Thread of my ideas shared on this platform

Another Japanese cybersecurity stock: Cyber Security Cloud $4493. A ¥18B small-cap running Japan's leading web-application-firewall SaaS: Shadan-kun guards SME websites, while WafCharm uses AI and threat intelligence to automate the cloud WAFs running on AWS, Azure and GCP. It's ~90% recurring revenue with roughly 1% monthly churn — sticky subscription plumbing that sits in front of a customer's website and rarely gets ripped out. 66% gross margins, a 22% return on equity, and revenue still compounding at 42%, the fastest grower on my list. This is the opposite end of the Trend Micro trade. Where Trend Micro is mature, cash-rich and pays you to wait, Cyber Security Cloud is small and all-growth: ¥18B buys the whole business, there's almost no net cash to hide behind (~¥455M), and you're paying ~21x forward earnings / ~13x EBITDA / ~3.4x sales. Flip the multiple and it's a ~5% earnings yield. A token ~0.3% dividend at a 6% payout tells you the character: cash gets plowed straight back into 40% growth. Operating margins have compressed hard and trailing earnings fell sharply even as revenue surged — growth is being bought, not banked yet. Add small-cap liquidity and a thin balance-sheet cushion, and this is the higher-risk, higher-variance end of the theme.

Trend Micro $4704 ¥766B global cybersecurity company, founded in Japan in 1988 and now one of the oldest pure-play security vendors in the world. It protects enterprise endpoints, servers, and hybrid-cloud workloads through subscription software (XDR, threat detection, cloud security) — the kind of mission-critical, contract-locked tooling that's deeply embedded in IT stacks and painful to rip out. The economics show it: 77% gross margins, a 34% return on equity, and earnings that grew 34.7% last year. The reason it's interesting is the price. ¥766B buys the whole business, but ~¥229B of that — nearly a third of the market cap — is just net cash sitting in the till, with zero debt. Strip it out and you're paying an enterprise value of only ~¥537B for a global security platform: roughly 6.3x EBITDA and under 2x sales. Flip the ~20x forward earnings multiple and that's a ~5% earnings yield on a 34% ROE compounder — at a time when US cyber peers like CrowdStrike, Palo Alto and Zscaler trade north of 50x earnings. Shareholders are paid to wait, too: a ~3.1% dividend at a ~71% payout plus steady buybacks, a ~4.5% shareholder yield, all resting on a fortress balance sheet. The catch is growth. Revenue at only ~9%, and the stock sits near its 52-week low (¥5,900 against a ¥10,935 high) as the market frets about share loss to faster-moving rivals and Microsoft's security bundling. Not investment advice.

Intelligent Wave $4847 This ¥25B small-cap quietly runs ~70–80% of Japan's credit-card authorization processing — the invisible rails nearly every card transaction in the country clears through. Deeply embedded, high switching costs, recurring revenue. FY24 operating profit grew 30.5%, and management targets ¥30B sales by 2030 (~16% CAGR). ¥24.6B is the price for the whole business — and ~¥4.8B of that (≈20% of the market cap) is just net cash in the till, zero debt. Net it out and you're paying ~5.6x EBITDA / ~1.2x sales for the dominant payment rails of an entire country. Flip the 13.5x forward multiple and that's a ~7% earnings yield on a 14% ROE compounder — while comparable Japanese software names routinely sit north of 20x. ~3.9% dividend yield at a ~72% payout, no buybacks. The watch-item is concentration: Japan-centric revenue and exposure to payment-tech shifts — the price of a small-cap that owns one country's plumbing rather than many. Not investment advice.

keep in mind anytime a VC throws around a gen z slang its OFF LIMITS FROM THEN ON

Growth rate per capita / P/E of the listed Exchange Poland and Greece are the standouts

$7735 SCREEN Holdings While the crowd piles into the obvious semicap names, SCREEN controls ~80% of the global single-wafer cleaning market — an unglamorous but mission-critical step that scales with every new node and every HBM stack feeding AI demand. That dominance shows up in the numbers: ~26% operating margins, sticky customers, and recent quarterly sales of ¥183B ( 32% YoY). Yet it still trades around 14–15x forward earnings vs peers at 19–21x — a 20% ROE compounder riding the AI/memory capex wave at a discount. On capital return: a conservative ~1.8% dividend at a steady ~30% payout — but the real lever is the ¥221B net-cash pile, reinvestment-led with room for opportunistic buybacks. You're paid to wait while it compounds, not for income. The watch-item is China (~49% of semi revenue): equal parts opportunity and export-control risk. A best-in-class GARP setup in semis.

