AI 광 사이클은 진짜다. 하지만 $946.90에서는 “좋은 회사”보다 “검증 대기 가격”에 가깝다.
LITE는 2023~2024년 Cloud and Networking 침체를 지나 2026년 AI 데이터센터 광부품 부족의 핵심 수혜주가 됐다. Q3 FY26 매출 $808.4M, non-GAAP GM 47.9%, Q4 매출 guide $960M~$1.01B는 방향을 확인한다. 반면 현재가 $946.90은 EV/Sales 29.66x와 forward P/E 59.78x를 요구하므로, 신규 풀비중은 가격 조정 또는 Q4 상단 초과 뒤에만 합리적이다.
판정
신규는 $820~900 또는 Q4 beat 후 재평가. 보유자는 $1,120 위에서 일부 현금화 규칙.
현재가/기준가
$946.90
2026-05-22 close / after-hours snapshot, StockAnalysis delayed snapshot
상단/위험선
$1,120 / $760
상단은 추격 제한선, $760은 thesis 재검증선
의사결정 카드
Buy가 아닌 이유: EV/Sales 29.66x는 FY27 매출 $5B~$6B와 45%+ margin 유지가 필요하다.
Hold가 가능한 이유: Cloud and Networking 매출 회복, 1.6T/200G EML, UHP laser, NVIDIA-linked capacity 자금이 thesis를 강화했다.
Watch로 바뀌는 조건: $820~900 가격 조정 또는 Q4 FY26 revenue $1.01B 초과와 GM 48.5% 이상.
Avoid로 바뀌는 조건: $760 아래 주간 종가, Q4 매출 $960M 미만, GM 45% 하회, capex intensity 22% 초과.
LITE의 사업 방향성은 강하다. 투자 판단의 핵심은 'AI 광통신이 맞는가'가 아니라 '현재 가격이 다음 4~6개 분기 성공을 얼마나 선반영했는가'다.
2. 현재 주가 위치 대시보드
기준가 $946.90은 2026-05-22 종가이고, after-hours snapshot은 $935.20이다. 52주 고점은 StockAnalysis/MarketBeat 기준 $1,085.68, 50일선은 $846.87, 200일선은 $442.62다. 가격은 50일선보다 약 11.8% 위, 200일선보다 약 113.9% 위에 있어 장기 추세는 강하지만 신규 진입 안전마진은 좁다.
52주 고점 대비
87.2%
고점 근처 재평가 구간
50일선 대비
+11.8%
단기 과열 완화 필요
200일선 대비
+113.9%
장기 rerating 반영
Short interest
12.12%
squeeze와 reversal 모두 가능
[현재가, 이동평균, 행동선] 현재가가 50일선·200일선·위험선과 어디에 있는지 보여준다. $900 위에서는 신규 안전마진이 좁고, $760 아래 주간 종가는 thesis 재검증 신호다.
source-backed input confidence: high
StockAnalysis 2026-05-22 close 기반
[Base target에서 bear target으로 내려가는 리스크 waterfall] Base 목표가가 어떤 사건에서 Bear 목표가로 낮아지는지 보여준다. GM 45% 하회나 Q4 매출 $960M 미만이면 valuation을 먼저 낮춘다.
model/proxy input confidence: medium
모델 차트. 수치가 공식 목표가가 아니라 리스크 감도다.
*AI의 견해입니다
3. 타임라인 - M&A 누적에서 AI 광 scarcity rerating으로
LITE의 10년은 단순한 성장 스토리가 아니라 M&A, customer cycle, 광부품 세대 전환, balance-sheet 부담이 겹친 사이클이다. 2018 Oclaro, 2022 NeoPhotonics, 2023 Cloud Light는 현재의 AI optical 포지션을 만들었지만, 2023~2024 downturn에서는 부채와 통합비용도 같이 드러났다.
[10년 주가/이벤트 압축] M&A와 AI 광통신 재평가가 주가 레짐을 바꾼 순서를 보여준다. 2026년 구간은 실적보다 기대가 빠르게 오른 구간으로 읽어야 한다.
model/proxy input confidence: medium
가격은 주요 국면을 보여주는 근사 series다. 실제 daily OHLC가 아니다.
