If you have ever opened the Xiaohongshu Aurora (聚光) dashboard and felt immediately overwhelmed by a wall of Chinese-language metrics, campaign toggles, and audience graphs, you are not alone. Aurora is Xiaohongshu’s native advertising intelligence platform, and while it is powerful, it is also genuinely dense — particularly for brands that are newer to the ecosystem or managing campaigns without a dedicated China-market specialist on the team.

The real problem, though, is not the volume of data. It is knowing which numbers to pay attention to and which ones to scroll past. Aurora surfaces dozens of performance indicators, but acting on the wrong ones can lead to budget misallocation, misguided creative pivots, and campaign decisions that look logical in isolation but fail to drive real business results. This guide cuts through the noise. Whether you are running paid promotions, seeding influencer content, or scaling a brand keyword strategy on Xiaohongshu, here is how to read the Aurora dashboard in a way that actually informs smarter decisions.

What Is Xiaohongshu Aurora (聚光) and Why It Matters

Aurora (聚光, literally “gathering light”) is Xiaohongshu’s centralised advertising and analytics platform, serving as the primary interface for brands and agencies running paid and organic performance campaigns on the platform. Think of it as the Xiaohongshu equivalent of Meta Ads Manager or Google Analytics — except that it is built specifically for a community-driven, discovery-first environment where purchase decisions are heavily influenced by peer content and authentic reviews.

What makes Aurora particularly important for performance marketers is that it bridges paid media and organic content intelligence in a single dashboard. You can monitor the performance of boosted notes (sponsored posts), track how your organic brand content is indexing within the platform’s search algorithm, and cross-reference audience behaviour data to refine targeting. For brands investing in Xiaohongshu marketing, Aurora is the single most important tool for understanding whether your content is actually reaching and resonating with the right consumers.

Navigating the Aurora Dashboard: A Quick Orientation

When you log into Aurora, the interface is organised around several primary modules: Campaign Management (投放管理), Data Reports (数据报表), Audience Insights (人群洞察), and Creative Management (创意管理). Each module feeds into a different layer of decision-making, and it is common for marketers to spend all their time inside Campaign Management while ignoring the depth of insight available in Audience Insights and Data Reports.

The top-level summary view gives you a snapshot of impressions (曝光量), clicks (点击量), engagement rate (互动率), and spend (消耗). These headline numbers are useful for a quick health check, but the real analytical value is one or two layers deeper. Before you can act intelligently on Aurora data, you need to understand which metrics belong to which decision layer — and that requires distinguishing between metrics that signal content performance and metrics that signal business performance.

Vanity Metrics vs. Actionable Metrics: Know the Difference

This distinction is critical on Xiaohongshu because the platform’s community-first design means that high impression counts can feel reassuring even when the campaign is underperforming commercially. Vanity metrics are numbers that look good in a report but do not reliably predict downstream value. Actionable metrics, by contrast, are the ones that correlate with real outcomes — lead quality, conversion rate, customer acquisition, or sustained brand interest.

On Aurora, raw impression volume (曝光量) is the most commonly over-weighted vanity metric. A note can be served to millions of users in a broad interest targeting scenario and generate almost no meaningful engagement or search lift. Similarly, total click volume without segmenting by click type (profile visits versus link clicks versus search triggers) tells you very little. The practice of content marketing on Xiaohongshu is fundamentally about earning trust and triggering discovery intent — and the metrics that reflect that are more nuanced than a headline impression number.

The Metrics That Actually Move the Needle

Exposure Volume and Click-Through Rate (CTR)

Exposure volume becomes meaningful only when paired with CTR. A healthy CTR on Xiaohongshu for a boosted note typically sits between 3% and 8%, depending on the category and creative format, though this benchmark varies by industry. If your CTR is sitting below 2%, the issue is almost always at the creative or headline level — users are being served the content but are not compelled to open it. This is a signal to test new cover images, rewrite the hook copy in the first line of the note, or reconsider whether the content format (image carousel versus video versus single image) matches what the target audience responds to in that category.

Conversely, a high CTR paired with low subsequent engagement is a red flag. It suggests the content preview is compelling but the body of the note fails to deliver on what it promised — a mismatch that erodes trust and can suppress organic distribution through Aurora’s quality scoring system.

Engagement Depth: Saves, Comments, and Follows

On Xiaohongshu, saves (收藏) are generally considered the strongest single engagement signal. A user saving a note is explicitly bookmarking it for future reference — a behaviour that strongly correlates with purchase consideration and repeat platform visits. Aurora breaks down engagement by type, and the save rate relative to impressions is one of the best early indicators of whether a piece of content has genuine utility or aspiration value for your audience.

Comments (评论) carry a different kind of weight. A high comment volume on a brand note suggests the content has sparked a community conversation, which Xiaohongshu’s algorithm actively rewards with extended organic reach. Pay close attention to the sentiment and content of comments — Aurora does not provide built-in sentiment analysis, so this requires manual review, but what users are asking or saying in the comments section is often the most direct signal you will get about unmet questions or purchase intent triggers in your target market.

Follows (关注) generated by a campaign note indicate that users found the content valuable enough to want more from the brand account. This is particularly relevant for brands in early awareness stages on the platform, where growing a credible follower base is a prerequisite for organic content to gain traction through Aurora’s distribution logic.

Cost Per Engagement (CPE) and Cost Per Acquisition (CPA)

CPE (每次互动成本) is Aurora’s efficiency metric for engagement-based campaigns, and it is the number that should anchor most budget optimisation decisions for brand awareness objectives. If your CPE is trending upward over a campaign period, it typically indicates audience fatigue — the same users are seeing your content repeatedly without responding — or that your targeting has become too narrow and the algorithm is struggling to find new relevant users efficiently.