It's pretty incredible that we can create custom visuals of pretty much anything we could ever want PLUG YOUR AGENT INTO FINANCIAL DATA

Hardest problem for small mid cap is a free news feed that actually sources good, relevant news

The best opportunities I'm looking at in Japan & Korea: 1. SCREEN Holdings (7735.T) — 91.6 · Semis Dominates single-wafer cleaning (~80% share), 26% op margins, riding the semicap up-cycle. Valuation: fwd P/E 14.4x · EV/EBITDA 15.8x · rev 9% · ROE 20% · net cash ¥221B. Re-rated 🔺 91% since pitch but still ~14x vs peers at 19-21x. Capital return: ~1.8% yield, conservative ~30% payout held steady through the cycle. The ¥221B net-cash pile is the real lever — funds room for opportunistic buybacks; reinvestment-led, not income. 2. Intelligent Wave (4847.T) — 90.1 · Software 70-80% share of Japan's credit-card authorization processing — sticky, non-bank payments infrastructure. Valuation: fwd P/E 13.6x · EV/EBITDA 5.7x · rev 15% · ROE 14% · net cash ¥4.8B. Cheap on both metrics; no re-rating. Capital return: income-style — ~4.0% yield on a high ~72% payout. Returns most of earnings as dividends; you get paid well to hold while it compounds. 3. OBIC (4684.T) — 89.6 · IT Services #1 ERP for Japanese SMEs (~30% share), 63% operating margin, 99% retention, fortress balance sheet, cloud optionality. Valuation: fwd P/E 25.0x · EV/EBITDA 16.6x · rev 11% · ROE 16% · net cash ¥207B. Re-rated 🔻 -26% — cheaper entry on a premium compounder. Capital return: ~2.4% yield, steadily rising dividend at ~50% payout. The ¥207B net cash is famously under-deployed — the bull case includes management finally returning more (buybacks / higher payout) under TSE governance pressure. 4. M&A Capital Partners (6080.T) — 89.1 · Capital Markets #1 domestic M&A intermediary for Japan's SME-succession wave. Asset-light, 81.5% equity ratio, zero long-term debt. Valuation: fwd P/E 16.7x · EV/EBITDA 6.2x · rev 82% · ROE 14% · net cash ¥44.8B. Capital return: ~2.0% yield, low ~28% payout — earnings mostly retained to fund the growing deal pipeline. Asset-light model throws off cash; return is via growth a modest, growing dividend. 5. SK hynix (000660.KS) — 89.0 · Semis HBM memory leader (MR-MUF packaging edge) with real pricing power on AI-driven, contractual demand; 60-70% HBM gross margins. Valuation: fwd P/E 5.4x · EV/EBITDA 15.6x · rev 198% · ROE 61% · net cash ₩32.5T. The 5.4x is on peak-cycle earnings — cyclical cheapness, not structural. Capital return: token ~0.1% yield, ~3% payout — cash is consumed by HBM capex. Policy is a fixed annual dividend floor (₩1,200/sh) plus cyclical top-ups. Buy this for the cycle, not the cheque. 6. Fukui Computer (9790.T) — 88.6 · Software Regional near-monopoly in construction CAD/software, 25% op margins, 70% recurring revenue, 18-22% 3yr CAGR. Valuation: fwd P/E 15.9x · EV/EBITDA 5.3x · rev 26% · ROE 15% · net cash ¥24.1B. One of the cleanest cheap-quality names here. Capital return: ~2.5% yield, ~35% payout — steady, well-covered dividend backed by recurring revenue and net cash. Balanced reinvest-and-pay profile. 7. Japan Exchange Group (8697.T) — 88.1 · Capital Markets Monopoly operator of the Tokyo & Osaka exchanges — the quintessential toll-road / market-infrastructure moat. Valuation: fwd P/E 32.4x · EV/EBITDA 15.2x · rev 45% · ROE 23% · net cash ¥58.0B. Best-in-class moat, but priced for it. Capital return: dividend-led — ~2.9% yield on a high ~80% payout. Capital-light monopoly returns the bulk of profit; progressive dividend policy. The reliability is the point. 8. NE Inc. (441A.T) — 87.8 · Software Japan's dominant e-commerce order-management hub (~33% share, 6,570 merchants), 40% EBIT margin, 110% ROIC, <1% monthly churn. Valuation: P/E 9.0x (ttm) · ROE 29% · net cash ¥3.0B. Re-rated 🔻 -26%; screens very cheap for the quality (small-cap, ¥8.7B mcap). Capital return: no dividend yet — recent IPO retaining 100% to reinvest at 110% ROIC. Return is pure compounding for now; a payout could initiate as growth matures. Watch for first dividend as a re-rating catalyst. 9. Miroku Jyoho Service (9928.T) — 87.5 · Software ~30% share of accounting-firm ERP, 99% renewal rates, capital-light, cloud transition lifting ARR ( 35%). Valuation: fwd P/E 10.8x · EV/EBITDA 5.1x · rev 9% · ROE 17% · net cash ¥8.9B. Cheap on a 99%-renewal recurring base. Capital return: shareholder-friendly — ~34% base payout topped up with special dividends (which inflate the trailing yield to ~7%). Recurring cash funds a generous, growing distribution. Normalize the yield to ~3% for planning. 10. RS Technologies (3445.T) — 86.6 · Semis Global leader in wafer reclaiming (~33% share), ~10x sales growth in a decade, less cyclical than prime wafers. Valuation: fwd P/E 21.3x · EV/EBITDA 9.6x · rev 9% · ROE 9% · net cash ¥80.6B. Re-rated 🔺 44% — less of a bargain, still reasonable. Capital return: low ~0.8% yield, ~13-30% target payout — cash is reinvested into wafer-reclaim capacity expansion. Growth-first; the dividend is a token, not the thesis. 11. Plus Alpha Consulting (4071.T) — 86.1 · Software Enterprise HR-SaaS leader (Talent Palette), ~50% margins, high switching costs from 6,000 features, ~20% CAGR. Valuation: fwd P/E 21.3x · EV/EBITDA 12.9x · rev 14% · ROE 30% · net cash ¥15.5B. Premium multiple justified by the ROE and stickiness. Capital return: ~3.7% yield, ~31% payout, plus an initiated ~6% buyback. Founder owns ~40% — interests aligned, and the buyback signals confidence. Dividend repurchase combo. 12. Cyber Security Cloud (4493.T) — 86.1 · Software WAF SaaS (Shadan-kun / WafCharm), 90% recurring, ~1% monthly churn, 6 straight years of 25% revenue & op-profit growth. Valuation: fwd P/E 21.6x · EV/EBITDA 13.0x · rev 42% · ROE 22% · net cash ¥455M. Fastest grower in the group (small-cap, ¥18.2B). Capital return: minimal ~0.3% yield, ~6% payout, plus a ¥450M buyback. Cash reinvested into 40% growth — return is overwhelmingly via compounding, with the buyback a small capital-discipline signal. ★ Trend Micro (4704.T) — 85.5* · Software (Cybersecurity) Global cybersecurity leader (endpoint, cloud, network security) with deep enterprise installed base and recurring subscription revenue. Valuation: fwd P/E 20.8x · EV/EBITDA 6.8x · rev 9% · ROE 34% · op margin 21% · net cash ¥229B (zero debt). The 6.8x EV/EBITDA 34% ROE net cash is the standout — much cheaper on cash-flow than the P/E suggests. Added pick; score estimated, not from the original screen. Capital return: dividend-led — ~3.0% yield on a generous ~71% payout, supplemented by periodic buybacks. Mature, cash-generative model that returns the bulk of earnings while still growing. Strongest income profile of the cybersecurity names here.

