NAURA Technology Group SUN-R Investment Analysis

If the trend runs another three to five years, the industry structure changes in two ways. China’s equipment wallet becomes less concentrated in a handful of foreign vendors and more split among several domestic champions, with NAURA as the broadest front-end platform, AMEC stronger in etch, ACM Research Shanghai and others stronger in selected clean or electrochemical steps, and more listed consolidation around packaging and subsystem niches. And the competitive battleground moves from basic availability to process depth: who can win higher-layer memory, advanced packaging, specialty materials compatibility, lower defectivity and the fastest yield ramps. NAURA’s breadth earns it a seat at that table. It does not guarantee a win.

Usage and Unit Economics

For a stock like NAURA, “usage” means revenue, installations, customer validation, repeat procurement and cash conversion. On those measures the business is undeniably real. In 2025, revenue reached CNY 39.35 billion, up 30.85% on the adjusted base; electronic process equipment contributed CNY 36.73 billion, or about 93% of total revenue, with a gross margin of 39.18%. Electronic components contributed CNY 2.58 billion at a gross margin of 52.95%. In 2026 Q1, revenue rose to CNY 10.32 billion, up 25.80%, while operating cash flow improved to CNY 748 million from negative CNY 1.73 billion a year earlier.

The 2026 Q1 income statement shows why this is more than a simple operating-leverage story. Revenue of CNY 10.32 billion against operating cost of CNY 6.11 billion implies a quarterly gross margin of about 40.8%, broadly consistent with 2025 segment economics, but R&D expense rose to CNY 1.40 billion, up 36.64% year on year. That is why net profit grew only 3.42% despite healthy top-line growth. The right reading: management is still choosing share and category expansion over near-term earnings smoothing. The model is not broken.

A few disclosed numbers reveal the economic model clearly enough to estimate beta. Because electronic process equipment generated CNY 36.73 billion of 2025 revenue at 39.18% gross margin, every 10% swing in that segment’s revenue would move segment revenue by about CNY 3.67 billion and gross profit by about CNY 1.44 billion before any offset from opex and mix. Every 1 percentage point move in that segment’s gross margin changes annual gross profit by roughly CNY 367 million on the 2025 base. This is why the stock responds violently to changes in fab-capex expectations and customer order timing. The usage is real, but it is still capex-sensitive industrial usage.

Shipment scale and application breadth show that customers have moved past trialing tools. NAURA’s 2025 report says etch equipment revenue exceeded CNY 10 billion, thin-film deposition revenue also exceeded CNY 10 billion, and the company had batch-selling products across 12-inch logic and memory as well as advanced packaging, compound semis, 8-inch specialty processes and LED applications. The report also details 12-inch etch tools for logic and memory, 12-inch ALD and PECVD tools for storage and advanced packaging, 12-inch LPCVD and tube-CVD platforms, 12-inch EPI, 12-inch electroplating tools, and several 8-inch tool families. This is the footprint of a supplier already embedded in production lines, not a subsidy-built ghost supplier.

The company does not disclose order backlog the way some Western industrial firms do, so the nearest public proxy is contract liabilities and customer concentration. Contract liabilities were CNY 4.29 billion at the end of 2025 and CNY 4.20 billion at the end of 2026 Q1. The top five customers accounted for 39.03% of 2025 sales, the largest at 12.72% and the second largest at 11.25%. Orders are meaningful and concentrated enough to matter, yet no single customer alone defines the company. The flip side is that the stock will react sharply to capex cadence at a handful of national-champion fabs.

The quiet strength in order stability is timing. In a 2025 SSE filing by a supplier serving NAURA and other domestic equipment makers, the issuer stated that semiconductor-equipment procurement is usually arranged one to two years in advance to match future capacity expansion, and that the major domestic equipment customers were actively expanding output. This is not NAURA guidance and should be used carefully, but it supports the common industry view that visible shipments understate the planning visibility upstream.

What I do not see is fake prosperity in the classic sense. The filings carry no public evidence of related-party sales inflating revenue; the annual report says top-five customer related-party sales comprised 0% of annual sales, and top-five supplier related-party purchases comprised 0% of annual procurement. The business is cyclical and policy-supported, but the revenue looks like real equipment shipments into real fabs rather than circular transactions or one-off subsidy dressing.

