XHS Marketing Analytics: The Latest Tools & Techniques for International Brands
Date Published
Table Of Contents
1. Why Analytics on Xiaohongshu Is Different
2. The CES Scoring System: How XHS Actually Ranks Your Content
3. Key Metrics That Actually Signal ROI on XHS
4. Native Analytics Tools Built Into the Platform
5. Third-Party Analytics Tools for Deeper Intelligence
6. Search Analytics: The Most Underused Edge on XHS
7. Measuring KOL and KOC Campaign Performance
8. Attribution Techniques for XHS's Multi-Touch Journey
9. Building Your XHS Analytics Framework Step by Step
10. Common Mistakes Brands Make When Measuring XHS Performance
If you've been running campaigns on Xiaohongshu (XHS) and struggling to connect your content performance to actual business outcomes, you're not alone. Most international brands enter the platform using analytics frameworks built for Instagram or Meta — and then wonder why the numbers don't tell a coherent story.
Xiaohongshu is not Instagram. It's not a pure social platform, and it's not a pure e-commerce site. It's a content-search-commerce hybrid where a post can drive zero immediate sales but quietly rank in search results for eight months, steadily feeding purchase intent. That means conventional metrics like reach and likes are not just incomplete — they can be actively misleading.
This guide breaks down the latest tools and techniques for XHS marketing analytics in 2026: the native platform instruments you should be using, the third-party solutions worth your budget, the CES scoring framework that determines how your content actually performs, and the search intelligence strategies most brands are still ignoring. Whether you're just establishing your analytics baseline or looking to build a more sophisticated measurement system, this is where to start.
Why Analytics on Xiaohongshu Is Different {#why-analytics-different}
Before diving into tools and frameworks, it helps to understand the structural reason why standard analytics approaches fall short on XHS. Most Western social platforms distribute content primarily through a recommendation feed — your post performs well in the first 24–48 hours, then loses momentum. Xiaohongshu operates on two independent engines simultaneously: a discovery feed driven by the algorithm's content recommendations, and a search engine driven by user intent. A post that underperforms in the discovery feed can still sit at the top of a high-intent search query for months.
This dual-engine reality changes everything about how you measure success. A piece of content with modest engagement metrics at launch might be doing significant work in the background — capturing searches from users actively researching your product category. Conversely, a post that goes viral in the feed might reflect strong entertainment value rather than purchase intent. Understanding which engine your content is winning, and why, is the foundational task of XHS analytics in 2026.
Layered on top of this is a cultural dynamic that matters enormously for international brands. Chinese consumers on Xiaohongshu have developed a sophisticated distrust of obvious advertising. The platform's community rewards authentic, peer-style content — and the algorithm reflects this. Metrics that look strong on paper (high views, high likes) can mask weak performance if users aren't saving posts, clicking through, or mentioning your brand in their own content. For a deeper look at how to navigate these nuances across different product categories, explore AllXHS's industry-specific Xiaohongshu marketing strategies.
---
The CES Scoring System: How XHS Actually Ranks Your Content {#ces-scoring}
One of the most important technical concepts for XHS analytics is the platform's Content Engagement Score (CES) framework. Rather than treating all engagement equally, Xiaohongshu weights different interaction types according to their signal value. The scoring system works as follows: likes earn 1 point, saves (collections) earn 1 point, comments earn 4 points, shares earn 4 points, and new follows from a post earn 8 points.
The weighting logic reflects the platform's priorities. A like is a passive signal — low effort, low commitment. A save signals that a user intends to return to your content, most commonly before making a purchase decision. Comments and shares require active effort and extend your content's reach. A follow is the strongest possible endorsement, indicating that a user wants a long-term relationship with your account. This is why brands obsessing over like counts are systematically misreading their own performance.
For your internal reporting, build CES calculations into your content scorecards. A post with 5,000 likes but 50 saves may have a lower CES than a post with 1,000 likes and 400 saves. The latter is doing far more work in terms of purchase consideration and algorithm favor. Track CES trends across content types, posting times, and topic clusters to identify what your audience actually values versus what they passively scroll past.
---
Key Metrics That Actually Signal ROI on XHS {#key-metrics}
With the CES framework in mind, here are the specific metrics that experienced XHS marketers prioritize for ROI measurement:
Save Rate (收藏率): The ratio of saves to views is the single most reliable early indicator of purchase intent on Xiaohongshu. For beauty and fashion categories, a healthy save-to-view ratio indicates that users are actively considering the product rather than passively consuming the content. High saves with low immediate conversions are not a failure — they represent users in the consideration phase who will convert over a longer window.
