ALAB: Scorpio breaks out; the stock still needs a cash-flow breakthrough

The 60-second read

Five things to know

1. Scorpio becomes the largest product family. The X-Series production ramp drives a step change in revenue; product-level dollars and profits are not disclosed.
2. Operating leverage outweighs a lower gross margin. GAAP operating income reached $179.1M even as gross margin fell to 72.3%.
3. Cash conversion trails the earnings headline. Receivables and inventory consumed $347.4M over nine months; FCF was $268.1M before acquisitions.
4. Dilution belongs in the valuation. Q4 guidance uses 187M diluted shares; our model uses 190M and retains compensation expense in operating margins.
5. The price asks for more than another good quarter. Our base cash model values the stock near $141; more generous earnings and sales routes lift the rounded blended value to $165.

Three scenarios, one probability-weighted number

Illustrative 12-month decision scenarios, not company guidance or a guaranteed trading range.

Signposts: what would change our mind

Dates are analysis checkpoints, not issuer-confirmed earnings appointments. Concentration is measured by direct customer, often a manufacturing partner; it is not a named hyperscaler market-share estimate.

The tape

The price used throughout this packet is the November 9, 2026 regular-session close of $216.87 from FMP, independently matching its daily history response. The quote was retrieved on November 10 UTC. We calculate close-to-close returns directly: the first full session after the release fell 4.4%, and the stock was 7.2% below the release-day close by November 9. The latest daily bounce therefore does not mean that investors rewarded the print overall. Revenue above the company’s own guidance can coexist with a falling share price when expectations and valuation were already demanding.

Our market capitalization uses the filing cover count of 174.782075M common shares outstanding on October 31, rather than the provider’s older implied share base. That produces $37.90B of basic equity value and $36.53B of enterprise value after September cash and marketable securities. This bridge deliberately combines dated observations and is not described as a freshly reported November balance sheet. The forward valuation denominator is a separate, conservative 190M diluted-share assumption. Mixing those denominators would make a multiple appear cheaper without any business improvement.

The print

Revenue of $584.899M grew 153.7% from the year-earlier quarter and 49.1% sequentially. Gross profit rose to $422.752M, and GAAP operating income reached $179.056M. This is real operating leverage: operating expenses grew far more slowly than revenue, even while research spending accelerated. The improvement is not merely a non-GAAP presentation. GAAP operating margin increased to 30.6%, compared with 24.0% a year ago and 22.7% in the preceding quarter.

The August 4 company release guided third-quarter revenue to $540–560M. Reported sales of $584.899M exceeded that range’s high end by $24.899M. This comparison is anchored to dated issuer guidance published before the quarter’s result. The FMP earnings snapshot was retrieved after the release; it does not establish the estimate’s pre-release timing or a reconciled EPS accounting basis. We therefore make no consensus revenue or earnings beat claim. Reported GAAP EPS was $0.99 and company non-GAAP EPS was $1.32; those are distinct actual measures, not evidence of a like-for-like estimate comparison. More useful for the next quarter, management guided revenue to $730–750M, implying 26.5% sequential growth at the midpoint. Adding that midpoint to reported nine-month sales gives an illustrative full-year total of $2,025.660M, not separately issued full-year guidance.

Products and concentration

Scorpio became Astera Labs’ largest product family by revenue during the quarter, led by high-volume deployment of the X-Series 320-lane fabric switch. That changes the investment debate from a narrow retimer supplier to a broader connectivity platform. Aries 7 extends signal conditioning across copper and optical connections, while Leo memory products and the Orion reference architecture broaden possible content per accelerator. These are real product and design developments, but introductions and design wins are not equivalent to recognized revenue or a contracted backlog.

The company has one operating segment and does not disclose quarterly revenue or profit by product family. We therefore do not manufacture a Scorpio revenue stack or treat product announcements as segment earnings. The geographic table uses customer billing addresses, including manufacturing partners and distributors. Singapore revenue of $310.237M is not a measurement of Singapore end-user AI spending. Likewise, Customer A accounted for 53% of revenue and 58% of receivables; this is a direct-customer concentration, not proof that any named hyperscaler buys precisely that share.

The economics cut both ways. A successful customer deployment can quickly absorb engineering costs and produce exceptional incremental profit. The same exposure makes shipment timing, qualification delays and purchasing negotiations powerful determinants of quarterly results. The filing warns that customers can shift purchasing between manufacturing partners, so a falling direct-customer concentration does not automatically prove end-customer diversification. We need both broader design activity and sustained cash collection before treating diversification as established.

