SIGMAXYZ HOLDINGS INC

1. Company Overview

SIGMAXYZ Holdings is a Tokyo-listed enterprise-transformation consultancy. It helps large Japanese companies redesign business models, implement SaaS core systems, run complex PMO and change programs, and launch new businesses. The legacy investment business has been shut and is being wound down, so the real analytical question is whether the consulting franchise has merely hit a timing air pocket or whether its earning power has been structurally impaired.

Data freshness. The latest clean official annual base is FY2025. More recent data is FY2026 full-year company results released on 8 May 2026 and June 2026 shareholder materials; for this report I use those figures as company-reported updates, while the long historical base remains the FY2025 audited annual filing. Market data below are current market data as of 23 June 2026.

On the audited FY2020-FY2025 base, revenue compounded at about 10% and net income at about 26%; adjusted EPS compounded at roughly 23%. The growth came from two simple drivers. First, SIGMAXYZ kept adding consultants: 511 at FY2022 year-end, 571 at FY2023, 625 at FY2025, and 692 in the FY2026 company update. Second, it won larger enterprise transformation programs, especially SaaS core-system, PMO, and digital-transformation work. The important point is that FY2026 broke a genuine growth streak; it was not exposing a chronically stagnant business.

Owner earnings sanity check. Starting from FY2026 net income of ¥3.97bn, I deduct sustaining capex of roughly ¥0.25-0.30bn. Depreciation was about ¥0.25bn and total property/equipment purchases were about ¥0.39bn, so that sustaining-capex range is reasonable for an asset-light consultancy. Working-capital movements matter in individual years, but I would not treat the FY2026 receivable release as recurring, so I normalize working-capital impact around zero. That yields rough owner earnings of about ¥3.7-3.8bn, or an owner-earnings yield of roughly 8.8-9.0% at the current market value. That is not meaningfully different from the P/E because the business is light on physical capital and I do not add back share-based compensation.

Capital efficiency. ROE has been consistently high: roughly 21-32% across FY2021-FY2025 audited data, and 27.8% on the FY2026 company update. Reported ROIC is also very high because little capital is tied up in fixed assets. Incremental operating capital is earning high returns. The caveat is that the reinvestment bottleneck is people and client access, not cash. This is a good business, but not a business that can automatically deploy unlimited capital at historical returns.

2. How the Company Makes Money

Almost all of the economics come from consulting. In FY2025 audited data, the consulting segment generated 99.3% of revenue and segment operating income of ¥7.73bn, while the investment segment contributed only ¥0.17bn of revenue and lost ¥0.37bn before being discontinued. In plain English: profits come from advising and executing difficult corporate change programs, not from venture-style investment gains.

The key nuance is revenue mix. SIGMAXYZ often uses outside specialists on large implementations. That outsourced work inflates reported revenue but carries much lower margin than the in-house layer that designs the program, manages stakeholders, governs execution, and de-risks delivery. When a large project reaches go-live, low-margin pass-through revenue can disappear even if the high-value consulting franchise is intact. That is exactly why FY2026 revenue fell sharply while gross profit held roughly flat and operating profit still rose.

Where do profits actually come from? From the higher-value parts of enterprise change: SaaS core-system renewal, program management, business-process redesign, and “transformation sherpa” work with large clients in transportation, finance, information and communications, retail, trading, and construction. This is closer to trusted change execution than to commodity IT staffing.

Why has this been a good business? It is asset-light, it sits close to real management agendas, and switching costs are meaningful once a complex program is underway. Client satisfaction has stayed extremely high at 97 points, which helps repeat work. But the moat is only medium-depth. The barriers are trust, delivery reputation, reference cases, and talent density. They are not patents, regulation, proprietary data, or a network monopoly. That distinction matters because it limits how much of the historical ROE should be treated as permanent franchise value.

3. Why the Stock Fell

The stock has de-rated violently. It traded around ¥1,331 at the 52-week high on 12 June 2025 and around ¥518 on 23 June 2026, only a few yen above the ¥513 52-week low set on 3 June 2026. That is a fall of roughly 61% in one year.

The first break was the November 2025 guidance cut. Management reduced FY2026 revenue guidance from ¥30.0bn to ¥24.5bn, operating profit from ¥6.95bn to ¥6.10bn, and net income from ¥4.90bn to ¥4.40bn. The stated reasons were concrete: sale of a consolidated subsidiary, delays in the start of new projects after several large projects went live, and removal of expected investment-security sales from guidance. For a stock that had been priced for smooth, high-ROE growth, that was a credibility hit.