Narrative Liquidity

NAURA’s one-line story is simple enough for the market to trade: China’s broadest domestic chip-tool platform, exposed to localization and AI-driven memory / logic capex. That is a powerful A-share narrative because it compresses policy, technology sovereignty and earnings momentum into one sentence. It also travels well across both institutional and retail frames, as a national substitute, an AI infrastructure supplier, and one of the few listed names with real exposure to CXMT / YMTC / logic-capex themes.

The narrative is healthy rather than empty because it rests on disclosed usage. NAURA’s 2025 filing describes two tool families with revenue above CNY 10 billion, several categories with share gains, and shipment milestones above 1,000 units for etch, PVD and vertical furnaces. Those are the facts a durable narrative needs. The dangerous case is one where price outruns proof; here the price is rich, but the proof is there.

There is still narrative risk, because the story is now widely understood. Reuters’ quote page shows NAURA on around 48.3x forward earnings and about 11.9x sales, while AMEC screens at around 79.5x forward earnings and 22.8x sales, and ACM Research Shanghai at about 112.8x adjusted P/E and 21.9x sales. So NAURA is not the cheapest domestic equipment name on relative multiples, yet it is not priced like AMEC or ACM Research Shanghai on the most stretched metrics either. The market is already paying for quality without yet assigning NAURA the highest scarcity multiple in the domestic semi-equipment cohort.

The disagreement is over what the narrative is worth. Bulls argue NAURA deserves a durable premium because platform breadth lowers single-tool risk and opens share gains across several steps. Bears counter that the narrative crams too many good things into one price: memory cycle strength, policy enforcement, high-end technical catch-up and successful integration of new categories. I sit closer to the middle. The narrative still has room to run if new categories turn into real profit pools, but the stock already reflects most of the easier “localization + AI = buy tools” trade.

Bridge

In SUN-R terms, NAURA is one of the clearer bridges from old-world capital into the domestic semiconductor localization trade. This is a state-backed listed company, not a private lab story. The controlling shareholder is Beijing Sevenstar, while the actual controller is Beijing Electronics Holding; Hong Kong Securities Clearing held 13.75% at 2025 year-end, the National Integrated Circuit Industry Investment Fund held 5.00%, and the second-phase Big Fund vehicle also appeared among the leading holders. That share register matters because NAURA already sits in the overlap between local SOE capital, national industrial capital and public-market liquidity.

The bridge is also operational. NAURA is a current vendor to strategic domestic fabs, not a future aspirant, and it has used the A-share market as a consolidation and financing platform. In 2025 it acquired control of Kingsemi after buying 9.49% via negotiated transfer and 8.41% via public solicitation transfer, reaching about 17.87% ownership but majority board control after board reconstitution. That transaction carried unusual weight because it widened NAURA’s product coverage into photoresist track and advanced-packaging tools without waiting for in-house development cycles.

The same bridge logic appears in CXMT’s IPO process. CXMT’s prospectus says the company is China’s largest and most advanced DRAM IDM, already global number four by capacity, and that its listing would fund line upgrades and forward R&D while also supporting upstream semiconductor equipment and parts makers. The Shanghai Stock Exchange shows the application was reviewed by the Listing Committee on 2026-05-27, and the CSRC disclosed its registration approval on 2026-06-12. A major domestic memory champion is moving deeper into public-market financing just as domestic tool intensity rises. That is precisely the institutional bridge NAURA shareholders want to see.

The bridge to CXMT is commercial and ecosystem-based rather than equity-based. The disclosed CXMT shareholder list around the major holders names GigaDevice, CMB-related vehicles, Beijing Peak Benefit, Alibaba-network affiliates, Xiaomi-related funds and others, but I do not find NAURA named as a direct major shareholder in the prospectus excerpts available through the SSE filing. NAURA’s annual report does show director Yuan Xun also serving as a director at CXMT subsidiaries ChangXin Xinqiao and ChangXin Jidian. That falls short of proving exclusive supply or a locked commercial relationship, but it confirms NAURA is embedded in the same state-directed industrial ecosystem, not standing outside it.

Anti-cycle

The industry phase looks more like recovery-to-expansion than late bubble. SEMI forecast continued WFE growth into 2026 and 2027, SEAJ showed 2025 global semiconductor-equipment billings rose 15% to $135.1 billion, and China remained the largest regional market at $49.3 billion. Reuters separately reported that China would still lead chipmaking-equipment investment in 2025, helped by policy support and AI-linked capacity demand. None of that describes a clean bottom. It describes an upcycle with policy support under it.