Search Traffic Percentage (搜索流量占比): This metric, visible in your professional account dashboard, tells you what proportion of your content views came from search queries versus the recommendation feed. A growing share of search traffic signals that your content is winning on intent — users are actively looking for what you're publishing. In 2026, search drives a significant portion of content discovery on the platform, and search-driven users convert at substantially higher rates than passive recommendation feed visitors.
Comment Quality and Sentiment (评论质量): The volume of comments is a CES signal, but the content of comments is your qualitative intelligence layer. Users on Xiaohongshu commonly use the comment section to ask purchase-intent questions: "Where can I buy this?" "Does this work for sensitive skin?" "Is it available outside China?" These comments are the closest thing to a real-time customer survey you can get — mine them systematically.
Follower Growth from Content (内容涨粉): When a specific post drives a notable spike in follower growth, it signals that users found the content valuable enough to want more from your account. This metric connects individual content performance to long-term audience building and compounds over time in ways that one-off viral posts do not.
Cost Per Engagement (CPE) for KOL Campaigns: For influencer investment specifically, CPE (calculated as KOL fee divided by total engagements) is a more useful benchmark than CPM alone. A CPE below ¥10 is generally considered solid, with elite performers delivering under ¥5. Track this alongside Cost Per Reading (CPR) to get a complete picture of both reach efficiency and engagement quality.
---
Native Analytics Tools Built Into the Platform {#native-tools}
Before investing in any third-party solution, international brands should fully leverage the three native analytics instruments that Xiaohongshu provides.
The Creator/Brand Professional Account Dashboard (专业号后台) is the starting point for any analytics setup. Accessible through verified professional accounts, it provides content-level data including views, impressions, engagement breakdowns, follower growth, traffic source splits (search vs. discovery), and audience demographic profiles. The traffic source data is particularly valuable — it allows you to monitor in real time whether your SEO efforts are translating into search visibility gains. To access the full suite of commercial analytics features, your account needs to be verified as an official brand account.
Lingxi (灵犀) is Xiaohongshu's brand marketing analytics center and arguably the most powerful native tool available to commercial accounts. Lingxi allows brands to analyze campaign performance across paid and organic content, optimize ad targeting, and gain visibility into the consumer journey from first content exposure to purchase. The platform has been progressively expanding Lingxi's capabilities, with a focus on making campaign measurement more transparent — addressing one of the platform's historically weakest areas for international brands. For SMBs and mid-market brands, Lingxi now offers insights that were previously only available to major enterprise accounts with dedicated platform support.
Pugongying (蒲公英) is Xiaohongshu's official creator marketplace and collaboration platform, with over 100,000 verified creators. From an analytics perspective, Pugongying is invaluable because it gives brands access to creator performance data before committing to a partnership — including historical cost per engagement, audience demographics, content category performance, and audience overlap with other brands. In late 2025, Pugongying enhanced its toolset with an Intelligent KOL Finder that uses AI matching across multiple data dimensions to surface high-fit creators for specific campaign briefs. This removes much of the guesswork from influencer selection and creates a more defensible data trail for ROI measurement post-campaign.
---
Third-Party Analytics Tools for Deeper Intelligence {#third-party-tools}
Native tools provide a solid foundation, but brands running serious XHS programs will benefit from supplementing with specialized third-party platforms.
XinHong Data (新红数据) is a Xiaohongshu-verified official data partner built by Newrank China. It functions as a comprehensive analytics hub, offering deeper content performance insights than the Creator Centre, competitive benchmarking, trend tracking, and influencer vetting capabilities. For brands that want a single platform to monitor their own content performance alongside competitor and category-level trends, XinHong Data is one of the most purpose-built options available.
AnyTag by AnyMind Group extended its platform to support Xiaohongshu in early 2025, following its recognition as a first-tier Xiaohongshu partner. AnyTag allows brands to run influencer discovery, campaign management, and campaign analytics for XHS within a unified interface — useful for teams managing multi-market campaigns across several Asian platforms simultaneously.
Social listening tools such as KAWO and WalktheChat provide broader monitoring of brand mentions and category conversations across Chinese platforms, including Xiaohongshu. For brands that need to track how their products are being discussed organically (not just in their own content), social listening adds an essential qualitative layer to quantitative performance data.
