← Back to 52-Week Low Lens

ALPHA PURCHASE CO LTD

1. Company Overview

Alpha Purchase Co., Ltd. is a Japanese B2B outsourcing company focused on two operational headaches that large organizations do not want to manage themselves. The first is indirect procurement: buying thousands of low-ticket, non-core items such as tools, safety supplies, office equipment, repair parts, and other MRO goods. The second is facility management for chain-store and commercial operators: repairs, maintenance, cleaning, store-related works, and renovation support. ASKUL is the controlling shareholder, owning roughly 62%.

The important point is that this is not a consumer-facing e-commerce story. It is an enterprise workflow and outsourcing business. Customers use Alpha Purchase because large organizations need one approved counterparty, controlled purchasing rules, ERP connectivity, and a catalog that can aggregate many suppliers under one contract and payment flow. That makes the core MRO business much stickier than the low reported operating margin would suggest. The weaker side of the company is the FM segment, which is more project-based, more operationally messy, and far less attractive economically.

2. How the Company Makes Money

Alpha Purchase makes money mainly from its MRO segment. In 2025, MRO produced JPY 44.3bn of revenue and JPY 1.19bn of segment profit, versus FM at JPY 14.6bn of revenue and JPY 0.20bn of segment profit. In other words, about three quarters of sales came from MRO, but more than 80% of segment profit came from MRO. If the investor is buying anything of real quality here, it is the enterprise procurement platform, not the FM business.

The accounting needs one clarification. Management states that roughly 99% of MRO sales are booked as gross merchandise sales rather than a net commission. System-use fees are less than 1% of MRO revenue. That means reported revenue is large, but margins are structurally thin because Alpha Purchase is recognized as the seller of record. Revenue growth therefore matters less than gross profit and operating profit growth.

What is driving growth? Two things, and only two really matter. First, existing enterprise customers ramp spend over several years after onboarding as more subsidiaries, locations, users, and approved suppliers are added. Management’s own cohort material shows that customer revenue often takes years to mature. Second, the newer “Infinite Catalog” recommendation and substitution logic has improved MRO gross profit by steering spend toward cheaper or better-stocked supplier options while keeping Alpha Purchase as the approved counterparty.

Owner earnings sanity check. A rough 2025 view is: net income of JPY 1.03bn, plus depreciation and amortization of roughly JPY 0.69bn, less sustaining software capex of roughly JPY 0.65bn to JPY 0.70bn, less a normalized working-capital drag of roughly JPY 0.10bn to JPY 0.20bn. That gives rough owner earnings of around JPY 0.8bn to JPY 0.9bn. On today’s market cap, that is an owner earnings yield of roughly 5.0% to 5.5%.

Is this meaningfully different from P/E? Yes. The stock looks cheaper on earnings than on owner earnings because software investment is real, even when capitalized, and working capital can swing materially. Reported earnings yield is about 6.5%, but normalized owner earnings yield is closer to 5%-plus. That gap matters.

Capital efficiency. The group-level ROE is solid rather than extraordinary. The better way to read the business is that MRO is a good business sitting inside a mixed-quality group. Incremental capital in MRO appears to earn high returns because the platform is embedded, fixed assets are mostly software, and supplier payables fund part of the operating base. Incremental capital in FM does not look nearly as good.

Business quality. Profits actually come from enterprise MRO outsourcing. This has been a good business because the value proposition is operational control, not low price alone: one approved vendor, ERP connectivity, standardized purchasing workflows, supplier normalization, and low large-customer churn. Management has disclosed that no customer spending more than JPY 10m per month canceled in the last four years. That is the moat signal that matters. The moat is process integration and trust, not brand glamour.

3. Why the Stock Fell

The stock fell because expectations reset much faster than the business deteriorated. The shares peaked around JPY 3,845 in August 2025 and have since fallen to roughly JPY 1,627, a decline of about 58%, leaving the stock near the bottom end of its past year’s range.

Three things drove the decline. First, the stock had become expensive after strong first-half 2025 results and a guidance raise; at the peak, the market was paying roughly 36x trailing earnings. Second, in October 2025 ASKUL suffered a ransomware attack that halted shipments. Alpha Purchase said its own systems were unaffected, but about 6.3% of procurement came from ASKUL and some customers delayed purchases while waiting for ASKUL-sourced items to return. Third, the FM segment stayed weak and low quality, which made investors question whether the group deserved a premium multiple at all.

The market’s plain-English worry is this: maybe the MRO business is less durable than it looked, more dependent on ASKUL than investors thought, and paired with an FM segment that drags on cash generation and quality. A second worry is that 2025’s margin improvement may prove less durable than the share price once implied.

4. What the Market Is Assuming

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

• The ASKUL ransomware disruption will leave a longer tail than management admits.
• The 2025 stock collapse is partly a reversal of an overheated 2025 rerating.
• FM project timing and emergency costs made 2025 look worse than underlying demand.

(b) Medium-term business headwinds

• Large-customer acquisition may be slowing; the count of very large MRO accounts was flat at 42 in 2025.
• The “Infinite Catalog” may lower customer prices faster than it raises Alpha Purchase’s gross profit.
• Cash conversion may stay weaker than accounting earnings because software capex and working capital absorb cash.
• FM may continue to dilute group returns even if MRO remains healthy.

(c) Potential long-term structural threats

• ERP, procurement SaaS, or large marketplace players could replicate Alpha Purchase’s enterprise procurement workflow.
• ASKUL control creates a permanent governance and counterparty discount.
• The FM segment may simply be a structurally lower-return business with limited moat.