Profit and tax bridge

The cleanest quarterly bridge begins at non-GAAP operating income of $261.646M and deducts $82.590M of stock compensation to reach $179.056M of GAAP operating income. There was no quarterly acquisition-cost adjustment. Interest and other income of $12.971M takes pretax profit to $192.027M; the $9.083M tax provision leaves $182.944M of GAAP net income. There is no minority-interest deduction in this consolidated bridge. The interest line is a benefit from the liquid balance sheet, and must not be capitalized in enterprise operating earnings and then counted again as cash.

At the net-income level, $182.944M plus $82.590M of compensation and minus a $22.134M tax effect equals $243.400M of non-GAAP net income. That tax adjustment is negative: removing compensation does not allow us to keep all the associated tax benefits. In the preceding quarter GAAP net income actually exceeded non-GAAP net income because the GAAP tax benefit was especially large. Presenting adjusted earnings as uniformly more conservative, or adding every tax line in the same direction, would distort the comparison.

Over nine months the company recorded a $48.076M GAAP tax benefit, helping net income exceed pretax income. Management attributes tax-rate differences to valuation allowances, equity-compensation tax benefits, foreign-derived intangible income and research credits. Future release of a valuation allowance could create another accounting benefit, but is not a recurring sales margin. Our forward cash and earnings models use a normalized 20% tax charge. That is an assumption, with upside if favorable cash taxes last longer, rather than a forecast copied from the current year’s benefit.

Earnings quality

Stock compensation is noncash at grant-date accounting recognition, but it transfers a claim on future value. Quarterly expense of $82.590M was 14.1% of sales; its dollar cost roughly doubled from the year-earlier quarter even as its revenue percentage declined. There were 8.463M unvested RSUs and $802.9M of unrecognized RSU expense, plus $30.7M related to performance awards. These amounts show why compensation should not be erased from long-run economics. We retain its expense in modeled GAAP margins and use a forward share base to reflect outstanding awards, without mechanically escalating shares again for every future compensation dollar.

Customer warrants are a separate issue. The quarter included about $18.1M of contra revenue, versus $1.9M a year earlier. The nine-month charge was $30.397M. It reduces reported revenue and is added back in the operating cash-flow reconciliation because it is noncash at recognition. It is not included in the company’s regular stock-compensation add-back. We retain the reported revenue treatment in our models rather than create a higher unofficial sales measure. The 2026 customer warrant covers up to 3.262299M shares with a $142.82 exercise price and purchase-based vesting conditions; not all of that maximum should be assumed immediately issued.

Receivables reached $284.253M and inventory $200.893M. Quarter-end balances imply 44.7 receivable days and 114.0 inventory days using the quarter’s sales and cost of sales over 92 days. Using beginning/end averages instead produces 37.5 and 89.3 days. We show the convention because rapid growth makes the choice consequential. Work in progress accounts for $160.735M of inventory, consistent with a production ramp but also exposing capital to a demand reversal. These indicators warrant monitoring; they do not by themselves establish bad debt or obsolete inventory.

Cash and balance sheet

Nine-month operating cash flow was $341.477M, and cash purchases of property and equipment were $73.363M. Our free-cash-flow definition is the difference, $268.114M. It excludes acquisitions and financing flows. The company spent another $69.214M on business combinations; subtracting that leaves $198.900M before other investing and financing activity. The quarterly cash-flow statement is derived by subtracting the six-month totals from nine-month totals: Q3 OCF was $179.201M and cash capex $45.309M, leaving $133.892M. These are calculated quarter values, not a separately issued standalone cash-flow statement.

Cash flow benefited from $195.495M of noncash compensation and $30.397M of warrant charges. Receivables and inventory used $201.245M and $146.173M respectively, partially offset by payables and accrued liabilities. Subtracting compensation from reported FCF gives a conservative economic cross-check of $72.619M over nine months. This is not another reported cash metric and should not be mistaken for money that literally left the bank. It asks what remains if owners recognize the compensation burden instead of treating it as free.

Cash and marketable securities totaled $1,378.243M at September 30. No financial debt is reported, but the company has operating leases, purchase obligations and accrued capital expenditures. Marketable securities include corporate debt and are not identical to insured bank cash. Our enterprise-value bridge subtracts the liquid portfolio while keeping ordinary working-capital liabilities within operating economics; operating lease costs remain in modeled margins. We do not deduct every liability again after charging those expenses, or count the same security as both cash and a separate investment asset.

Valuation and the price-implied case

Our valuation is intentionally forward-looking. A trailing multiple cannot represent a business whose quarterly sales are changing this quickly. The base forecast assumes $3.4B of revenue in 2027, roughly matching the retrieved $3.390B consensus, followed by $4.4B, $5.4B, $6.4B and $7.4B through 2031. GAAP operating margins are 33%, 35%, then 36%. Thus the cautious call does not require weak growth: 2027 sales rise 67.8% from the illustrative 2026 guidance-based total. It requires that growth have a cost and that the stock’s starting price matter.