Custom API and data pipelines are increasingly being explored by more technically sophisticated teams. Tools like XHS Insights provide programmatic access to note data, comment analysis, and creator profiles for research workflows, enabling brands to build custom dashboards that integrate XHS performance data with broader marketing attribution systems.
For a curated set of ready-to-use tools and templates across these use cases, AllXHS's free resources library covers 25+ instruments built specifically for international brands on Xiaohongshu.
---
Search Analytics: The Most Underused Edge on XHS {#search-analytics}
Search analytics is where most international brands leave significant opportunity on the table. In 2026, Xiaohongshu's search engine has matured significantly — it no longer relies on simple keyword matching. The platform now uses intent-based semantic understanding, surfacing content that addresses the underlying question behind a query even when exact keywords don't match. A user searching "dry skin routine" will encounter content about hydration, sensitive skin care, and seasonal skincare — not just posts containing the literal phrase.
This evolution has important implications for analytics. Tracking which search terms are driving traffic to your content is no longer just an SEO task — it's a direct window into your audience's purchase journey. When you see a specific long-tail query repeatedly driving saves and comment interactions, you've identified a content-commerce opportunity worth doubling down on.
For keyword intelligence, use a layered approach: start with Xiaohongshu's own search bar autocomplete (which surfaces real user queries), then analyze the title patterns and hashtag structures of top-performing competitor posts in your category, then mine the comment sections of high-engagement posts for the organic language your potential customers use when discussing your product space. This triangulated approach to keyword research tends to surface higher-converting long-tail terms than any single-source method.
The 70/30 principle is a useful framework for keyword allocation: aim for roughly 70% of your content to target long-tail, lower-competition search terms (such as "sensitive skin moisturiser for beginners") and 30% to target broader trending keywords. Long-tail content is easier to rank and typically serves users further along in the consideration process — making it disproportionately valuable for conversion.
---
Measuring KOL and KOC Campaign Performance {#kol-koc-performance}
Influencer investment is often the largest variable cost in an XHS marketing budget, which makes rigorous KOL and KOC performance tracking non-negotiable. The metrics that matter differ depending on whether you're running awareness-focused KOL campaigns or trust-building KOC seeding programs.
For KOL campaigns focused on reach and brand lift, prioritize impressions, CES score, share of voice in your category, and brand search volume lift following the campaign. Top-performing KOL content should ideally deliver CPM rates in the ¥50–200 range depending on niche and audience quality, with engagement rates of 5% or above considered healthy and 10%+ elite.
For KOC seeding campaigns, the quality of engagement matters more than volume. Focus on the save rate, comment quality, the number of posts that achieve minor viral status (defined as outperforming the KOC's typical average by a meaningful margin), and downstream search volume changes for your branded and category keywords. KOCs with commercial post ratios above 30% of their total content tend to have eroded audience trust — an important vetting criterion before committing budget.
For both KOL and KOC campaigns, build attribution mechanisms in from the start: assign unique promotional codes per creator (e.g., BRAND10), create creator-specific UTM-tracked landing pages for any off-platform traffic, and track branded search volume changes in the days and weeks following content publication. These combined signals give you a far more complete picture of campaign ROI than engagement metrics alone.
To understand how KOL and KOC strategies perform differently across specific product categories and industry verticals, AllXHS's industry-specific marketing strategies provide data-driven guidance for 20+ sectors.
---
Attribution Techniques for XHS's Multi-Touch Journey {#attribution}
Xiaohongshu sits at a specific and important stage in the Chinese consumer journey: discovery and consideration. Users frequently encounter a brand on XHS, save the content, research further on Baidu or brand websites, and then purchase on Tmall, JD.com, or through a WeChat Mini Program. This means last-click attribution will systematically undervalue XHS's contribution to your marketing mix.
A multi-touch attribution model that credits first-touch and mid-funnel interactions is far more appropriate for this platform. Linear attribution (equal credit across all touchpoints) is a reasonable starting point for brands new to the model. Position-based attribution, which weights the first and last interactions more heavily, works well when Xiaohongshu is your primary awareness channel and a separate platform handles the final purchase step.
Practically, attribution infrastructure should include platform-specific promo codes for tracking conversions originating from XHS content, cohort analysis of users exposed to your content to identify purchase pattern timelines, post-purchase surveys asking customers how they first discovered your brand, and — where technically feasible — custom landing pages accessible only through XHS that allow clean source attribution. Extending your measurement window to 6–12 months is also important: unlike most social platforms where content decays quickly, a well-optimized XHS note can continue driving search traffic and conversion for many months post-publication.
