Industry stance: Divergent (mixed conditions across value chain segments) | Research date: 2026-11-15 | Scope: Domestic micro short dramas + overseas short drama apps, covering four monetization links: user acquisition, payments (IAP), advertising (IAA), and IP licensing | Horizon: 12–24 months
The short drama industry is in a transition period of "volume growth, stagnant pricing, thin profits": volume on the demand side (downloads, time spent, user pool) is still growing rapidly, but the growth rate of USD revenue is decelerating, and the profit pool structurally accrues to traffic and payment channels. In H1 2026, overseas ad creative volume grew +324% YoY while in-app purchase revenue grew only +13% — user-acquisition inflation is systematically shifting profits in this business from content and platform players to Meta, ByteDance, and Apple/Google. On the platform side, 2026 has seen defensive, razor-thin profitability of "trading revenue for profit" (Fengye Interactive cut H1 revenue by -43% for a ~3.5% net margin), not an organic improvement in unit economics; the segments truly "making real money" are ByteDance's owned-traffic ecosystem (Hongguo + Juliang/Ocean Engine) and the rent-collecting payment/advertising channels. A-share short drama concept stocks show a pattern of "industry not expensive, individual stocks expensive": the SW Media Index PE sits at the 52nd historical percentile, while concept stocks generally trade at extreme PS/PB percentiles with profit elasticity yet to materialize. We judge the industry's momentum as "divergent": volume and ad inventory trending up, paid revenue growth and production-side returns trending down; we judge the quality of returns in this business as "poor" and the value-trap risk as "high" (specifically for pure user-acquisition IAP platforms and high-valuation concept stocks). We recommend participating via the "picks-and-shovels" approach (traffic, tools, payment workarounds) and being cautious about "bought growth."
Data caveat (the #1 data risk in this report): There is no unified official statistics for the short drama industry, and four data sets exist for the overseas market with large discrepancies — Sensor Tower's USD 2.8 billion overseas IAP revenue for 2025 (narrow definition, used for this report's headline), the China Audio-Visual Conference's USD 2.329 billion, DianDi Data's USD 2.058 billion, DataEye's broad definition (including social media and ad monetization) projecting over USD 6 billion for 2026, and MPA's USD 3.6 billion for 2026; the rumor that Hongguo generates "RMB 40 billion in annual revenue" has been officially refuted. All market-size figures in this report are labeled by source; broad definitions are used only as range references.
The research objects are micro short dramas (web dramas with episodes lasting tens of seconds to a few minutes, primarily vertical-screen format) and their overseas expansion value chain, with monetization covering four links: user acquisition (paid media buying + organic platform traffic), payments (IAP: per-episode unlock / coin top-ups / subscriptions), advertising (IAA: free viewing monetized via rewarded video / feed ads), and IP licensing (IP licensing, revenue sharing, secondary distribution).
One-line value chain map: Web novel IP/scripts → producers (live-action/AI-generated) → distribution platforms (overseas standalone apps / domestic free platforms like Hongguo / Douyin mini-programs / social channels like TikTok) → acquisition channels (Meta/Google/TikTok/Ocean Engine) and payment channels (Apple/Google/WeChat). The domestic market is dominated by the free IAA model (about 66.3% of the domestic market in 2025, per the China Netcasting Services Association white paper), while overseas remains dominated by paid IAP (about 80% of overseas revenue), with the free model scaling rapidly. This report is organized around a consumer internet unit-economics framework; commodity supply-demand balance sheet and term structure templates do not apply.
Domestic micro short drama users: 662 million (Dec 2024) → 696 million (Jun 2025, white paper data); industry MAU 718 million (Feb 2026, QuestMobile), reaching 56.3% of mobile internet users — the user penetration dividend has largely been realized, exceeding mobile gaming and long-form video user bases. The structural change: the free IAA model's share rose from 11% to 66.3% within 18 months; Hongguo's DAU climbed from above 100 million in Jan 2026 to 168 million in July and about 203 million in September (QuestMobile), already exceeding the combined DAU of iQiyi, Youku, Tencent Video, and Mango TV, with 125 minutes of daily per-user usage; over the same period, long-form video platforms' monthly time spent fell -24% to -47% YoY. Paid mini-program dramas have been squeezed to about one-third share, with average acquisition ROI falling to 1.03 and the paid segment declining continuously through H2 2025. Domestic market size is about RMB 100 billion for 2025 (white paper broad definition, including IAA ads; iiMedia narrow definition RMB 67.79 billion — a nearly 1.5x gap between the two), projected at RMB 105–119.2 billion for 2026; considering regulatory cleanup's compression of AI-generated drama inventory (see I6), we take the lower end of the range, with domestic market growth of about 5–10% in 2027.
