Excluded from today's screen — already covered in the last 7 days.
| Company | Researched on |
|---|---|
| SONY FINANCIAL GROUP INC (8729) | 2026-04-27 |
| ASAHI GROUP HLDGS (2502) | 2026-04-28 |
| TOHO CO LTD (9602) | 2026-04-29 |
| NINTENDO CO LTD (7974) | 2026-05-01 |
| GMO INTERNET INC (4784) | 2026-04-30 |
| SONY GROUP CORPORATION (6758) | 2026-05-02 |
| Company | Opportunity | Core moat damage | Rationale |
|---|---|---|---|
| ZOZO INC (3092) Selected | 9 | 2 | Core ZOZOTOWN network effects and brand density appear intact; current pressure is mainly promotion intensity, LYST mix, and weather, with downside cushioned by an asset-light model and variable spend. |
| TOKYO METRO CO LTD (9023) | 8 | 1 | The central Tokyo rail moat is essentially untouched; current weakness is mostly capex timing and input costs, leaving bounded downside and several normalization levers, though regulated pricing caps the upside. |
| BANDAI NAMCO HOLDINGS INC (7832) | 8 | 2 | IP ownership, cross-media scale, and toy/licensing distribution remain intact; present weakness is largely title-mix and amortization timing, while diversified cash generators buffer the left tail. |
| M3 INC (2413) | 7 | 3 | The Japan physician-network moat remains strong and current weakness is largely cyclical and mix-driven, but overseas/services exposure and acquisition-led dilution make the setup less pure than the best opportunities. |
| KOEI TECMO HOLDINGS CO LTD (3635) | 6 | 4 | Catalog monetization and partner relationships provide a floor, but rising development costs and weaker title tails have started to erode the historical mid-scale efficiency advantage. |
| TOKYU CORP (9005) | 5 | 2 | Rail-plus-real-estate control around scarce Tokyu corridors remains intact, but leverage, higher rates, and construction inflation push cash realization out and reduce near-term asymmetry. |
| SEKISUI CHEMICAL CO (4204) | 5 | 4 | Core materials franchises remain defensible, yet housing demographics and diagnostics pricing/churn are structural drags, so upside depends more on cyclical recovery and execution than on true convexity. |
| SUBARU CORPORATION (7270) | 4 | 3 | Brand, dealer network, and AWD/safety differentiation remain intact, but tariff and compliance exposure create a policy-driven, concave earnings setup with limited near-term self-help. |
| WEST JAPAN RAILWAY CO (9021) | 4 | 4 | The rail barrier remains intact, but structural commuter-demand softness and fare-lagged cost inflation weaken monetization of the moat and make upside incremental rather than asymmetric. |
| AIR WATER INC (4088) | 4 | 6 | The installed gas network and switching costs survive, but accounting/governance failures, lower true scale, and higher funding costs meaningfully damage the expansion moat and cap upside to partial normalization. |
| FUJITSU (6702) | 3 | 6 | Japan remains solid, but the UK public-sector trust/procurement moat is structurally impaired and Europe faces a scale-erosion loop until liability clarity and bidding access return. |
| MITSUBISHI MOTOR CORP (7211) | 2 | 7 | ASEAN cost/scale advantages and alliance support are eroding under price wars, underutilization, and EV cost gaps, leaving recovery dependent on hard-to-prove structural fixes. |
| REVOLUTION CO LTD (8894) | 1 | 8 | Crowdfunding trust and capital-access advantages are directly impaired; low switching costs and regulatory scrutiny create a negative feedback loop with little downside protection. |
| ANGES INC (4563) | 1 | 9 | The approval-based moat and partner channel have largely been removed, and future upside is long-dated, binary, and likely diluted by continued financing needs. |
| V-CUBE INC (3681) | 1 | 10 | Product scope, client trust, and scale have already broken down, while solvency and audit stress make further permanent customer loss, dilution, or worse highly plausible. |
Why this company was selected: 3092 offers the best risk-adjusted asymmetry in the set: the core domestic marketplace moat is still intact, the current problems are mostly temporary monetization and integration noise, and the asset-light model plus variable marketing spend bound downside better than the industrial, rail, auto, or distressed platform names. Compared with 9023, the upside is less capped by regulation and capital intensity; compared with 7832 and 2413, the normalization path is simpler and less dependent on hit-driven execution or acquisition mix.