---
Building Your XHS Analytics Framework Step by Step {#analytics-framework}
Pulling these elements together into a functional analytics framework doesn't have to be complex. Here's a practical sequence for brands at any stage:
1. Verify your professional account and access your dashboard — Without a verified brand account, you're operating blind. Native analytics access is the minimum viable starting point for any measurement effort.
1. Define your primary objective and select your tier-one metrics — Brand awareness programs should prioritize impressions growth, branded search volume, and follower growth rate. Community-building programs should focus on save rate, CES, and UGC volume. Conversion-focused programs should track CPE, click-through rate, attributed revenue, and cost per acquisition.
1. Establish a pre-campaign baseline — Run your account for four to six weeks before a major campaign, collecting consistent data across your chosen metrics. This baseline is what makes before-and-after comparisons meaningful.
1. Layer in attribution infrastructure — Set up promo codes, UTM parameters, and any applicable post-purchase surveys before content goes live, not after.
1. Choose supplementary tools based on budget and needs — Pugongying and Lingxi are free or low-cost native options that all commercial accounts should use. Third-party tools like XinHong Data or AnyTag are worth evaluating once you have a regular cadence of campaigns to measure.
1. Review and iterate on a monthly cycle — XHS's algorithm evolves frequently. Monthly performance reviews allow you to catch algorithm-driven shifts early, before they significantly affect your results.
---
Common Mistakes Brands Make When Measuring XHS Performance {#common-mistakes}
Even with the right tools in place, several patterns consistently lead international brands astray when measuring XHS performance.
Optimizing for likes instead of saves. Likes are the lowest-value engagement signal on the platform and the easiest to inflate artificially. Save rate is the metric that correlates most strongly with purchase intent — don't let like counts dominate your reporting.
Measuring too early. XHS content has a much longer lifespan than content on Instagram or TikTok. Evaluating a post's performance at 48 hours is almost meaningless. Set measurement windows of at least 30 days for standard content and 3–6 months for search-optimized evergreen posts.
Ignoring search traffic data. Many brands set up their analytics dashboards but never investigate their traffic source breakdown. The discovery-to-search ratio in your traffic data is one of the most actionable signals for refining your content strategy.
Using follower count as a KOL proxy. Follower count is a vanity metric on XHS. Engagement rate, CPE, save rate on past sponsored posts, and commercial post ratio are far more predictive of campaign ROI. Use Pugongying's filtering tools to vet by performance data, not audience size.
Failing to account for cross-platform conversion. If you're only tracking conversions that happen inside the Xiaohongshu ecosystem, you're likely attributing only a fraction of the platform's actual revenue contribution. The discovery-to-purchase gap is real and significant — build your attribution infrastructure accordingly.
For brands looking for expert guidance on setting up analytics systems that connect to real business outcomes, AllXHS's expert Xiaohongshu marketing services offer hands-on support from specialists who work with the platform daily.
Getting Analytics Right Is How You Scale on XHS
Xiaohongshu analytics in 2026 is not about tracking more numbers — it's about tracking the right numbers within a framework that reflects how this platform actually works. The brands that consistently outperform on XHS treat analytics not as a reporting exercise, but as their primary decision-making engine: informing which content formats to double down on, which creators deliver genuine ROI, which search queries represent untapped demand, and how to allocate budget across an increasingly competitive platform.
The good news for international brands is that the tool landscape has matured significantly. Native instruments like Lingxi and Pugongying now provide levels of measurement transparency that were unavailable just two years ago. Third-party solutions are increasingly XHS-native rather than retrofitted from other markets. And the frameworks — CES scoring, intent-based search analytics, multi-touch attribution — are well-established enough to build reliable reporting systems around.
The remaining challenge is almost entirely one of knowledge and implementation. If you're ready to move from guesswork to a data-driven XHS program, explore AllXHS's free resources, including industry reports, tool templates, and training materials built specifically for international brands navigating the platform.
---
Ready to Build a Smarter XHS Marketing Strategy?
AllXHS is the #1 English-language resource hub for international brands marketing on Xiaohongshu. From analytics frameworks to KOL strategy, content optimization to compliance — our team of XHS specialists helps brands turn platform data into real growth.
[Get in touch with our team today →](https://www.allxhs.com/contact)