*AI의 견해입니다
[시기] 이 표는 시기 항목을 기준으로 주요 수치, 판단 기준, 주가/투자 의미 정보를 비교한다. 가격선, Green/Red, 매출·GM·FCF 임계값이 있는 행을 먼저 읽고 나머지는 근거 확인용으로 쓴다.
| 시기 | 이벤트 | 사업 의미 | 주가/투자 의미 |
|---|---|---|---|
| 2018 | Oclaro 인수 | 고속 coherent/transport 역량 강화 | M&A 시너지는 긍정, 이후 부채/통합 부담도 축적 |
| 2021 | Coherent 인수 시도 실패 | 산업 레이저/포토닉스 scale-up 시도 | 승자는 COHR-II/VI 통합, LITE는 다른 경로 선택 |
| 2022 | NeoPhotonics 인수 | coherent/ROADM/transport portfolio 확대 | CN 침체 전 비용기반 확대 |
| 2023-10 | Cloud Light 인수 발표 | $750M, 400G+ transceiver 직접 노출 | AI data-center module 시장 진입 |
| 2024 | CN 침체/재고 조정 | 매출과 margin 동반 압박 | 2025 회복 전 valuation discount |
| 2025 | CN 매출 회복 | FY25 CN +30%, segment profit +112% | AI/ML cloud customer 매출화 시작 |
| 2026-03 | NVIDIA strategic investment/Greensboro facility | 자금+capacity+customer validation | 유동성 우려 완화, share/dilution 체크 필요 |
| 2026-05 | Q3 FY26 record quarter | 매출 $808.4M, non-GAAP GM 47.9% | valuation이 2027 성공까지 선반영 |
과거 사이클에서 현재로 가져올 규칙
[과거 구간] 이 표는 과거 구간 항목을 기준으로 주요 수치, 판단 기준, LITE 보고서 반영 정보를 비교한다. 가격선, Green/Red, 매출·GM·FCF 임계값이 있는 행을 먼저 읽고 나머지는 근거 확인용으로 쓴다.
| 과거 구간 | 좋았던 점 | 놓치기 쉬운 리스크 | 현재에 적용할 규칙 | LITE 보고서 반영 |
|---|---|---|---|---|
| 2017~2018 iPhone/3D sensing enthusiasm | VCSEL/Face ID content expansion 기대 | customer concentration and seasonality | single customer cycle should not receive full scarcity multiple | IT segment cannot carry whole valuation |
| 2018 Oclaro integration | transport/coherent scale | integration and debt complexity | M&A expands TAM but raises fixed-cost base | Cloud Light synergy must show margin |
| 2021 Coherent attempt | scale ambition | failed bid avoided larger leverage but lost asset | LITE still pursued portfolio expansion through other deals | COHR is necessary peer |
| 2022 NeoPhotonics | coherent/ROADM product depth | telecom downturn followed | right asset can still meet wrong cycle | transport recovery must be verified |
| 2023 Cloud Light | datacom transceiver entry | module competition and ASP pressure | TAM expansion is not margin expansion by default | track module GM proxy |
| 2024 CN trough | inventory correction and low utilization | negative FCF and valuation discount | trough can hide high operating leverage | TTM rebound validates but does not price every future year |
| 2025 AI/ML cloud recovery | CN +30% and segment profit +112% | IT remains smaller and weaker | core engine shifted to CN | Apple cycle is secondary |
| 2026 NVIDIA/1.6T rerating | capital validation and scarcity premium | share/dilution, capex, multiple risk | scarcity assets can overshoot | buy zones must be price-sensitive |
4. 사업 구조 / 산업 사이클 분석
FY2025 기준 Cloud and Networking은 매출 $1.4108B, 전체 85.8%다. Industrial Tech는 $234.2M, 14.2%다. 그래서 현재 LITE는 산업 레이저/3D sensing 회사라기보다 AI 데이터센터 광부품 공급망 주식으로 평가된다. Apple/3D sensing은 하방 완충재와 옵션이지만, 주가의 12개월 승패는 Cloud and Networking 매출과 margin이 결정한다.
사업부를 숫자로 읽는 법
[표 해석] 사업부별로 무엇을 팔고, 왜 주가에 중요하며, 어느 KPI로 돈이 실제로 찍히는지 보여준다. Cloud and Networking이 85%대 mix와 high-40s GM을 유지하면 Bull, Industrial Tech/3D Sensing만으로는 현재 valuation을 방어하기 어렵다.