Overseas is the industry's main growth driver: full-year 2025 IAP revenue of USD 2.8 billion (+116%, Sensor Tower), 2026Q1 IAP about USD 750 million (+20%), H1 about USD 1.27 billion (+13%) — growth has plunged from triple digits to low double digits; meanwhile, downloads grew +140% in 2026Q1 and +79% in H1, with six short drama apps entering the global Top 40 by downloads. The implication of the gap: paid conversion cannot keep pace with the surge in user acquisition (creatives +324% vs IAP +13%, revenue elasticity only 0.04), and incremental users are concentrated in low-eCPM markets. By region: North America contributes about 54.6% of global revenue (most mature, slowing, most expensive acquisition); Japan and Korea have strong payment capacity (Japan at 7.3% of global revenue); Southeast Asia accounts for 32% of global downloads in 2026Q1, with nearly 40 minutes of daily usage already at OTT levels; Latin America and India combined account for about 45% of downloads but less than 5% of revenue — "high growth, low monetization," the core testing ground for the IAA model.
The "IAP→IAA" switch completed domestically is replaying overseas: ByteDance's Melolo has 30-day retention of 10.08% (5x Adjust's 2% benchmark), PineDrama surpassed 100 million downloads within 9 months of launch, and TikTok's short drama revenue-sharing pool grew from USD 7 million in Jan 2026 to USD 31 million in August (+343%). But it must be emphasized: there is no first-hand ARPDAU disclosure validating overseas IAA unit economics — 82% of downloads for leading free apps come from the lowest-eCPM markets like Indonesia and Brazil (Southeast Asia rewarded video eCPM as low as USD 0.28–1.10 per thousand impressions); the domestic Hongguo model relies on ByteDance's owned traffic at near-zero acquisition cost, a condition third-party platforms lack overseas. We therefore characterize IAA as "validated domestically, a potential successor awaiting validation overseas," without making a headline-level assertion.
Demand bridge methodology (bottom-up, Sensor Tower IAP base + conservative IAA increment): user scale-up (overseas short drama MAU ~220 million → 360 million, short-video user pool 1.8 billion, penetration 20%→25–30%) contributes +USD 1.78 billion; paid conversion improvement (rising share of local-language dramas, weighted conversion 9.1%→9.6%) contributes +USD 250 million; annual ARPPU USD 140→150 contributes +USD 340 million — IAP 2028E ~USD 5.18 billion (implied CAGR 22.8%); IAA increment conservatively at +USD 800 million (range USD 0.8–1.5 billion, awaiting quantification). 2028E overseas market midpoint ~USD 6 billion (range USD 5–7.2 billion); DataEye's broad definition (including social and AI dramas) extrapolating to ~USD 9.8 billion serves only as an aggressive upper bound.
Nature of demand: predominantly structural (about 60–70% of the increment) — vertical short dramas substituting long-form video time, emerging-market penetration, local-language dramas substituting translated dramas (paid conversion 8–12% vs 3–5%), and the AI supply revolution; cyclical pulses (about 30–40%) that are shrinking — paid media-buying pulses (average ROI 1.03) and domestic regulatory cleanup compressing low-quality supply. The "impulse payment wall" model — annual ARPPU USD 110–150 per paying user, USD 20–50 total unlock cost per drama — remains the mainstay of overseas revenue.
Domestic launches: ~24,000 titles in 2024 → 33,000 in 2025 (NRTA) → ~550,000 in the first 10 months of 2026, 16.7x full-year 2025, of which over 94% are AI-generated short dramas / animated dramas (NRTA Development Research Center). Per-title costs have been broken through the floor by AI: overseas live-action shoots at USD 180,000–200,000 per title → AI overseas dramas at ~USD 10,000 per title (an ~18–20x reduction); domestic live-action vertical dramas RMB 200,000–300,000 (2024) → RMB 500,000 (2025) → RMB 100,000–150,000 full-AI pipeline; production cycles compressed from 3 months to 1 week. Any hit theme is replicated and saturated by AI capacity within 2–4 weeks.
