track of Recommendation & Search
Xiaohongshu traffic mainly comes from two types. Recommendation traffic (Discovery Page) accounts for about 60% of total traffic. The system proactively pushes content based on users' interest tags, making it the primary channel for gaining initial cold-start exposure. Search traffic accounts for 30-70% — users actively search with keywords, offering strong long-tail effects that allow quality content to continuously drive conversions.

In terms of search, a key change in 2026 is the addition of an "intent recognition" module — the algorithm can distinguish between "informational searches" (e.g., "how to remove blackheads") and "transactional searches" (e.g., "blackhead remover recommendations"), prioritizing content that best matches the searcher's intent. Meanwhile, the AI semantic understanding system can now parse deep semantic relationships across text, images, and video simultaneously. This means notes that used to rely on keyword stuffing for traffic are now flagged as low-quality. The algorithm favors naturally expressed, clearly focused content. Brands that want to win the battle for search traffic must combine keyword SEO with search ads to capture high-intent audiences.

2. CES Score: The Core Metric Determining Traffic
CES (Community Engagement Score) is at the heart of Xiaohongshu's traffic distribution mechanism, directly determining whether a note enters a larger traffic pool. The 2026 updated formula is:
1 like + 1 save + 4 comments + 4 shares + 8 follows

As the formula shows, the platform has significantly increased the weight of comments, shares, and follows — these "strong engagement signals" are worth 4 to 8 times more than likes. Simply racking up likes is useless. Content that genuinely sparks thoughtful comments and active shares is what the algorithm recognizes as "high-quality content." Additionally, long comments (thoughtful comments over 15 characters) carry more weight, and behaviors like screenshotting or saving content are also considered extremely high-value signals. Therefore, the focus of content design should not be "how to get people to like" but "how to get people to spend time reading, leave thoughtful comments, save, and want to come back."

3. Tiered Traffic Pool: A Layer-by-Layer Filtering Mechanism
Xiaohongshu uses a tiered traffic pool system to filter content. After publishing, notes go through multiple stages of testing. The initial exposure is about 100-500 views, and the system decides whether to expand recommendations based on click-through rate (CTR) and engagement rate. The 2026 key thresholds are:

Image-text notes: CTR ≥ 2.5%

Video notes: CTR ≥ 4%

Engagement rate: ≥ 5%

Completion rate (video): > 40%

It's worth noting that performance in the first hour is critical — the system uses a real-time race mechanism, comparing content against similar pieces every 5, 30, and 60 minutes. Only winners move to the next traffic tier. This means actively encouraging initial engagement right after posting is a golden window for breaking through traffic levels. Moreover, in 2026, Xiaohongshu has further prioritized authenticity and uniqueness as core criteria for content recommendation. Signals like user dwell time and interaction authenticity are used to uate content authenticity. Notes that meet authenticity standards can even jump directly into a high-traffic pool. Additionally, video notes receive clear traffic support — their initial exposure is 30% higher than image-text notes, and videos with a completion rate above 50% are eligible for a second round of recommendation.

4. Pitfalls to Avoid
Keep an eye out for several common pitfalls:

"Hard-sell" phrasing: The system can now identify unapproved commercial content through multi-dimensional signals (semantics, images, etc.) and will directly limit its distribution.

Ignoring search keywords: The first 10 characters of your title should include the primary keyword, which should naturally appear 3-5 times in the body, combined with precise hashtag strategies.

Information overload on the cover, leading to a low CTR.

Inconsistent posting frequency: Long gaps in posting will lower your account's authority.

In summary, gaining traffic on Xiaohongshu requires continuous optimization across three areas: content quality, engagement guidance, and keyword layout. While the traffic mechanism evolves, returning to user value and focusing on authentic content is always the foundation for success. It's recommended to focus on the golden 2 hours after publishing to drive high-weight interactions, systematically avoid algorithm pitfalls, and continuously refine your content strategy based on data feedback.

Core Services

News

Xiaohongshu Traffic Knowledge: A Complete Guide from Mechanism to Practice