| 사업 | 무엇 | 왜 중요 | 어디서 돈 | 무엇 보면 됨 |
|---|---|---|---|---|
| Cloud and Networking | 클라우드 데이터센터와 통신망에 들어가는 광칩·레이저·광트랜시버(optical transceiver)·ROADM/EDFA 모듈을 판다. | FY25 매출 $1.4108B, 전체 85.8%; FY25 segment profit $264.5M. 현재 주가의 대부분을 설명하는 사업부다. | hyperscaler(대형 클라우드 사업자), AI/ML 인프라 고객, 네트워크 장비사가 고객이다. 매출은 제품 출하·qualification·장기 capacity allocation에 따라 인식된다. | Q4 FY26 매출 가이드 $960M~$1.01B, non-GAAP OPM 35~36%; CN 추정 매출 $880M 이상이면 Green, $780M 이하면 Red. |
| Industrial Tech | 산업용 fiber laser, ultrafast/UV laser, diode/gas laser와 sensing용 레이저를 묶은 사업부다. | FY25 매출 $234.2M, 전체 14.2%; segment profit $12.1M. 현재 valuation의 본체는 아니지만 하방 완충재다. | 정밀가공·반도체/마이크로일렉트로닉스·EV/배터리 제조 장비 고객에게 제품을 판다. | IT 매출이 2분기 연속 YoY 성장하면 옵션가치 회복, 2분기 연속 감소하면 CN-only multiple로 낮춘다. |
| 3D Sensing / LiDAR | VCSEL(수직공진 표면방출 레이저)과 laser light source가 모바일 3D sensing 카메라, AR/VR, 산업·자동차 LiDAR에 들어가는 영역이다. | 공식 별도 매출은 미공시이며 FY25 Industrial Tech $234.2M 안에 포함된다. Apple/iPhone LiDAR 매출 비중은 이번 공식 자료에서 확인되지 않았다. | 스마트폰·AR/VR·산업/자동차 고객의 설계 채택과 부품 단가로 돈을 번다. | 공식 customer name 또는 IT 매출 YoY 회복 없이는 Bull target에 크게 반영하지 않는다. Apple 관련 숫자는 user/inference로만 표시한다. |
| Cloud Light / Systems | Cloud Light 인수로 datacenter module·transceiver 매출에 직접 노출되는 경로다. | 인수 발표 당시 Cloud Light LTM 매출은 $200M 초과였고 400G+ 비중이 90% 초과였다. Q3 FY26 product-type Systems 매출은 $275.1M이지만 Cloud Light 단독 매출은 아니다. | 고속 광모듈을 cloud/datacenter 고객에게 판다. 부품보다 TAM은 커지지만 module ASP와 yield 리스크도 직접 맞는다. | Systems 매출 성장과 GM 48%대 유지가 같이 나오면 긍정. Systems 성장에도 GM <45%면 dilution risk다. |
[CN/IT 매출 믹스: FY2023~FY2027E] Cloud and Networking이 회사의 중심으로 커진 변화를 보여준다. CN이 85% 아래로 내려가면서 GM이 같이 낮아지면 AI 광통신 thesis가 약해진다.
model/proxy input confidence: medium
FY2023~FY2025는 10-K/회사 발표, FY2026E~FY2027E는 모델
*AI의 견해입니다
[분기별 CN/IT 모델] FY26 분기 매출을 CN/IT로 배분한 추정 경로다. Q4 FY26에서 CN 추정 $880M 이상이면 Bull, $780M 이하면 Red 신호다.
model/proxy input confidence: medium
Q4 FY25 CN $424.1M은 회사 발표 기반, 나머지는 total revenue와 mix를 연결한 model
*AI의 견해입니다
[CN 매출 회복 곡선과 margin score] CN 매출 회복과 마진 품질이 같이 움직이는지 보여준다. 매출만 오르고 margin score가 떨어지면 Cloud Light/module mix dilution을 의심한다.
mixed input input confidence: medium
AI optical cycle이 CN을 다시 성장 engine으로 바꾸는지 확인
[LITE 투자 thesis 6축 score] LITE 투자 thesis의 강점과 약점을 한 화면에 압축한다. 점수 70 미만 축이 3개 이상이면 비중 확대보다 검증 대기가 맞다.
model/proxy input confidence: medium
점수는 정량 공시와 모델 판단을 합친 decision aid
*AI의 견해입니다
Disclosure bridge: FY25 segment, FY26 product type, modeled quarter mix
critic round 1의 핵심 오류 가능성은 CN/IT segment와 Components/Systems product type을 같은 분모로 섞는 데 있었다. 아래 표에서는 공식 segment, 공식 product type, 모델 quarter mix를 분리한다.