The DCF charges 20% tax on operating income, adds depreciation of 2% of sales, deducts cash capex of 4% of sales and invests 15% of incremental sales in working capital. It discounts year-end cash flows at 10%, uses 4% terminal growth and recalculates terminal reinvestment at that slower growth rate. September liquid assets are added once. Dividing by 190M forward shares gives $141.23. About 82.0% of enterprise value lies in the terminal component, so this is highly sensitive to durable margins and the discount rate. We use complete annual periods as a simplified valuation convention, without pretending that a partial-quarter calendar adjustment adds certainty.

The earnings route assumes $1,122M of 2027 GAAP operating income plus $50M of interest and other income, taxed at 20%. Net income of $937.6M over 190M shares gives $4.93 per share. A generous 40-times multiple yields $197.39. The sales route applies 10 times $3.4B of revenue, adds $1,378.243M of net cash and divides by 190M shares to reach $186.20. Sales multiples embed profit assumptions rather than replace them. Weighting DCF 50%, earnings 30% and sales 20% gives $167.07, rounded to a $165 base estimate. The rounding is editorial, not hidden model precision.

At $216.87 and 190M modeled shares, forward equity value is $41.21B and enterprise value $39.83B. Holding our margins, taxes, capex and working-capital rules constant, scaling the whole revenue path until DCF equals that price requires approximately $5.34B in 2027 and $11.62B in 2031. This is one conditional reverse-DCF solution, not the only path or management guidance. Better margins, a lower discount rate or slower dilution could also close the gap. The bull case is possible; the current price makes much more of it necessary than our base underwriting accepts.

Compared with Wall Street

FMP’s aggregate price target is $363.36, with a $325 median and a $275–500 range. The aggregation does not expose every contributor’s effective date. It should not be presented as a clean post-print consensus. The dated November 4 observations are more useful: Northland $315, Needham $300 and UBS $290. Those are provider-reported targets, not our independent access to the underlying research reports. The news records describe reductions at Needham and UBS while Northland upgraded its rating. A high average target therefore masks different actions and potentially stale contributions.

The revenue estimate for 2027 is $3.390B across 18 contributors, very close to our $3.4B assumption. The disagreement is not principally next year’s sales. FMP’s $7.08 EPS estimate does not document a reconciled GAAP basis; we do not multiply it by a GAAP multiple and claim an apples-to-apples valuation. Our $4.93 earnings route explicitly charges compensation and normal tax. For 2030, the provider’s average revenue is $7.442B, above our $6.4B forecast. Faster sustained growth would improve value, while the more distant estimate has only ten revenue contributors and should carry less confidence.

Management and capital allocation

Management’s delivery earns credit. The prior release guided third-quarter revenue to $540–560M, and actual revenue exceeded the high end. The Scorpio ramp supports the claim that newer products are contributing materially, rather than relying solely on an addressable-market presentation. At the same time, the new fourth-quarter guide assumes gross margin near 72%, not a return to the year-earlier mid-seventies. Scale is improving operating profit despite mix and warrant effects on gross margin. A fair scorecard must recognize both developments.

Research and development expense rose to $170.957M in the quarter, including $49.395M of compensation. Investment can sustain product leadership, but the return must appear in future customer deployments. The February acquisition had $74.0M of consideration, mostly allocated to goodwill; another smaller acquisition followed in May. Goodwill rose to $91.557M. These transactions are ordinary capital-allocation decisions, not an eligibility exclusion, but their accounting does not prove that promised technology synergies will produce economic returns.

The company reports effective disclosure controls and disclosed an ERP implementation during the quarter. That is an implementation risk to monitor, not evidence of a present material weakness. Common stock has one vote per share; there is no ADS conversion in this analysis. PricewaterhouseCoopers LLP signed the 2025 annual audit. We found no current filing-delay issue preventing this period’s analysis. The company’s annual risk disclosures and quarterly updates remain the source of structural and concentration risks; historical IPO-related expenses are retained in the dataset rather than projected as recurring cash charges.

Risks and the strongest counterargument

The strongest bull case is that inference and training systems need much more connectivity per accelerator, while Astera’s software and interoperability expertise increase its content and switching costs. Scorpio’s ramp could be the start of a broader platform cycle. If new products diversify revenue, gross margin stabilizes and the company maintains expense discipline, our assumed revenue path could prove too low. The balance sheet supplies resilience and funding flexibility. A valuation critique should not erase these competitive and financial advantages.