If the industry falls sharply, NAURA is part victim, part buyer. Victim because customer capex still drives the order book, contract liabilities are finite, and the multiple is too high to shrug off a 50% capex scare. Buyer because the balance sheet is deeper than most domestic peers, and management has already shown a willingness to consolidate strategic assets, Kingsemi above all. End-2025 cash and cash equivalents were CNY 17.18 billion, and 2025 financing cash flow was strongly positive. That gives NAURA real optionality in a stress period, though it stops well short of immunity to earnings downgrades.

Counter-cyclical expansion widens the moat only when it adds missing process steps or hard-to-build validation assets. Kingsemi clears that test. A broad domestic platform becomes structurally more valuable in downturns because fabs standardize vendors, preserve engineering resources and prefer suppliers able to support several adjacent processes. In the next downcycle, the thing to watch is whether NAURA buys assets that deepen recipe integration, service density and customer share of wallet, rather than cheap assets for their own sake.

Regulatory and Ruin Risk

The main risks are real and visible. Each one runs a different path from operating event to valuation damage.

Regulatory and sanctions risk carries medium probability and high severity. Reuters reported that the U.S. included NAURA in the December 2024 restrictions targeting part of China’s semiconductor tool chain, while a later Reuters report described China’s response as a stronger domestic-equipment push. What matters is whether NAURA keeps qualifying replacement components, sustains deliveries into restricted applications and holds its new-product launch cadence; press rhetoric is noise. When this risk fires hard, tool deliveries slip and engineering cost rises first, then the multiple compresses as investors stop paying for “smooth localization.”

Technology-iteration risk carries medium probability and high severity. The company has genuine scale in mature and mid-advanced steps, but the industry’s profit pool keeps moving toward tighter geometries, higher aspect ratios, stricter defect control, more layers in memory and harder advanced-packaging recipes. The signals to watch are yield-related customer wins, sustained shipment milestones beyond legacy nodes, and wider disclosed product coverage in 12-inch memory and advanced packaging. If this risk fires, NAURA does not go bankrupt. It becomes a lower-ceiling domestic industrial supplier, and the valuation slides toward a much lower industrial multiple.

Order-concentration and capex-cycle risk carries high probability and medium-to-high severity. The top five customers were 39.03% of 2025 sales, contract liabilities were CNY 4.29 billion at year-end and CNY 4.20 billion in Q1, and the business remains exposed to the investment cadence of a handful of large fabs. Here the signal is the combination of contract-liability drift, operating-cash-flow quality and supplier commentary on upstream procurement timing. When this risk fires, revenue downgrades hit quickly and operating leverage works in reverse, because R&D is sticky.

Valuation reflexivity carries high probability and medium severity on ordinary setbacks, but high severity on a true cycle break. Reuters shows the stock had moved from a 52-week low near CNY 300 to above CNY 700 at the high, and the 2026-06-16 quote still implied about 48x forward earnings. A rising price hands management acquisition currency, lifts employee confidence and broadens the buyer base. Run that in reverse and even a modest miss can trigger both estimate cuts and a lower multiple. The signal is a widening gap between revenue growth and profit growth, much like what Q1 already showed when R&D ran ahead of earnings.

Integration and governance risk carries medium probability and medium severity. The Kingsemi acquisition is strategically sound, but the synergies are not automatic: the acquired portfolio, customers and operating culture still have to be integrated without losing speed. The governance structure itself is fairly plain for an A-share SOE-backed industrial company. The bigger question is whether capital allocation keeps favoring process-depth assets over empire-building, which means watching whether NAURA can expand cross-selling into track / packaging while keeping returns on incremental capital sensible. If this risk fires, the damage usually comes slower, as the stock quietly stops earning a scarcity premium.

Liquidity risk in the stock is low probability and low severity for ordinary investors. NAURA is a very large-cap A-share name with active turnover; Reuters showed nearly 8.7 million shares of volume on 2026-06-16. The liquidity that actually matters here is valuation liquidity: whether buyers still pay growth-stock multiples if semiconductor policy rhetoric cools or the memory cycle softens.

Key-person and reputation risk carries low-to-medium probability and medium severity. NAURA is not a one-founder cult stock, which helps. Even so, the company has become symbolically important in China’s “domestic ASML” discourse, and Reuters quoted chairman Zhao Jinrong in a March 2026 call for stronger national support to build lithography capabilities. The signal is whether public ambition keeps matching engineering delivery. If symbol outruns substance, the stock can derate sharply even with the business intact.