ZOZO is the operator of ZOZOTOWN, Japan’s largest fashion e-commerce platform, plus WEAR, used-fashion services, advertising, and a growing commerce alliance with LINE Yahoo. Economically, it is much better understood as a high-margin fashion marketplace and fulfillment platform than as a classic inventory-heavy retailer. That distinction matters: most of the business runs on consignment-style fees, logistics, and advertising rather than taking large inventory risk.
Data freshness matters here. The latest clean official annual base is FY2025, from the audited annual securities report. More recent data is the FY2026 full-year earnings release disclosed on April 30, 2026. Those FY2026 numbers are official company results, but they are unaudited until the annual securities report is filed, which the company said is scheduled for June 8, 2026. Current share price and market cap below are market data as of May 1, 2026.
| Core metric | Value | Type |
|---|---|---|
| Market cap | About JPY 906.4bn | Market data |
| Net cash / (net debt) | About JPY 49.4bn net cash | FY2026 full-year earnings release, unaudited; cash JPY 69.4bn less short-term borrowings JPY 20.0bn |
| Net income, TTM | JPY 47.9bn | FY2026 full-year earnings release, unaudited |
| Current P/E | About 18.9x | Market data divided by FY2026 EPS from earnings release |
| Normalized P/E | Not meaningfully different; roughly 18.7-18.9x | My estimate; FY2026 included only a small JPY 0.7bn pre-tax exit charge |
| Revenue CAGR | About 9.7% from FY2021 to FY2025 | Audited annual data |
| Net income / EPS CAGR | Net income about 10.0%; EPS about 10.8% from FY2021 to FY2025 | Audited annual data |
Growth has clearly slowed from the market’s old premium-growth narrative, but it has not stopped. FY2026 net sales still rose 7.2% and net income 5.7% on the earnings release. The two concrete drivers are straightforward. First, more buyers: ZOZOTOWN annual buyers rose from 12.2m to 13.2m, and active members from 11.4m to 12.5m. Second, LINE Yahoo commerce kept growing fast: GMV there rose from JPY 69.6bn to JPY 78.9bn, up 13.4%. The weaker counter-trend is that spend per active member is slipping, so growth is coming more from breadth than depth.
| Owner earnings sanity check | JPY bn | Type |
|---|---|---|
| Net income | 47.9 | FY2026 full-year earnings release, unaudited |
| Less sustaining capex | 6-7 | My estimate; recurring fixed/intangible investment excluding the LYST acquisition |
| Plus / minus working capital | Roughly neutral to slightly negative | My estimate; not a structural drain |
| Owner earnings | About 41-43 | My estimate |
| Owner earnings yield | About 4.6%-4.7% | My estimate versus current market cap |
The owner-earnings yield is only modestly below the earnings yield implied by the P/E. That is what you would expect from a business with real but not crushing maintenance capex. ZOZO is asset-light, but not capex-free: software and fulfillment infrastructure are real ongoing costs.
Capital efficiency remains exceptional. ROE has run from roughly 47% to 60% over FY2023-FY2026. Reported group ROIC is less useful because consignment liabilities, excess cash, and acquisition accounting distort the denominator, but the economic point is clear enough: the domestic core earns very high returns on incremental capital. The caveat is equally clear: the same cannot yet be said for the newer global and adjacent bets.
Where profits actually come from is the best quality check. In FY2026, ZOZOTOWN generated JPY 157.4bn of sales, LINE Yahoo commerce JPY 24.2bn, and advertising JPY 11.9bn. LYST contributed only JPY 5.8bn of sales. Direct buy/manufacture was tiny: just JPY 2.6bn of sales and 0.4% of GMV. So the cash engine is still overwhelmingly the domestic platform business. It has been a good business because brands want traffic and outsourced fulfillment, customers want selection and convenience, and ZOZO mostly avoids inventory risk while monetizing data, logistics, and audience reach.
As of May 1, 2026, the stock closed at JPY 1,025. That is down about 38% from the 52-week high of JPY 1,649.5 and only about 3% above the 52-week low of JPY 993.9. The headline move looks like something broke. The actual business data says something narrower: earnings did not collapse; the valuation did.