5. Temporary or Structural?

Reality check versus the market narrative. The data say the business is mixed, not broken.

Structural diagnosis, focusing only on structural risks.

• FM business quality. The damaged mechanism is utilization and pricing in chain-store maintenance and project work. This segment is exposed to customers delaying renovations when stores are busy, and to costly emergency responses when demand returns suddenly. It does not damage the core MRO value-creation mechanism, but it does weaken group economics. Time could help a little within three years, but not fully, because the problem is industry structure and business mix, not just timing. Classification: real structural but survivable.
• Enterprise procurement disintermediation. The damaged mechanism would be the customer acquisition and retention funnel if ERP suites, procurement SaaS vendors, or large marketplaces could offer the same integration, approval workflow, catalog normalization, and vendor-of-record convenience. If that happened, it would be hard to reverse within three years. But the current evidence does not show it happening: revenue is still growing, large-customer churn is negligible, and MRO margins improved in 2025. Classification: not truly structural yet.
• ASKUL dependence and control. The damaged mechanism is supply continuity perception and strategic independence. Operational supply dependence is lower than the market feared, because only about 6.3% of procurement came from ASKUL, but the October 2025 incident proved that even limited exposure can disrupt customer behavior. Ownership dependence is not realistically reversible within three years unless ASKUL changes its stake. This does not appear to weaken the customer moat directly, but it is a permanent quality discount on the equity. Classification: real structural but survivable for the business; structural for valuation.

Bottom line: the core MRO issue is time, not essence. The FM issue is closer to essence, but it is not the core profit engine. The stock’s problem is that a good MRO business sits inside a mixed-quality group structure.

6. Is the Market Wrong? By How Much?

Time-as-a-moat test. If I had the current market cap in cash, I could not realistically rebuild the MRO business in two years. I could write software, hire salespeople, and even assemble supplier feeds, but I could not replicate years of ERP integrations, enterprise approvals, product master-data normalization, and trust with blue-chip customers that quickly.

• Within 2 years: No, not for MRO. Parts of FM could be rebuilt, but not the embedded procurement workflow.
• Within 5 years: Partially. A strong competitor, especially an ERP or marketplace player, could build a serious alternative. What would still block me is long enterprise sales cycles, installed approval workflows, and switching friction at large customers.
• Within 10 years: Yes, probably. That means the moat is real, but mostly time-based rather than absolute.

At today’s roughly JPY 16.1bn market cap, the stock implies an earnings yield of about 6.5%, but only about a 5.0% to 5.5% owner earnings yield on a rough normalized owner earnings estimate of JPY 0.8bn to JPY 0.9bn. If one simply subtracts reported net cash, the market is valuing the operating business at roughly JPY 10.7bn. Even after haircutting cash because a meaningful portion supports working capital, the implied value of the operating business is still only around the low-to-mid teens in billions of yen.

That is too low if one believes the MRO moat is intact and the ASKUL shock was temporary. It is not absurdly low, because the market is correctly charging for three real issues: FM is a weak segment, software upkeep is a real expense, and ASKUL control is a permanent discount. My judgment is that the market is somewhat wrong, not wildly wrong. It is pricing too much permanence into the 2025 ASKUL disruption and not enough stickiness into the MRO installed base.

Mispricing in yen terms. I would frame the gap as roughly JPY 2bn to JPY 4bn of undervaluation at the equity level, not a dramatic multiple-bagging setup. Put differently, the market today seems to be valuing the strong MRO core only a little above what I would assign to a no-growth, mixed-quality operating business, despite evidence of continued customer stickiness and growth. That is a modest but real disconnect.

Moat & mispricing score: 6/10. The moat in MRO is real: customer workflows are embedded, large-customer churn is near zero, and the business still grew through a supplier shock. But the moat is not wide enough to ignore FM weakness, low reported margins, capitalized software spend, and parent-control risk. The market is getting wrong the permanence of the 2025 disruption; it is getting right that this is not a pristine, high-conviction compounder. This is a moderate time-mispricing, not a clear-cut essence-mispricing.

7. Key Facts, Estimates, and Judgments

Facts

• 2025 revenue was JPY 58.9bn, operating profit JPY 1.47bn, and net income JPY 1.03bn.
• MRO generated JPY 44.3bn of revenue and JPY 1.19bn of segment profit in 2025; FM generated JPY 14.6bn and JPY 0.20bn.
• Cash at 2025 year-end was JPY 5.36bn and debt was negligible.
• The stock is down about 58% from the August 2025 high.
• ASKUL’s ransomware incident in October 2025 disrupted Alpha Purchase even though Alpha Purchase’s own systems were independent.
• Management has disclosed zero cancellations in the last four years among customers spending more than JPY 10m per month.
• Stock-based compensation does not appear material; dilution has been modest.

Estimates

• Normalized owner earnings are roughly JPY 0.8bn to JPY 0.9bn.
• Economically excess cash is lower than reported net cash because the business needs a meaningful working-capital buffer.
• Operating ROIC is likely above 20%, but public disclosure does not allow a precise number.
• The equity appears undervalued by roughly JPY 2bn to JPY 4bn.

Judgments

• The core MRO damage is temporary, not structural.
• FM weakness is structural, but survivable because it is not the main profit engine.
• The company’s real quality is better than the current stock narrative suggests, but worse than the 2025 peak valuation implied.
• This is not a broken business. It is a good MRO business inside a mixed-quality group, currently priced with too much fear about a temporary shock and about the wrong segment.

CoffeeAnd — 52-week low lens