For brands with e-commerce or lead generation objectives, CPA (获客成本) is the north star metric. Aurora supports CPA tracking when your campaign is connected to a Xiaohongshu store, a linked landing page, or a lead form. If your CPA is significantly higher than your product margin allows, the fix rarely lives at the targeting level alone. More often, the issue is a disconnect between the content’s promise and the landing experience — a gap that requires cross-functional coordination between your content team, your website design team, and your campaign strategists.

Audience Insights and Interest Tags

The Audience Insights module inside Aurora is consistently underused, and this is where brands leave significant strategic value on the table. This section shows you the demographic breakdown of users who engaged with your content — age, gender, city tier, device type — alongside interest tag clusters (兴趣标签) that reveal what else your engaged audience cares about beyond your immediate category.

These interest tags are particularly powerful for refining future targeting and for informing influencer marketing strategies. If the data shows that users who save your skincare brand’s content also heavily index for fitness, wellness supplements, and healthy eating content, that is a targeting and creator partnership signal that most brands simply miss when they focus only on top-level campaign metrics. Sophisticated use of Aurora’s audience data is one of the clearest differentiators between brands that scale efficiently on Xiaohongshu and those that plateau after initial traction.

Search Conversion and Note Heat Score

One of the most distinctive features of Aurora is its ability to surface search-related performance data, which reflects Xiaohongshu’s dual nature as both a social platform and a product discovery search engine. The Note Heat Score (笔记热度) indicates how well a piece of content is performing within the platform’s organic search index — a metric that matters enormously for brands that want their notes to surface when users search for category-relevant keywords.

If your paid notes are generating impressions and clicks but not improving brand keyword search volume over time, it suggests your content is winning attention in-feed but failing to seed the kind of word-of-mouth that drives users to actively search for your brand. This is the signal to revisit your content strategy — specifically, whether your notes are structured around the search queries your target customers actually use, and whether they include the keyword-rich language that Xiaohongshu’s discovery algorithm favours. This type of search-content alignment is closely related to the principles behind answer engine optimisation and generative engine optimisation, where content must be structured to surface in discovery-first environments rather than relying solely on paid placement.

How to Read Optimization Signals in Aurora

Aurora surfaces several automated optimisation recommendations within the campaign management interface, particularly for campaigns running on CPM (cost per thousand impressions) or oCPX (optimised cost per action) bidding models. These recommendations — such as budget increase suggestions, audience expansion prompts, and creative refresh alerts — should be treated as hypotheses to test rather than instructions to follow blindly.

The most reliable optimisation signal in Aurora is the delivery curve. A healthy campaign shows relatively stable delivery with gradual spend acceleration as the algorithm learns. A sharp spend spike followed by a plateau almost always indicates the algorithm has exhausted its initial learning phase and needs either a creative refresh, a targeting adjustment, or a bid strategy change. Watching the delivery curve alongside your CPE trend gives you a much more nuanced read on campaign health than looking at aggregate numbers at the end of a reporting period.

Common Mistakes Brands Make When Reading the Dashboard

The single most common mistake is optimising exclusively for impressions and clicks while ignoring downstream behaviour. Xiaohongshu is a platform where the path from content discovery to purchase can span days or weeks, and Aurora’s attribution window may not capture the full conversion journey if your campaign is not configured correctly. Brands that evaluate Aurora data on a single-session basis routinely underestimate the platform’s actual contribution to sales.

Another frequent error is failing to segment performance data by content format and creator type before drawing conclusions. A mixed campaign running both brand-produced notes and KOL-driven content will show blended metrics that obscure the fact that the two content types are likely performing very differently. Pulling disaggregated reports by creative unit is essential before making any decisions about budget reallocation or creative direction. Using tools like AI influencer discovery platforms can help you pre-qualify creators whose content performance history aligns with your campaign benchmarks, reducing the variance in your Aurora results before a campaign even launches.

When to Bring in Expert Help

Aurora is a capable platform, but it rewards experience. Brands that are new to Xiaohongshu, operating without Mandarin-language proficiency on their marketing team, or managing campaigns across multiple markets simultaneously often find that the time cost of correctly interpreting and acting on Aurora data outweighs the perceived savings of in-house management. This is particularly true during campaign scaling phases, when small misreadings of optimisation signals can translate into significant wasted spend.

Working with an experienced AI marketing agency that has hands-on Aurora management experience means your dashboard data is being read in context — benchmarked against category norms, cross-referenced with platform algorithm updates, and translated into campaign decisions that account for the full Xiaohongshu consumer journey rather than isolated metric snapshots. For brands that are serious about scaling on Xiaohongshu across Southeast Asia and beyond, that context is not a luxury. It is the difference between a campaign that looks good in a report and one that actually builds lasting brand equity on the platform.

Final Thoughts

The Xiaohongshu Aurora dashboard gives you access to one of the richest data environments in Asian social commerce — but raw access to data is not the same as the ability to act on it intelligently. The brands that win on Xiaohongshu are not necessarily the ones spending the most on Aurora campaigns. They are the ones who have learned to focus on the metrics that genuinely reflect consumer intent: save rates that signal purchase consideration, comment patterns that reveal unmet questions, audience interest tags that unlock smarter targeting, and search conversion trends that show whether brand awareness is compounding over time.

Stop reporting on impressions as a success metric in isolation. Start building a reading habit around CPE trends, engagement depth ratios, and audience insight overlaps. And if the dashboard is still feeling opaque, that is a strong sign it is time to bring in specialists who speak the language — literally and analytically.