Financials

Financials in One Page

Auros is a sub-₩260 billion KOSDAQ small-cap designing wafer overlay and inspection tools — a niche of semiconductor capital equipment so cyclical that its FY2025 and FY2024 income statements look like two different companies. Revenue ran from ₩39.5B (FY21) to a record ₩61.4B (FY24), then dropped 15% to ₩52.1B in FY25 as Korean memory capex paused. Gross margin compressed from a 66.8% peak in FY23 to 48.6% in FY25; operating income flipped from +₩6.1B (a 9.9% margin) to −₩8.4B in twelve months. Reported earnings have never converted cleanly to cash — free cash flow has been negative in four of the last five years — and the working-capital build (AR + inventory tied up roughly ₩34B by FY24) plus a ₩14.7B FY25 capex program have flipped the balance sheet from net cash to net debt for the first time. The market currently prices the equity at ~4.0x book and ~4.9x EV/sales, discounts to global peers (KLA, Onto, Nova, Camtek trade at 13–20x EV/sales) — but peers earn 25–40% operating margins while Auros earns negative ones. The single financial metric that matters most over the next two quarters is gross margin recovery alongside inventory drawdown.

Terms the reader needs once:

• Free cash flow (FCF) — operating cash flow minus capital expenditures. The cash left over after the company has paid to run the business and to maintain or expand its plant.
• Net debt — total debt minus cash and short-term investments. A positive number means the company owes more than it has on hand.
• ROE / ROIC — return on equity / return on invested capital. How much profit the company generates per won of shareholder capital, or per won of total capital deployed.
• EV/Sales — enterprise value divided by revenue. Useful when a business is losing money and you cannot use a P/E ratio.

Revenue, Margins, and Earnings Power

Auros sells lumpy capital equipment to a small set of front-end memory fabs (the FY2025 사업보고서 names Korean memory and a single foreign foundry as the customer concentration). Its income statement therefore reads as order-cycle revenue with a fixed-cost overlay — operating expenses run ₩30–34B every year regardless, so gross-margin and revenue swings drop almost entirely to the operating line.

Two things stand out. First, the FY24 result was a peak, not a steady state: revenue jumped 35% YoY off a strong Korean memory cycle and operating margin nearly doubled to 9.9%. Second, the FY25 collapse was sharper than the top line implies — revenue fell 15% but operating income swung by ₩14.5B (about 28 points of margin) because the cost base did not flex.

The gross-margin chart is the single most important slide in this report. Auros earned a 66.8% gross margin in FY23 — close to what KLA does globally — but lost 18 points of gross margin by FY25. That cannot be explained by volume; revenue is roughly 14% higher than FY23. It is a mix and pricing story: a heavier weight of newer package-inspection products, a heavier mix of standard tools to a smaller customer base, and absorbed warranty and start-up costs on tools shipped into a slowing cycle. The earnings-quality test in the next section confirms this is not a one-quarter blip.

The quarterly view shows the cycle clearly: revenue ramped through FY24, peaked at ₩28.9B in 4Q24 (47% of full-year revenue in one quarter — a sign of order timing, not steady demand), then fell in five of the next six readings, troughing at ₩7.6B in 3Q25. 4Q25 ticked back up to ₩13.8B but operating income stayed negative. Earnings power is genuine when the cycle is up, but the operating leverage cuts both ways and the company has no recurring-revenue floor.

Cash Flow and Earnings Quality

Reported net income is not a reliable proxy for cash at Auros. Working capital absorbs cash every up-cycle (receivables and inventory grow with bookings) and only partially releases it in down-cycles. Heavy capex on its own metrology platforms compounds the gap.

The chart tells a hard story. FY2024 — the company's best reported year — was its worst on operating cash flow, with OCF of −₩5.3B against net income of +₩5.9B. The reason is straightforward: receivables jumped ₩8.6B and inventory jumped ₩10.6B as the cycle peaked. Free cash flow was negative in every year except FY21, including +₩5.9B and +₩3.4B reported-profit years. The cumulative 5-year free cash flow is roughly −₩37 billion, against ₩4.7B of cumulative reported net income.

Read the table this way: a negative number in Change in inventory or Change in receivables means cash was consumed (those balance-sheet items grew). FY24 alone consumed ₩19.2B of working capital. FY25 finally released ₩6.3B of inventory but consumed another ₩14.7B in capex — twice the prior-year average — for a new tool platform. Result: even the inventory unwind in FY25 was swallowed by capital investment.

The FCF margin chart is the cleanest summary of earnings quality. A semicap peer like Nova converts 25% of revenue to free cash flow every year; KLA, almost 31%. Auros has averaged −18% FCF margin over the last four years. Until working capital normalizes and capex falls, reported profits should not be treated as available cash.

Balance Sheet and Financial Resilience

For four years Auros's balance sheet was its strongest feature — IPO proceeds left ₩34B of cash and short-term investments against zero long-term debt in FY21. That cushion has now been spent.

Five-year arc: from ₩32B net cash (FY21) to ₩4.5B net debt (FY25). Total debt grew from zero in FY23 to ₩16.2B by year-end FY25, and the borrowing is overwhelmingly short-term (₩14.2B of ₩16.2B, 88%) — meaning it has to be rolled annually. With FY25 EBITDA negative, traditional "net debt / EBITDA" is not meaningful; the relevant question is liquidity coverage.

The quick ratio (current assets excluding inventory, divided by current liabilities) collapsed from 4.56x in FY21 to 1.04x in FY25 — just barely above the 1.0x line that defines short-term solvency without selling inventory. The balance sheet is still solvent — ₩63.5B of equity remains, intangibles are modest at ₩2B, and there is no goodwill — but the margin of safety has gone from extraordinary to thin in two years.