Ruin risk, in the strict sense, is low. The realistic worst case is a very large permanent-loss scenario, not a literal zero from today: advanced-category catch-up disappoints, memory / logic capex cools, export controls tighten around critical parts, and the market rerates NAURA from a national-platform multiple toward a mature industrial-equipment multiple. In that state, a 40% to 60% drawdown is easier to imagine than insolvency. A true zero would take a far darker stack of events: sanctions that significantly break sourcing, a collapse in customer qualification, plus failed integration and governance damage. Public evidence today does not point there, but the framework still requires spelling out the path.

SUN-R score

I would score NAURA this way.

• Structural shift: 14/15. China’s equipment localization push is real, policy-backed and already visible in market structure.
• Real usage: 18/20. Revenue, shipment milestones and product breadth show a real installed-base business, not an aspirational one.
• Unit economics: 12/15. Segment gross margins remain strong, but heavy R&D means earnings lag revenue when management accelerates new categories.
• Network effects: 12/15. This is platform depth, recipe accumulation, service density and cross-category customer lock-in rather than a consumer-style network effect.
• Narrative liquidity: 10/15. The story is simple and powerful, but the market already understands it and partially pays up for it.
• Bridge to the traditional world: 8/10. SOE control, Big Fund ownership, Stock Connect participation, listed-share consolidation and CXMT’s IPO process all strengthen the bridge.
• Counter-cyclical opportunity: 7/10. NAURA can consolidate and keep investing through downturns, but it still depends on fab capex for demand.
• Regulatory and ruin risk: -10/40. The business is sturdier than the stock multiple, but sanctions, qualification risk and cycle exposure still deserve a meaningful deduction.

That yields a total SUN-R score of 71/100, which places NAURA in High-beta watch. What pulls the score down is not business reality but the amount of good news the market already prices into the equity.

【Project Standard Rating】Watch

Position sizing and tracking

For a balanced investor, the stock is not an obvious heavy position at the current quote. The business quality is high enough for long-term research ownership, but the price is not forgiving. I would stay watch-only above the fair-buy band, and move to a medium position only if the stock re-enters a better range or new disclosures prove the newer categories are becoming material without a fresh multiple expansion.

A few disclosures would justify more work over the next three months. Evidence that etch and deposition keep outgrowing the broader equipment segment. Whether contract liabilities stabilize or rise again after the 2025–Q1 2026 decline. How much of the Kingsemi integration starts showing up in bundled customer wins. And whether CXMT’s listing proceeds into issuance, with its capacity / technology-upgrade program accelerating disclosed procurement across domestic tool vendors.

What would overturn the thesis is a pattern, not a single soft quarter: contract liabilities trend down, operating cash flow weakens, new-category R&D fails to translate into shipped systems, and the company loses the right to claim widening process coverage. At that point the stock stops being a platform premium story and starts trading like a cyclical industrial at a still-too-high multiple.

The markers I would track are straightforward. Over the next 3 months: CXMT registration-to-issuance progress, NAURA contract liabilities, delivery cadence, and any further disclosures on Kingsemi integration. Over 12 months: electronic process equipment revenue growth versus total revenue, gross-margin stability, the gap between revenue growth and R&D-led profit growth, and whether new categories stay a “promise” or move into disclosed scale. Over 3 years: whether NAURA can hold double-digit or better growth and credible upward migration in the technology mix at the same time.

My ideal buy range is CNY 500 to CNY 580. That range assumes the business stays intact and the investor is paid for cycle, sanctions and execution risk rather than pretending those risks away. I read CNY 580 to CNY 700 as fair hold territory, and prices materially above CNY 700 as leaning ever harder on new upside proof rather than existing fundamentals.

【Valuation Range】

• current: 682.22 CNY (as of 2026-06-16 previous close)
• bear (conservative · ideal buy zone): [500, 580]
• base (fair · acceptable hold zone): [580, 700]
• bull (optimistic · fully-priced zone): [700, 850]
• mode: price

Scenario analysis and final research verdict

The valuation below rests on a blended A-share framework. A pure DCF does not fit, because earnings are heavily shaped by policy, fab capex and category mix; a pure narrative multiple does not fit either, because the company already has large-scale disclosed revenue and segment margins. I anchor on a reasonable 12-month to medium-term earnings band and then apply peer-aware A-share multiples. Reuters’ 2026-06-16 quote pages show NAURA at about 48.3x forward earnings and 11.9x sales, versus AMEC at about 79.5x forward earnings and 22.8x sales, and ACM Research Shanghai at about 112.8x adjusted P/E and 21.9x sales. So NAURA still trades at a discount to the more narrowly focused high-growth domestic tool names, but the discount is a long way from “cheap.”