This has been a re-rating more than an earnings implosion. At FY2025 year-end, official company data implied the stock was trading around 28.2x earnings with a market cap of roughly JPY 1.28tn. Today it is around 18.9x on a larger earnings base. In other words, the market has removed roughly JPY 370bn of equity value even though FY2026 net income increased.
The selloff happened in stages. FY2025 results in April 2025 showed a still-good business, but not the kind of upside surprise a premium multiple needed. Then on July 31, 2025, management revised FY2026 guidance: revenue went up because of LYST, but operating profit and net income went down versus the original plan because LYST came with lower-quality revenue and more amortization. Q2 and Q3 then suffered from warm-weather softness in seasonal apparel and weaker-than-planned promotional effectiveness. By the time FY2026 closed, profit had still grown, but the market had already decided ZOZO should be valued as a mature, high-quality platform rather than a premium growth stock.
(a) One-time / cyclical / sentiment-driven factors
(b) Medium-term business headwinds
(c) Potential long-term structural threats
| Market concern | Quantitative reality check | What it means |
|---|---|---|
| The core business is stalling. | GMV excluding “other” rose from JPY 462.2bn in FY2022 to 501.1bn in FY2023, 536.9bn in FY2024, 574.7bn in FY2025, and 646.2bn in FY2026. | The core is not stalling. It is slowing from earlier peaks, but still compounding. |
| Margins are breaking. | Operating profit rose from JPY 49.7bn in FY2022 to 56.4bn, 60.1bn, 64.8bn, and 69.4bn in FY2026. Operating margin on sales stayed roughly flat at 30%+ across FY2023-FY2026. | Consolidated mix changed, but the domestic operating engine has not broken. |
| Cash generation is deteriorating. | Operating cash flow was JPY 39.9bn in FY2022, 36.7bn in FY2023, 42.6bn in FY2024, 60.1bn in FY2025, and 52.5bn in FY2026. FY2026 year-end cash was still JPY 69.4bn against JPY 20.0bn of short-term borrowings. | This is still a strong cash machine with a cushion, not a stressed balance sheet. |
| Customer quality is deteriorating. | Active members rose over the last eight quarter-ends from 10.92m to 12.48m, but annual spend per active member slipped from JPY 42,947 to JPY 41,323. Annual purchase points per active member fell from 10.9 to 10.6. | This concern is real. Growth is coming from more users, not higher wallet share per user. |
| LYST broke the economics. | In FY2026, LYST contributed JPY 42.2bn of GMV and only JPY 5.8bn of sales. Group gross margin on GMV excluding “other” fell from 34.5% to 33.0%, but SG&A ratio improved from 23.2% to 22.2% and EBIT still rose 7.1%. | LYST is dilutive to mix and visibility. It has not yet damaged the domestic core. |
| The production-business exit signals deeper trouble. | Direct buy/manufacture GMV fell from JPY 3.7bn in FY2025 to JPY 2.8bn in FY2026 and was only 0.4% of FY2026 GMV. The exit charge was JPY 0.7bn pre-tax versus JPY 47.9bn of FY2026 net income. | This was a cleanup of a tiny non-core activity, not damage to the main engine. |
Most of the recent share-price damage is time, not essence. The only true essence issue is that ZOZO’s long-run growth ceiling is lower than the market once hoped. That is different from saying the moat is broken.
| Structural concern | Damaged mechanism | Can time heal this within 3 years? | Classification |
|---|---|---|---|
| Japanese market maturity and demographics | The long-run customer-acquisition funnel and spend-per-buyer ceiling in domestic branded fashion. This lowers growth duration. | No. Demographics do not reverse on a 3-year view. ZOZO can offset some of it through share gains and adjacent categories, but not erase it. | Real structural but survivable |
| Dependence on LINE Yahoo / parent ecosystem | Traffic acquisition economics and minority-shareholder independence. If terms or priorities changed, ZOZO’s economics could be affected. | Not easily. This is a permanent feature of the ownership and distribution structure, though the relationship is currently additive rather than harmful. | Real structural but survivable |
| LYST and lower-quality global expansion | Consolidated take-rate, margin mix, and capital-allocation quality. This does not currently damage the domestic flywheel. | Yes, probably. Within 3 years management can improve monetization, cap investment, or keep the drag contained. The risk is more about capital discipline than franchise breakage. | Not truly structural to the core |
| Low-cost cross-border platforms | The low-end customer-acquisition funnel. These players pressure price-sensitive traffic. | Ongoing pressure, yes. But there is still no hard evidence of irreversible damage to ZOZO’s branded-fashion position. | Not truly structural |
The cleanest diagnosis is this: the core domestic value-creation mechanism remains intact. The real structural issue is a slower runway, not a broken moat. That matters for valuation, but it is not the same as franchise impairment.