Returns, Reinvestment, and Capital Allocation

Returns on capital tell the same cyclical story as the income statement but with more pronounced highs and lows because the equity base is small.

Peak ROE was 9.1% (FY24) — a respectable single year for a small-cap industrial but well below the 20–30% returns the global metrology peers print steadily. Through-cycle ROE looks closer to 1–2%, almost entirely consumed by the FY22 and FY25 losses. A business with 60%+ peak gross margins should not be ending five years at break-even on capital; the problem is operating-expense intensity (research and SGA combined run 55–65% of revenue) on a revenue base that is too small to absorb it.

The chart frames the full IPO cycle. FY21 imported ₩39.8B of fresh equity from the IPO. Management then ran modest opportunistic buybacks in FY22–FY24 (about ₩5B cumulative — meaningful given the float), and recently bought back zero. Capex spiked to ₩14.7B in FY25 — by far the largest year — funded by new short-term debt. The capital story is now reinvest-first, return-cash-second, and the company has used essentially all of its IPO cash.

Diluted shares have been essentially flat at 9.2–9.3 million since the IPO — no large secondary, no SBC-driven dilution, and the small buyback programs offset stock-based compensation. Book value per share traced an arc from ₩7,128 (FY21) → ₩6,584 (FY22 loss) → ₩6,699 → ₩7,367 (FY24) → ₩6,874 (FY25 loss). Net, five years after the IPO, book value per share is essentially unchanged. That is the cleanest summary of capital allocation: management has not destroyed per-share value, but has not compounded it either.

Segment and Unit Economics

Auros does not publish a segment income statement. Public filings disclose product-family revenue (wafer overlay metrology, wafer inspection, package metrology, package inspection) and customer concentration in qualitative terms, but no segment operating income, asset base, or geography split is staged in the financial files. The mix story is therefore narrative, not numerical:

• The legacy wafer overlay metrology business is the historical profit engine — high gross margin, single-domestic-vendor position competing with KLA at Korean memory fabs.
• Newer package inspection / advanced-packaging products (WaPIS, AEliT, METIS-PS family) are the FY24–FY25 revenue accelerant and the most likely source of margin compression: lower-margin entry products into a more crowded TAM (Camtek, Onto are direct competitors here).

If management discloses segment economics in FY26 — particularly the gross margin of advanced-packaging products vs. core overlay — that would be the single most decision-useful upgrade to this analysis. As staged today, the segment numerical work cannot be done.

Valuation and Market Expectations

You cannot use a P/E ratio on Auros — FY25 was a loss year and the FY22 loss makes any five-year average meaningless. The two metrics that work are P/B (because tangible book value is real and the cycle will recover) and EV/Sales (because revenue is the cleanest available scale of the franchise).

At ₩27,450, the stock trades around 4.0x book and 4.95x EV/sales — close to the 5-year average on both metrics. The market is pricing a recovery, not capitulation. That is the central valuation tension: investors are paying a peak-cycle P/B multiple while the company is printing trough-cycle margins.

Three scenarios frame the range. The base case below uses through-cycle FY24-style economics (revenue near ₩60B, 8% operating margin, 12% normalized FCF margin) and a target multiple of 12x normalized EBIT, discounted to today.

In English: at today's price, the base-case outcome is essentially fair value — the market is already discounting a cycle recovery in line with FY24's economics. The bear case (no recovery, additional working-capital damage) implies roughly ₩14,000 a share, a 50% downside. The bull case (memory upcycle plus successful packaging-platform wins lifting through-cycle revenue 15% above prior peak) implies roughly ₩38,000, a 38% upside. Asymmetry is currently unattractive at the entry price — pay-off is roughly 1-for-1 but only if the bull-case operational improvements happen.

Peer Financial Comparison

Each peer is shown in its own reporting currency (USD for KLA, Onto, Nova, Camtek; KRW for Auros and Park Systems) to avoid stale FX conversions inside a moving comparison. Ratios, margins, and multiples are unit-free and directly comparable.

Two patterns matter. First, every metrology specialist except Auros is currently profitable, and profitably so — Nova at 29% operating margin and ~25% FCF margin is the closest peer-equivalent margin profile, and it trades at 18x EV/sales. KLA, the franchise that Auros directly competes against in its core overlay product, prints 39% operating margins and trades at 20x EV/sales but with a 52x P/B (a different conversation about ROE durability). Second, Park Systems — the only listed Korean small-cap metrology peer — currently runs a 20% operating margin and trades at 10x EV/sales and 9.4x P/B. Park Systems is what Auros has to look like operationally before its valuation can re-rate toward Korean specialist levels.

The peer gap is real but earned: Auros is not currently a peer of these companies on quality. If margin recovery happens, the discount should compress; if it does not, the discount is justified.

What to Watch in the Financials

What the financials confirm: Auros has a real specialty-equipment franchise with cyclical high-margin earnings power (66.8% gross margin at peak, 9.9% operating margin at peak). The IPO balance sheet kept the company alive through the FY22 downturn.

What the financials contradict: the narrative that Auros is a cash-generative compounder. Cumulative five-year free cash flow is roughly −₩37 billion, and reported earnings exaggerate the cash economics by at least an order of magnitude. The new net-debt position also contradicts the "fortress balance sheet" story that held through FY23.

The first financial metric to watch is gross margin in 1Q26: if it prints above 55%, the FY25 margin damage looks mix/absorption-driven and reversible; if it stays below 50%, the franchise looks structurally weaker than its FY21–FY23 history implied, and the current price-to-book is hard to defend.