The optimistic case assumes CXMT and other domestic fabs keep spending, Kingsemi integration widens wallet share instead of just broadening the slide deck, and NAURA proves that ion implantation, electroplating and adjacent tools can scale into meaningful revenue without crushing returns through R&D. The neutral case assumes sustained growth but slower earnings conversion, with R&D and integration staying heavy. The pessimistic case assumes a real capex pause, a less favorable mix and a lower A-share multiple for domestic semi-equipment names. The table is my judgment, grounded in the company’s current revenue base, segment margin profile, disclosed R&D trajectory and the peer multiple backdrop.

My disagreement with the market is over how much future certainty belongs in today’s price, not over whether NAURA is good. The market largely agrees NAURA is a national champion. The open question is whether that automatically deserves a premium close to the top of the domestic-tool cohort while the company is still spending heavily to widen its category set. I sit more conservative than the momentum market. NAURA probably earns more over time than a simple cyclicals framework would allow, but the current price already reflects a large share of the first-order localization and AI-memory narrative.

My final verdict: NAURA is one of the strongest listed expressions of China’s semiconductor-equipment localization shift, and it passes the hard part of the SUN-R test because real usage clearly carries the narrative. This is substance, not narrative dressed up as substance. At the current quote, though, it is also not an old-world asset the market has simply missed and left underpriced. It is a high-quality, state-backed platform toolmaker with real shipment scale and wide product coverage, trading at a valuation that assumes much of the structural case keeps working. That leaves upside if execution keeps broadening, and little protection if memory spending, export-control friction or valuation liquidity turn against the stock.

So I keep the name in High-beta watch with a Watch rating. Over the medium-to-long term the company’s prospects stay good: breadth, validation depth, R&D intensity and policy alignment all argue for a larger domestic share of China’s tool wallet. I turn more constructive on a price reset into the fair-buy band, or on fresh evidence that new categories are becoming an earnings engine rather than mostly an engineering project. I turn more negative if contract-liability support, cash conversion and category monetization deteriorate together while the stock still commands a growth multiple.

【Fair Buy Price】500–580 CNY Basis: blended 12-month and 3-5-year A-share valuation using NAURA’s disclosed segment economics, continued heavy R&D, and a peer multiple band anchored to NAURA’s 48.3x forward P/E versus AMEC’s 79.5x and ACM Research Shanghai’s 112.8x as of 2026-06-16.

The main information gaps are worth stating plainly. NAURA does not publicly disclose backlog in the industrial-company sense, so order visibility has to be inferred from contract liabilities, customer concentration and industry commentary. Nor does it publish segment revenue or margin by each tool family beyond a few category revenue landmarks, so any fine-grained valuation by etch versus deposition versus clean stays inferential. On CXMT, the IPO process, the industrial linkage and the ecosystem overlap are all public; exclusive supply share and detailed commercial exposure are not. Treat the valuation bands above as disciplined ranges, not false precision.

This is not investment advice, only analysis under a research framework.

Other tickers mentioned

• 688012.SHG: AMEC, used as a domestic etch-equipment valuation and product-focus comparable.
• 688072.SHG: Piotech, mentioned as a domestic deposition-focused comparable within China’s tool ecosystem.
• 688082.SHG: ACM Research Shanghai, used as a domestic clean / electrochemical-equipment valuation comparable.
• 688120.SHG: Hwatsing Technology, mentioned as a domestic CMP-equipment peer in China’s broader tool stack.
• 688037.SHG: Kingsemi, discussed because NAURA acquired control to extend coverage into track and advanced packaging.
• 688981.SHG: SMIC, referenced as a representative domestic fab and advanced-node qualification benchmark in localization discussions.
• 603986.SHG: GigaDevice, mentioned because it appears among the named shareholders in CXMT’s prospectus.
• 005930.KS: Samsung Electronics, referenced in CXMT’s prospectus as a global DRAM incumbent benchmark.

This report is based on public information and does not constitute investment advice. Markets carry risk; invest with caution.