Assume you had ZOZO’s current market capitalization in cash, roughly JPY 906bn, and wanted to build a direct competitor from scratch.
| Rebuild horizon | Realistic? | What still blocks you |
|---|---|---|
| 2 years | No | You can build an app, warehouses, and subsidize shipping. You cannot quickly recreate 13.2m annual buyers, 12.5m active members, 1,710 shops, brand trust, fit-and-behavior data, and a working brand/consumer/logistics ecosystem. |
| 5 years | Maybe a credible niche platform, but not a true replacement | The blockers remain brand relationships, customer habit, traffic density, fulfillment quality, secondhand integration, and the LINE Yahoo promotional engine. Capital alone does not buy habit fast enough. |
| 10 years | Yes, in theory | Given enough money and patience, a serious rival could be built. But the economics may still be worse than ZOZO’s because you would likely have to overpay both brands and customers to switch. |
That is the right way to think about the moat. It is real, time-intensive, and category-specific, but not untouchable. Even very well-capitalized Japanese and global platforms have not displaced ZOZO in branded fashion. Still, this is not a regulated utility or a global software monopoly. Time helps ZOZO a lot; it does not make the business unassailable.
Moat & mispricing score: 5/10. ZOZO’s moat in domestic branded fashion is still real. The market is wrong if it treats recent weather softness, a small production-business shutdown, and LYST noise as proof the core franchise has broken. But the market is right to stop valuing ZOZO like a 25-30x premium grower. Long-run growth is slower, the domestic market is maturing, and capital allocation outside the core is not yet proven. The result is a stock that looks roughly fair, not obviously mispriced.
Put differently, the market has probably misdiagnosed the issue type more than the price. This is mostly a time problem in the recent results, but only a modest opportunity in the stock because the multiple has already reset hard.
Valuation basis: This is a FY2026 earnings-release-based intrinsic value estimate. FY2026 figures are official company results from the April 30, 2026 earnings release, but they are unaudited until the annual securities report is filed. I value the operating business off normalized operating owner earnings, then add official FY2026 net cash.
| Case | Normalized operating owner earnings | Required equity yield | Net cash add-back | Equity value | Value per share | Vs. current price |
|---|---|---|---|---|---|---|
| Bear | JPY 38bn | 6.5% | JPY 49bn | About JPY 634bn | About JPY 720/share | About -30% |
| Base | JPY 43bn | 5.0% | JPY 49bn | About JPY 909bn | About JPY 1,030/share | About flat |
| Bull | JPY 46bn | 4.5% | JPY 49bn | About JPY 1.07tn | About JPY 1,210/share | About +18% |
The bridge is simple. Start from FY2026 operating profit of JPY 69.4bn, apply a roughly normal tax load, subtract real sustaining capex, and you get a normalized operating owner-earnings range of about JPY 38-46bn. Then add JPY 49bn of official net cash. At the current price of JPY 1,025, the market is valuing the operating business at about JPY 857bn after net cash, which implies roughly a 5.0% yield on a JPY 43bn base of operating owner earnings. That is close to my base required return.
| Item | Classification |
|---|---|
| FY2025 annual report numbers | Audited annual data |
| FY2026 full-year results | Official company earnings release; unaudited |
| Current share price and market cap | Market data |
| Sustaining capex and owner earnings | My estimate |
| Moat intact but growth runway lower; stock roughly fair | My judgment |
The bottom line: good business, intact core, lower growth ceiling, limited current mispricing. ZOZO looks more like a fairly priced high-quality compounder than a broken franchise or a bargain-bin special situation. This is an intrinsic value estimate, not a price target.
CoffeeAnd — 52-week low lens