The cost of supply is collapsing monetization efficiency: Douyin saw 221,900 new AI/animated dramas in H1 2026, of which only 1,055 exceeded 100 million views (hit rate 0.47%); using 50 million views as the breakeven line, only ~1.3% of AI dramas recoup costs (DataEye). Live-action supply has collapsed: monthly new live-action launches in Jan–Jun 2026 fell over 90% from the ~20,000-title level; Tinghuadao has largely halted live-action production in favor of AI, COL's live-action monthly output fell to single digits, and Jiuzhou cut live-action capacity in half.
Media buying is the core cost of this business: domestic media-buying spend in 2025 was ~RMB 42.3 billion (+28%, DataEye), about 2.6x total production spend; typical media-buying-to-production ratio per title is 82:18 to 92.5:7.5; by channel, Ocean Engine holds ~73–74%, Tencent Ads 16–17%, Kuaishou ~8%. Overseas: 65% of acquisition depends on Meta, ~20% on Google, ~10% on TikTok (DianDi Data); CPI ~USD 5.28 in North America, ~USD 0.93 in APAC, ~USD 0.34 in LatAm. In H1 2026, overseas ad creative volume hit 19.61 million (+324%) against downloads +79% and IAP +13% — creative growth is ~4x download growth, with acquisition marginal efficiency decaying rapidly; domestic traffic costs rose over 100% YoY. Mature teams achieve media-buying ROI of only 1.03–1.07 (on gross billings), with ~80–90% of projects industry-wide failing to break even.
Supply-side conclusion: content is no longer scarce (AI broke through the floor); the monetization bottleneck lies in "buying the right traffic" and localized operations; ad inventory is quasi-rigid supply, with 1,200+ new dramas daily competing for slots, so bidding density can only rise. On the licensing side, web novel IP procurement and the licensing chain remain one source of differentiation (Hongguo guarantees RMB 50,000–120,000 per script plus 20% revenue share to "cherry-pick" writers), while overseas piracy (nearly 5,000 episodes stolen in a single case, copyright enforcement spending exceeding RMB 30 million) continues to erode the revenue pool.
The RMB 100 top-up funnel for the IAP (paywall) model: payment channel fees ~RMB 19.6 (assuming 60% via app stores at 30% commission, 40% via web direct top-up at ~4% fees) → media buying ~RMB 70 (upper end of the 55–70%-of-billings range) → content and operations ~RMB 6.9 (post-AI cost reduction, production at 0.8–1.5% of billings vs 7.5–18% in the live-action era) → platform keeps ~RMB 3.5. This decomposition cross-validates Fengye Interactive's actual H1 2026 net margin of ~3.5%.
The IAA (free + ads) model: monetization price drops to zero in exchange for the full user base, revenue = ARPDAU × DAU. Domestic Hongguo ARPDAU is ~RMB 0.27/day (calculated from media-estimated annual revenue of RMB 20 billion ÷ DAU 203 million ÷ 365 days; this revenue base is unofficially confirmed — Hongguo has refuted the "RMB 40 billion annual revenue" rumor, used only as an order-of-magnitude proxy), with content creators receiving ~36% of ad revenue (monthly revenue share exceeding RMB 600 million). The key difference: IAP outsources its channel take to Meta + Apple (~RMB 90 per RMB 100 of billings combined), while IAA internalizes it within ByteDance (ByteDance is simultaneously the traffic source, ad platform, and distributor); IAA's "free acquisition" advantage holds only within ByteDance's owned-traffic ecosystem — third-party platforms buying traffic for IAA in Southeast Asia see LTV/CAC of only ~0.25 (LTV30 ~USD 0.23 ÷ CPI USD 0.93), failing to break even.
Global short drama apps' average 30-day retention is 8.6% (day-one 26.9%), versus Adjust's benchmark of only 2%; TikTok is 29.8% — short drama is a "high-frequency, low-retention" format, with LTV driven mainly by impulse payments in the first two weeks and continuous media buying to replenish volume. Structural improvement is appearing in IAA free apps: Melolo's 30-day retention at 10.08% (5x benchmark), Hongguo nearly doubling retention through coins and binge-tracking mechanics; engagement is the industry's strongest metric (Hongguo 125 min/day, Melolo 96 min, overseas average 17 min). The retention center of gravity is slowly shifting from "acquisition-pulse-driven" to "product-operations-driven."