The bear case does not require the AI industry to stop growing. Large buyers can change architectures, internalize components, negotiate price reductions or shift deployment timing. Competition and standards transitions can compress margins while engineering costs remain high. Customer and supplier concentration can turn a single delay into a large quarterly miss. A stock priced for sustained compounding can fall even when reported revenue is still growing quickly. That is why we distinguish an attractive technology market from an attractive entry valuation.

Taiwan manufacturing exposure, export controls, tariffs and geopolitical disruption remain consequential. We do not assign a fabricated probability to a conflict or count all China billing revenue as prohibited business. The financial model is also vulnerable to lower gross margins, additional equity awards, working-capital intensity and a valuation multiple reset. These risks interact: a slower ramp can leave inventory elevated just as investors demand a higher return. The scenario range is deliberately wide and is not a stop-loss mechanism or assurance of limited downside.

Catalysts and decisions

The next major checkpoint is the fourth-quarter release and annual filing, expected in early 2027. FMP lists February 9 as a future date, but we have not treated that provider calendar entry as issuer confirmation. The immediate tests are revenue within the $730–750M range, gross margin around 72%, compensation near the guided $79M and diluted shares near 187M. More growth with materially worse working capital would be a mixed outcome, not an unqualified victory.

For a more constructive rating we need either a lower price with the operating thesis intact or stronger evidence that durable cash economics exceed this model. A potential entry near $124 represents a 25% discount to the $165 base value, not an automatic order or a forecast that the market must visit that price. Conversely, revenue below guidance, inventory days above 140 or a persistent compensation burden above 16% of revenue would pressure the valuation even after a price decline. The framework changes when evidence changes; it does not mechanically average up with every price target.

Data, sources and method

All financial statement amounts below are US dollars in millions, except per-share data and share counts, which are explicitly labeled. The quarter dataset begins with Q3 2024 and includes nine observations through Q3 2026. Fourth-quarter income-statement amounts are computed as annual totals less nine-month totals when a direct tagged quarter is unavailable. Quarterly cash flow is computed from successive cumulative statements. Non-GAAP figures and detailed reconciliation rows are retained from the contemporaneous SEC-filed earnings releases, including negative tax adjustments and early GAAP losses.

Annual cash flow covers 2023, 2024 and 2025. The 2024 compensation surge includes IPO-related accounting and is not treated as a steady-state expense rate. Early loss-period GAAP diluted shares exclude antidilutive instruments, while non-GAAP pro forma denominators can differ. We retain those distinctions in the source dataset; dividing adjusted profit by the loss-period GAAP share count would overstate comparability. There is no invented product profit allocation, backlog figure, end-customer identity or management long-range revenue commitment.

Computed means arithmetic from reported or assumed inputs, not an additional audited disclosure. Free cash flow here is OCF minus cash property-and-equipment purchases; valuation cash flow is after normalized operating tax, reinvestment and compensation. Neither should be conflated with GAAP net income. DCF means discounted cash flow; enterprise value is equity value less liquid assets plus relevant financial claims. Our three valuation routes are cross-checks, not statistically independent forecasts. All forward assumptions, scenario probabilities and rating judgments belong to Charged Alpha.

The source archive includes the current 10-Q, all nine quarter releases, intervening annual and quarterly filings, SEC company facts and dated FMP responses. Prior-quarter research was not recovered as a verified published packet in this candidate; no historical Charged Alpha forecast performance is asserted. This packet establishes the dated signposts that the next update should grade. The current quarter is fully identified as the September 30, 2026 period; the release date is November 3 and the market reference date is November 9. Keeping those dates separate prevents a price move or a guidance update from being assigned to the wrong quarter.

• quote.json
• profile.json
• historical-price-eod_full.json
• grades.json
• grades-consensus.json
• price-target-consensus.json
• price-target-news.json
• analyst-estimates.json
• earnings.json
• 0001736297-26-000045_alab-20260930.htm
• 0001736297-26-000035_alab-20260630.htm
• 0001736297-26-000020_alab-20260331.htm
• 0001736297-26-000010_alab-20251231.htm
• 0001736297-25-000147_alab-20250930.htm
• 0001736297-25-000098_alab-20250630.htm
• 0001736297-25-000031_alab-20250331.htm
• 0001736297-25-000003_alab-20241231.htm
• 0001736297-24-000043_alab-20240930.htm
• 0001736297-24-000018_alab-20240630.htm
• 0001736297-24-000006_alab-20240331.htm
• 2026-11-03_release.htm
• 2026-08-04_release.htm
• 2026-05-05_release.htm
• 2026-02-10_release.htm
• 2025-11-04_release.htm
• 2025-08-05_release.htm
• 2025-05-06_release.htm
• 2025-02-10_release.htm
• 2024-11-04_release.htm
• 2024-08-06_release.htm
• 2024-05-07_release.htm