I have to say, Fable can write pretty good investment research BUT you have to plug it in to live financials for it to produce accurate financials x.com/FirstHillcap/status/20…

i fear the substack notes feed is turning into Linkedin if it hasn't already

119850.K META 005935 Kioxia 8299.TW All interesting

Someone find those WSB edits of the ape army

7735.T interesting

AI still can't make investment decisions but it is very good at finding specific ideas and lookalike companies - the kind of stuff that takes forever like going A-Z. If you can just make good decisions I think a lot of the hard work is over, esp as models get better. Going A-Z through an exchange for ideas could not be a better LLM task - Synthesize a large data set where the cost of a few errors is meaningless. Can even load Claude code with up to date financial data for every stock. Any one else working on this and getting good results?

$GPW was top of my list for anyone hunting under a $1b market cap. cheap PE, plenty of growth, untapped upside in data sales in a growing economy. has a public stock exchange ever gone bankrupt? (not as cheap as it was now)

wrote this one up a while back. a japanese company with a $531m market cap sitting on $520m of cash and $800m of marketable securities. yes those numbers were right. and they returned cash to shareholders. rerated a lot since

Li Lu on how to invest under $100k: interesting one, in Korea there's a form of securities called preferred securities. basically non voting common stock that pay a slightly higher dividend. not much more, but no voting rights. cheap way to own the same business

You can borrow at 1.8% per year to buy a Japanese company that pays you 8% per year. Costs you nothing