At the leading-company level: Fengye Interactive 2025 revenue RMB 5.721 billion, net loss RMB 85.84 million (net margin -1.5%) → H1 2026 revenue RMB 1.57 billion (-43%), net profit RMB 55.69 million (net margin ~3.5%) — essentially a defensive move of cutting media buying to preserve profit; MPA projects full-year 2026 revenue of USD 1.05 billion and net margin of 3.81%, but H1 annualizes to only ~USD 440 million, requiring more than a doubling QoQ in H2 to hit the forecast — that forecast looks aggressive; we characterize 2026 leadership profitability as a 2–4% net margin "micro-profit" level. Dianzhong Technology self-reports short drama net margin below 1% (the largest platform by revenue); COL's 2025 revenue was RMB 1.657 billion (+43%) but selling expenses were RMB 953 million (57.5% expense ratio) with an attributable net loss of RMB 671 million — a typical case of "revenue up, profit down." Under MPA's framework, Fengye's profit expansion hinges on a single variable: acquisition costs falling from 55% to 44% of revenue (each percentage point worth ~USD 10.5 million of EBITDA, ~26% of the projected full-year net profit) — this assumption clashes head-on with the reality of acquisition inflation, and the sustainability of the profit inflection is in doubt.
Growth quality assessment: 2025 was purely "bought growth" (revenue +116% built on media buying, leaders losing money across the board); 2026 industry-wide remains bought growth (creatives +324% vs revenue +13%, elasticity 0.04), with leaders achieving defensive micro-profits through contraction. "Profitable growth" center: on a billings basis, overseas market 2026E ~+20% (IAP ~USD 3.36 billion), 2027E base +15–25%; on a net profit basis, leaders at 2–4% in 2026, 2027E base 3–6% — net profit growth (~10–20%) is significantly below billings growth (20–40%), and the gap is the "traffic inflation tax". Reference trajectory: overseas dual-platform IAP monthly billings ~USD 290–310 million in 2026Q4 → ~USD 340–380 million in 2027Q3 (contingent on creative inflation peaking and receding; see D2 inflection points).
Overseas IAP revenue (Sensor Tower, ~USD 750 million globally in 2026Q1): DramaBox (Dianzhong) and ReelShort (Fengye Interactive) each ~USD 140 million per quarter, CR2 ~37.3%, CR5 estimated at 50–60% (challengers NetShort, DramaWave, GoodShort, etc.); Chinese developers hold nine of the top ten paid apps. The landscape is not settled: Maiya's NetShort surged to the top two in overseas revenue within a year, and free apps (FreeReels, Melolo, PineDrama) are rewriting download charts. Domestic free side: Hongguo holds ~81% MAU share (a runaway #1, ~5x the #2 player Hema), free app CR5 ~99%; the paid mini-program side is fragmented.
Four layers of moats are shifting: ① the media-buying data flywheel (creative-to-payback ROI models + daily spend scale) is the core moat, with acquisition costs rising over 100% annually raising the bar; ② localization tiering — Europe/US requires local original content teams (ReelShort's LA studio, a deep moat), while Southeast Asia can rely on translated dramas + AI dubbing (easily replicable); ③ content library moats are collapsing (per-drama life cycle ~3 months, 1,200+ new AI titles daily); ④ user switching costs are near zero — consumers face no switching cost across homogeneous multi-app content, making this an "operational efficiency + capital" business rather than a "user lock-in" business, with inherently shallow moats.
Of the domestic RMB 63.4–100 billion market, media buying accounts for ~RMB 42.3 billion (42–67%); overseas, 55–90% of paid billings go to media buying plus 15–30% to payment fees. The only high-certainty profit-retaining segments are traffic platforms (Meta/ByteDance/Google ad businesses with 50%+ gross margins) and payment channels (Apple/Google risk-free commissions; USD 2.98 billion of overseas IAP implies USD 450–890 million in fees). Platform net margins are 0–5% (Dianzhong <1%, Fengye 3.5%); on the production side, 80–90% of dramas fail to break even, and 98.7% of AI dramas do not break even within six months. Over the next 2–3 years, profits will continue migrating from platform/production players to traffic and channel players: acquisition inflation (AI hands over every dollar saved on production to media buying) plus the monetization mix shifting from IAP to IAA (ad dividends accrue to whoever owns the ad inventory). The exceptions are ByteDance's owned-traffic ecosystem (Hongguo captures both traffic and advertising ends) and leaders with local original production capacity plus IP sources (beginning to accumulate thin profits).
Return quality: overall "poor," with genuine money-making segments existing across the layers. Making real money: paid-traffic channels (tax-like), payment revenue shares, ByteDance's ecosystem (dual capture of both ends via its own closed-loop traffic), and thin-margin head platforms starting 2026; prosperity illusion: second-tier overseas platforms (burning money for DAU), paid mini-program production contractors (hit-driven profits eaten up by ad buying and revenue splits), and most A-share concept stocks (revenue elasticity without profit elasticity).