The second leg lower came when FY2026 results confirmed that revenue and bottom-line growth had indeed broken. FY2026 revenue came in at ¥23.83bn, down 9.4%, and net income fell 9.6% to ¥3.97bn. Operating profit still rose to a record ¥6.06bn, but the market mostly looked through that because cost of revenue fell 17% as outsourcing dropped roughly 40%. Investors read the result as: demand was softer than hoped, margins were being protected by mix and cost rather than by clean top-line strength, and the old growth narrative was over.

There was also non-core noise. The old investment book created extraordinary losses during the cleanup, and after year-end management began building a stake in Core Concept Technologies, adding a new capital-allocation question on top of the operating one. None of that breaks the core consulting franchise, but it does make the equity messier and lowers the multiple investors are willing to pay.

4. What the Market Is Assuming

(a) One-time / cyclical / sentiment-driven factors

• The FY2026 revenue drop is not just mix; it reflects genuine demand weakness.
• The gap between go-live on large projects and the start of replacement projects will last longer than management says.
• The FY2026 revenue miss means FY2027 guidance is at risk.
• The old investment-book losses will keep muddying reported earnings.

(b) Medium-term business headwinds

• Lower outsourcing means reported sales growth will stay slower even if profit holds.
• Internal utilization can remain uneven when large projects roll off.
• Project timing and client concentration make results lumpier than the market previously assumed.
• Buybacks and equity shrinkage may be flattering ROE more than true organic operating improvement.

(c) Potential long-term structural threats

• AI will commoditize lower-value consulting hours and pressure pricing.
• The talent model may weaken if top consultants become harder to recruit or retain.
• Capital allocation could drift away from the clean consulting model into minority stakes and alliances.
• The moat may be narrower than historical margins suggest because the business is still fundamentally people-based.

5. Temporary or Structural?

Reality check versus the market narrative.

Structural risks only.

The bottom line: the core value-creation mechanism has not been broken. The current evidence says SIGMAXYZ can still win and deliver difficult enterprise-change work. What has been damaged is the market’s willingness to value that work as a smooth, high-growth compounder. The most important structural risk is not “consulting disappears”; it is that lower-value, hour-based work becomes less valuable and forces the firm to earn its returns through harder, scarcer, more outcome-linked work. That is a real risk, but it is not the same as franchise collapse.

6. Is the Market Wrong? By How Much?

This looks mainly like time, not essence. The revenue air pocket, subsidiary exits, and investment-book cleanup are temporary issues. The structural caveat is real: AI will pressure lower-value consulting hours, and the company will need to keep shifting toward higher-value transformation work. I think the market is too pessimistic on the operating franchise, but it is right to refuse the old 20x-plus multiple.

Time-as-a-moat test. Assume you had the company’s current market capitalization, roughly ¥42bn, in cash.

Valuation. This is a FY2026 company-reported full-year valuation adjusted with the latest capital-allocation updates. It is an intrinsic-value range, not a price target.

The bridge is straightforward. I start with owner earnings rather than reported net income because capex is low and the balance sheet is clean. The non-operating asset adjustment is a conservative haircut to excess cash and securities, including the remaining former-investment-book assets and the post-year-end Core Concept Technologies stake. I do not give full credit for headline asset values because realization timing, taxes, and management’s future use of that capital are uncertain.

At roughly ¥42bn market cap, the stock implies about a 9%-plus owner-earnings yield on my normalized range. For a debt-free consultancy with still-strong margins, improving client breadth, and a real but not deep moat, that is cheap enough to be interesting but not so cheap that the market must be obviously wrong. My base case says the market is underpricing the business by about ¥7bn, or roughly ¥80 per share. That is a real gap, but not a giant one.

7. Key Facts, Estimates, and Judgments

Moat & mispricing score: 6/10. The moat is real but not impregnable. The market is getting wrong the idea that a 9% revenue decline automatically means the consulting franchise is impaired; most of the evidence points to lower outsourcing pass-through, subsidiary exits, and temporary utilization gaps rather than client rejection. The market is correctly skeptical that a people-based consultancy facing AI-driven pricing pressure deserves the old peak multiple. So this is not a 10/10 mispricing. It is a good business with a medium moat, temporarily messy reported numbers, and a stock that looks somewhat cheap rather than absurdly cheap.