[구분] 이 표는 구분 항목을 기준으로 주요 수치, 판단 기준, 해석 한계 정보를 비교한다. 가격선, Green/Red, 매출·GM·FCF 임계값이 있는 행을 먼저 읽고 나머지는 근거 확인용으로 쓴다.
| 구분 | 수치 | 근거/산식 | 해석 한계 |
|---|---|---|---|
| FY2025 official segment | Cloud & Networking $1,410.8M / Industrial Tech $234.2M | SEC FY2025 10-K segment revenue table | 공식 reportable segment. CN 85.8%, IT 14.2%. |
| FY2026 official product type | Q3 FY26 Components $533.3M / Systems $275.1M / total $808.4M | Q3 FY26 press release product-type disclosure | product type이지 CN/IT segment가 아니다. Systems에는 Cloud Light module exposure가 섞여 있다고 보는 보조 신호. |
| Modeled quarterly mix | Q3 FY26 CN $740M / IT $68.4M, Q4 guide CN $910M / IT $75M | model: total revenue x modeled CN mix | segment 공시가 아니므로 CN error band ±$30~45M, IT error band ±$10~20M 적용. |
| Formula | Modeled CN = quarter total x CN mix assumption; Modeled IT = total - modeled CN | mix assumption anchored to FY25 official CN 85.8%, Q4 FY25 CN $424.1M, AI optical product-type ramp | 공식 segment와 product-type을 섞어 단정하지 않고 valuation sensitivity용으로만 사용. |
Apple / iPhone / Vision Pro optionality
Apple iPhone LiDAR 또는 3D Sensing 매출 비중은 공식 자료에서 확인되지 않았다. 아래 표와 차트는 FY25 Industrial Tech 매출 안에 포함된 optionality를 따로 떼어 본 추정 모델이며, Cloud and Networking thesis와 분리해서 읽어야 한다.
*AI의 견해입니다[3D sensing TAM model: iPhone / AR·VR / industrial / auto option] 3D sensing의 iPhone/AR/산업/자동차 옵션 크기를 보여준다. 공식 Apple 매출표가 아니므로 IT 매출이 2분기 연속 YoY 성장할 때만 target에 반영한다.
model/proxy input confidence: medium
모델 차트. source: FY25 IT revenue, user-provided Apple thesis, public 3D sensing TAM ranges; capture-rate와 target impact는 valuation input
*AI의 견해입니다
[시나리오] 이 표는 시나리오 항목을 기준으로 주요 수치, 판단 기준, Target impact 정보를 비교한다. 가격선, Green/Red, 매출·GM·FCF 임계값이 있는 행을 먼저 읽고 나머지는 근거 확인용으로 쓴다.
| 시나리오 | IT revenue input | TAM input | LITE capture-rate input | Target impact |
|---|---|---|---|---|
| Base iPhone sensing | FY25 IT $234.2M, FY26E IT $260~300M | iPhone sensing TAM $3.8B 2025E | LITE capture 5~7% model | $0~30 target impact; CN thesis와 분리 |
| iPhone content/share upside | IT +$80~150M annualized | iPhone/Face ID/LiDAR content regain | capture +1~2pp model | +$20~50; 단 2Q 연속 IT YoY 성장 필요 |
| Vision Pro / AR option | near-term revenue $0~30M until volume launch | AR/VR sensing TAM $0.5B 2025E -> $2.2B 2028E model | capture 3~6% model | +$0~20 until volume; headline만으로 매수 근거 아님 |
| Share/price-down downside | IT -$50~120M | VCSEL ASP pressure 또는 Apple share normalization | capture -1~3pp model | -$20~60; top customer/IT YoY 확인 전까지 inference |
[사이클] 이 표는 사이클 항목을 기준으로 주요 수치, 판단 기준, 검증해야 할 숫자/사건 정보를 비교한다. 가격선, Green/Red, 매출·GM·FCF 임계값이 있는 행을 먼저 읽고 나머지는 근거 확인용으로 쓴다.