Industry valuation position: Shenwan Media Index PE at 38.71x (52nd percentile since 2014), PB at 2.65x (51st percentile, as of 2026-11-13) — the industry index sits at the historical midpoint. But concept stocks are far more expensive than the industry: COL Group market cap approx. RMB 20.2 billion, PS approx. 12.2x (PB of over 100x is distorted due to nearly depleted net assets); Zhangyue PB 5.61x (90th percentile over the past year); Shanghai Film PB approx. 11x (99th percentile over 3 years); Jiecheng Technology PE(TTM) approx. 105x — "industry not expensive, individual stocks expensive": a concept premium, not an earnings premium. Prosperity conclusions must be checked against valuation: volume-driven prosperity × extreme individual-stock valuations ≠ allocation value at industry mid-cycle valuations.
Domestic: The first special regulation, the Administrative Measures for the Development of Micro-Dramas (NRTA Order No. 16), effective September 1, 2026 — classified filing + distribution licensing, filing threshold raised approx. 3x from 2024 online levels (Category 1 investment ≥ RMB 3 million), mandatory AI-generated content labeling per episode, explicit disclosure for paid content, algorithms prohibited from inducing addiction, and dramas shot domestically for overseas broadcast are also subject to filing review (Article 52, a critical impact on the overseas supply chain). Special governance of AI micro-dramas: from April 1, 2026, unregistered AI dramas were forcibly removed across all platforms, and daily new AI drama releases plunged from approx. 2,000 to 350 (-82.5%), with over 70,000 dramas cumulatively removed industry-wide — the clean-up benefits head players' market share but is a net compression of total domestic ad inventory (in tension with the IAA monetization logic; quarterly data on new releases and revenue-sharing pools must be tracked). Free platforms like Hongguo have been brought under the same regulatory standard (December 2024 interview and rectification).
Overseas: ① TikTok US: USDS joint venture settled in January 2026 (Oracle/Silver Lake/MGX 15% each, ByteDance retains 19.9%), ban lifted — the largest social media ad-buying channel for short dramas is preserved, and TikTok has entered the space itself (LimeShorts paid short drama testing); however, PAFACA's "foreign adversary application" designation mechanism remains a structural tail risk for other Chinese-background short drama apps (both ReelShort and DramaBox are Chinese-operated). ② Major payment-side positive: after the Epic v. Apple case, external-link payments are commission-free in the US market (from June 2025), legitimizing the web-direct top-up path for IAP-heavy short dramas — but the Ninth Circuit left room for a "reasonable commission," and the Supreme Court has scheduled oral arguments for January 11, 2027; there is a give-back risk before the final ruling. ③ EU DSA enforcement against addictive recommendation algorithms (three preliminary infringement findings against TikTok in 2026), member-state age verification in place by end-2026, and the AI Act transparency obligations (applicable from August 2026) are raising European compliance costs; US comprehensive privacy laws in 24 states + FTC enforcement on minors' payment mechanics ($20 million settlement precedent) point to tightening monetization of underage traffic. ④ Vietnam's PDPL (effective January 2026) and Middle East data localization are raising local operating costs.
Copyright: Translation/adaptation dramas carry numerous inherent flaws in the web-novel licensing chain; the "one-to-one adaptation" habit in going overseas persists; the mainstream piracy format is overseas servers + counterfeit apps (nearly 5,000 episodes in a single case), with difficult evidence collection and cross-border enforcement; AI-generated content copyright ownership remains unsettled both in China and abroad. Net direction: neutral to positive — regulation is shifting from campaign-style rectification to institutionalized routine, policy uncertainty is declining, and rising compliance costs will squeeze out tail players, benefiting existing head players and licensed platforms.
