XZ｜公司探针

China/HK company research｜商业模式、现金流与证伪｜中文 / EN

[XZ｜公司探针 @Zhao_XU_90]

Mixue’s Moat Is Intact. Its Growth Engine Isn’t. https://t.co/u1TeJ0MDvL

[XZ｜公司探针 @Zhao_XU_90]

研究完泡泡玛特和蜜雪，我发现消费股最容易犯的错误，是把“规模扩张”直接等同于“利润增长”。
泡泡玛特要验证的，是注意力能否在多个IP之间轮动；
蜜雪要验证的，是门店扩张能否继续转化为单位网络供货和毛利。
一个依赖IP组合，一个依赖供应链密度。护城河都成立，但增长是否成立，要看完全不同的指标。门店数、IP数和海外扩张都只是过程，单位经济和现金流才是结果。

[XZ｜公司探针 @Zhao_XU_90]

蜜雪2026H1最值得看的，不是门店数，而是门店与总部经营结果的背离：
全球门店 +20.7%
集团收入 +2.3%
商品和设备收入 +2.1%
归母净利润 -14.7%
体系没有崩，但“开店越多、利润越高”的增长逻辑已经失效。研究重点已经从门店数量，转向单店供货与加盟商回报。
https://t.co/fFKDrRjvWS

[XZ｜公司探针 @Zhao_XU_90]

蜜雪集团 护城河成立 增长逻辑降级 https://t.co/DHHZqW18tT

[XZ｜公司探针 @Zhao_XU_90]

蜜雪集团 护城河成立 增长逻辑降级 https://t.co/DHHZqW18tT

[XZ｜公司探针 @Zhao_XU_90]

@robot2trade1 Where I differ: the asset is not any single hit, but Pop Mart’s ability to reallocate shelf space, formats and overseas rollout across IPs. In 1H26, THE MONSTERS fell 7.5% while five other artist IPs grew. The warning is inventory days rising to 201—not LABUBU fading alone.

[XZ｜公司探针 @Zhao_XU_90]

@liqian_ren Agreed on the demand gap. But a cheaper alternative proves manufacturing efficiency, not yet a consumer-brand moat. The real test is whether a Chinese entrant can own design, distribution and repeat purchase while preserving margins after overseas acquisition and service costs.

[XZ｜公司探针 @Zhao_XU_90]

@algotradingdesk The platform thesis is reasonable, but the figures appear inconsistent with Pop Mart’s 1H26 filing: group revenue grew 23.8%, not 200%+, while THE MONSTERS fell 7.5% to RMB4.45bn. The stronger evidence is IP rotation—not headline growth. What is the source for these numbers?

[XZ｜公司探针 @Zhao_XU_90]

@investing_china Policy can scale visibility, but brand durability is harder. The real test is whether heritage converts into repeat purchase, pricing power and profitable distribution. Scale measures size; unit economics show whether a brand can compound.

[XZ｜公司探针 @Zhao_XU_90]

@ChinaGSProject Brazil is strategically interesting, but store count is not the key evidence. Mixue’s overseas value depends on local unit economics: sales, payback, closures and whether supply density can sustain low prices. Until then, expansion is an option—not a proven growth engine.

[XZ｜公司探针 @Zhao_XU_90]

I just published Pop Mart Isn’t Just Selling Toys — It Is Building an IP Monetization Machine https://t.co/ZsiTcOKg5K

[XZ｜公司探针 @Zhao_XU_90]

Mixue (https://t.co/3gsyuV17br) still has a moat, but its growth engine is weakening.
In 1H26, stores rose 20.7%, product/equipment revenue only 2.1%, and net profit fell 14.7%.
The next test is revenue and gross profit per franchise store—not how many stores open.

[XZ｜公司探针 @Zhao_XU_90]

蜜雪集团（https://t.co/3gsyuV17br）的门店仍在扩张，但增长飞轮已经降级。
2026H1全球门店同比+20.7%，商品和设备收入仅+2.1%，归母净利润-14.7%。
护城河没有消失，只是从推动增长变成维持低价、心智和加盟网络。下一步要验证的是：单位网络供货能否止跌。

[XZ｜公司探针 @Zhao_XU_90]

Inventory—not slower growth—is Pop Mart’s clearest bear signal.
Inventory rose from RMB1.53bn to RMB6.10bn, while turnover days climbed from 102 to 201.
If this normalizes without impairments, the IP-platform thesis strengthens. If not, it remains hit-driven retail.

[XZ｜公司探针 @Zhao_XU_90]

@business Slower topline growth is real, but the mix matters: THE MONSTERS fell 7.5% in 1H26 while Twinkle Twinkle rose 580.6%. The bull case is portfolio rotation; the bear case is that inventory days reached 201. The next test is overseas SSS and stock efficiency.

[XZ｜公司探针 @Zhao_XU_90]

@yicaichina The buyback is secondary. The key question is whether 23.8% growth can coexist with inventory days rising from 123 to 201. If inventory normalizes without impairments while several IPs scale, the platform thesis strengthens; otherwise capital return does not fix the risk.

[XZ｜公司探针 @Zhao_XU_90]

@vivilinsv @POPMARTGlobal Good framing. The key distinction is between repeat purchase driven by random assortments and repeat purchase driven by IP attachment. Pop Mart reports member repurchase near 50%, but not the split—important when judging durability beyond the blind-box mechanic.

[XZ｜公司探针 @Zhao_XU_90]

9/9 Pop Mart is becoming an IP commercialization platform—but the thesis remains unproven. It fails if: (1) one IP dominates revenue; (2) inventory days and write-downs rise; or (3) overseas stores disappoint on same-store sales, payback and repeat purchase beyond leading IPs.

[XZ｜公司探针 @Zhao_XU_90]

1/9 Pop Mart (HKEX: 9992) can look like a one-hit LABUBU story. Yet in 1H26, THE MONSTERS revenue fell 7.5% to RMB4.45bn, Twinkle Twinkle rose 580.6% to RMB2.65bn, and group revenue still grew 23.8%. Luck—or an emerging IP commercialization platform?

[XZ｜公司探针 @Zhao_XU_90]

8/9 Inventory is the strongest counterevidence. It rose from RMB1.53bn at YE24 to RMB5.47bn at YE25 and RMB6.10bn at 1H26; reported inventory turnover days rose 102 → 123 → 201. If IP rotation needs ever more stock, the “platform” may still be fashion retail carrying hit risk.

[XZ｜公司探针 @Zhao_XU_90]

7/9 The operating loop is IP → controlled distribution → sales feedback → resource reallocation → more formats and markets. At YE25, Pop Mart operated 630 stores and 2,637 roboshops across 20 countries. In FY25, wholesale and other channels accounted for just 4.7% of revenue.