Xiaohongshu Anti-Scam Campaign

Why

Incoming mainland students face three overlapping risks: information anxiety and inexperience, which makes phrases like “your account is frozen” or “an official call” destabilising; heavy reliance on Xiaohongshu for rental, banking and pre-departure guidance; and trust in peers over institutions. Conventional static outreach — institutional pushes, posters, assemblies — rarely penetrates that attention and trust network.

How

• Three recruitment rounds: direct screening of Hong Kong-based creators → recruitment notices through campus societies and the official account → outreach to over a hundred newly arrived student creators in campus groups
• 820 creators contacted, 353 replied, 203 published; every participant signed an informed-consent form, and no identity documents or account credentials were ever collected
• Two phases: Hong Kong-resident creators shared life experience before 31 August; newly arrived student creators documented their first weeks after 1 September
• One shared tag, #我在香港防詐騙, with no restriction on format so that posts stayed natural
• Back-end metrics were collected only after seven days live, so that short-term fluctuations would not distort the assessment

Results

• From 144 collected back-end datasets: 1.87 million impressions, 310,000 views and 5,651 shares; roughly 13,140 impressions and 2,202 views per post
• Projected across all 203 publishing creators: about 2.66 million impressions, 440,000 views and more than 8,000 shares
• Student survey: 78.5% had seen anti-scam content on the platform, and 64.7% of them noticed the shared tag
• Content organically split into four formats: embedded reminders, consolidated know-how, first-person accounts and screenshot-driven posts

Interrupting scams

Comments included immediate reactions such as “I just got a call like this — good thing I saw your post”, “now I know it was fake” and “I clicked the link, what do I do?”. Creators and other users replied with official hotlines, handling advice and warnings. Because these scenes were reported by users while the scam was still unfolding, the content intervened at an early stage; more than ten cases of creators successfully interrupting an active scam were observed.

Challenges

• Platform mechanics: without enterprise verification the laboratory could not use the official collaboration system, so outreach was manual and initial response rates were very low
• Words such as “scam”, “police” and “report” are treated as sensitive by moderation; some posts were throttled or removed, and appeals were answered by machines
• Interference from scam operators: some creators received deletion threats issued in the platform’s name, and posts were repeatedly mass-reported
• Posts touching on “proxy purchasing”, “transfers” or “commission” were markedly more likely to be throttled or removed

Second round, 2026

A second round ran in August 2026 under the same tag, reaching 575,000 tag views and 4,864 discussions — evidence that the peer-trust model is repeatable across cohorts.