| Scenario | Probability | Narrative | Implications |
|---|---|---|---|
| Bear | 0.35 | Ad-buying inflation structurally persists (creative growth >150%), overseas IAA fails (ARPDAU < $0.05), industry ROI falls below 1 | Overseas market growth <10%, head net margin 0-2%, mass exit of second-tier apps, A-share concept valuations mean-revert |
| Base | 0.50 | Creative inflation slowly recedes with hot money's exit, AI cost reduction + web-direct top-ups deliver, IAA ramps by eCPM in emerging markets with limited incremental size ($0.8-1.5 billion) | Overseas market +15-25% (including IAA), head net margin 3-6%, domestic market +5-10% |
| Bull | 0.15 | ByteDance fully replicates Hongguo's internal loop overseas (own traffic at zero CPI + 5x retention), IAA expands monetization to the full user base, AI-native drama hit rates improve | Overseas market +30%+, head net margin 8-15% |
Catalyst calendar (corresponding to the key battlegrounds): 2026-12-31 EU age verification in place (D1/D3, European ad-buying costs) | 2027-01-11 US Supreme Court Epic v. Apple oral arguments (D2, risk of giving back the external-link commission-free dividend) | 2027-01 A-share earnings season: COL Group's equity-method P&L or impairment on Maple Leaf Interactive (D2) | 2027-03 Sensor Tower/DataEye annual reports: first verifiable window for IAA ARPDAU (D1) | 2027Q2 Maple Leaf Interactive IPO developments and COL Group litigation progress (D2/stock mapping) | 2027-06 Vietnam data localization transition period ends (policy) | 2027-09 Order No. 16 one-year implementation data (D3). Also continuously track: Hongguo monthly revenue shares and DAU, TikTok short drama monthly revenue-share data, Sensor Tower quarterly in-app purchases.
One sentence to expose the value trap logic: bullish on industry prosperity ≠ worth buying. The points in the bottom-right quadrant — Dianzhong (No. 1 by scale with net margin <1%), COL Group (large revenue elasticity but no profit realization, PS approx. 12x), Zhangyue (No. 3 in downloads ≠ revenue, PB at the 90th percentile over the past year) — share a common trait: the prosperity dividend of industry volume growth is eaten up by ad-buying inflation and channel take rates, yet valuations are priced for prosperity; the A-share concept stocks and live-action contractors in the bottom-left quadrant are the "declining prosperity × poor returns" avoidance zone.
Relative beneficiaries:
Avoid:
| Scenario | Biggest Beneficiaries | Biggest Losers |
|---|---|---|
| Bear (0.35) | Almost no platform-side beneficiaries; ByteDance (channel taxation), Kuaishou (small exposure) relatively defensive | Mass exit of second-tier IAP platforms; collapse of A-share concept valuations; COL Group (core business losses + Maple Leaf impairment risk) |
| Base (0.50) | Maple Leaf Interactive (thin profits + share), Dianzhong (scale defended), Kuaishou (advertising + tools) | Live-action contractors (capacity clearing), Zhangyue (downloads fail to convert to revenue) |
| Bull (0.15) | ByteDance ecosystem (Hongguo model replicated globally), Kuaishou (Kling AI + ad inventory), Maple Leaf (net margin recovering toward 8%+) | Pure-paid model platforms (displaced by the free model), long-video platforms (further time-share loss) |
① Structural deterioration of ad-buying inflation: creative volume bidding against quasi-rigid ad inventory pushes industry-wide ROI below breakeven, with the AI cost-reduction dividend fully harvested by traffic platforms (the single biggest risk); ② falsification of overseas IAA unit economics: emerging-market eCPM and ARPDAU fail to cover acquisition costs, with the 2028E market midpoint revised down below $5 billion; ③ geopolitical and platform risks: PAFACA mechanism extended to Chinese-background short drama apps (forced restructuring/removal), the Epic case's final ruling giving back the external-link commission-free dividend, EU minors' rules compressing the ad-buying pool; ④ domestic regulatory enforcement exceeding expectations: strict enforcement of Order No. 16 Article 52 (domestic filing for overseas-bound dramas) would raise overseas supply-chain costs, and the AI clean-up continues to compress domestic ad inventory; ⑤ data caliber risk: no unified official industry statistics, market-size calibers differing by nearly 3x, with systematic bias in elasticity estimates based on third-party data; ⑥ AI content homogenization depressing retention and repurchase, undermining the long-term LTV assumptions of the impulse-purchase model.
Methodology note: This report is a full-archival research study on the short drama and content-export industry, with data as of today (2026-11-15); all market-size figures are labeled with their calibers (Sensor Tower's narrow in-app-purchase caliber leads the figures, with DataEye/MPA and other broad calibers provided as a range reference), and key ratios were recomputed with deterministic tools. A-share concept stocks are discussed only as industry mapping and do not constitute individual stock ratings; for target prices and position recommendations on specific names such as Kuaishou, COL Group, and Zhangyue Technology, a separate in-depth archival study can be commissioned.