| 사이클 | 시기 | 무슨 변화인가 | LITE에 주는 것 | 검증해야 할 숫자/사건 |
|---|---|---|---|---|
| 800G EML | 2025~2026 | AI cluster leaf-spine, DR4/FR4 | CN 매출 step-up | gross margin 45%+와 Q4 매출 $1B 근접 |
| 1.6T EML | 2026~2027 | 200G/lane PAM4, 8x200G | scarcity premium | volume shipment, not only demo |
| UHP/CW laser | 2026~2028 | silicon photonics external light source | NVIDIA/hyperscaler strategic supply | Greensboro ramp schedule and capex |
| ROADM/EDFA | 2025~2027 | DCI/metro transport upgrade | legacy telecom recovery plus AI DCI | transport order recovery without margin erosion |
| Cloud Light transceiver | 2024~2027 | 400G/800G/1.6T module entry | served opportunity expansion | module GM does not dilute component GM |
| Apple 3D sensing | iPhone annual cycle | Face ID/LiDAR/VCSEL | IT revenue floor | new model content or share normalization |
| AR/VR 3D sensing | 2026~2028 option | Vision Pro follow-on, display sensing | optional TAM | volume device launch and VCSEL content |
| Automotive LiDAR | long option | vendor LiDAR vs Tesla self-approach | low near-term weight | design win with production volume |
| Industrial laser | 2026 recovery candidate | EV/metal/semicap microprocessing | IT segment recovery | 2Q consecutive YoY growth |
| ASP curve | 세대 전환마다 반복 | launch premium -> volume discount | revenue/margin conflict | unit growth faster than ASP erosion |
| Optical vs copper | distance/bandwidth threshold | DAC/AEC/LRO/CPO/pluggable | architecture mix risk | optical content per GPU cluster |
| M&A consolidation | 2026~2028 | COHR-II/VI, LITE-Oclaro/NeoPhotonics/Cloud Light | scale and customer access | debt cost and integration execution |
CN 매출이 빨라질수록 LITE는 COHR보다 높은 optical beta를 받을 수 있다. 하지만 Cloud Light의 module exposure가 커질수록 AAOI/Innolight/Eoptolink식 ASP pressure도 더 직접적으로 맞는다.
5. 핵심 이벤트 의미 - Cloud Light, 1.6T, NVIDIA-linked capacity
Cloud Light는 LITE를 부품 공급사에서 광트랜시버(optical transceiver)·module 매출까지 확장시켰다. 회사 발표에 따르면 Cloud Light의 지난 12개월 매출은 $200M 초과였고, 400G+ 제품이 대부분이었다. 2026 OFC에서 LITE는 1.6T DR4 OSFP prototype, 400G differential EML, UHP laser, DWDM laser를 보여줬고, Q3 FY26 발표자료는 1.6T transceiver가 Q4 FY26 ramp on track이라고 밝혔다. 다만 NVIDIA Spectrum-X/Quantum-X와 H100/H200/B200별 매출액은 공식 미공시다.
[FY25 revenue에서 FY27E revenue로 가는 bridge] FY25 매출에서 FY27 매출 모델로 넘어가는 가정별 기여도를 보여준다. 1.6T와 UHP laser가 빠지면 Bull 경로가 성립하지 않는다.
model/proxy input confidence: medium
800G/1.6T/UHP/IT recovery는 모델 입력
*AI의 견해입니다
[광 vs 동 인터커넥트 임계값] 거리와 대역폭이 커질수록 광 연결이 구리 연결을 대체하는 구간을 보여준다. AI 클러스터가 커질수록 LITE의 optical content가 늘어나는 구조다.
mixed input input confidence: medium
거리와 대역폭이 커질수록 optical content가 늘어나는 구조를 단순화
[이벤트] 이 표는 이벤트 항목을 기준으로 주요 수치, 판단 기준, 반대 확인점 정보를 비교한다. 가격선, Green/Red, 매출·GM·FCF 임계값이 있는 행을 먼저 읽고 나머지는 근거 확인용으로 쓴다.
| 이벤트 | 검증된/공개 수치 | 긍정 메커니즘 | 반대 확인점 |
|---|---|---|---|
| Cloud Light | $750M transaction value | cloud intra-data-center opportunity를 크게 확장 | module margin과 customer concentration을 같이 추적 |
| 1.6T 2xDR4 OSFP | OFC 2026 live demonstration | 200G/lane EML과 UHP laser 양쪽 architecture 노출 | demo가 아닌 volume shipment 문구 필요 |
| Greensboro InP facility | 240,000 sq ft, mid-2028 ramp | U.S. InP laser capacity option | 2026~2027 실적에는 capex/qualification lag |
| Q3 FY26 result | revenue $808.4M, GM 47.9% | AI optical이 숫자로 전환 | Q4 guide 상단 초과 여부 |
| NVIDIA strategic context | NVIDIA $2B investment, multibillion purchase commitment, future capacity access rights | balance-sheet fear 완화와 customer validation | diluted shares, customer dependence, exact product revenue 미공시를 동시에 확인 |
Cloud Light bridge: acquisition claim vs actual run-rate proxy
[Bridge 항목] 이 표는 Bridge 항목 항목을 기준으로 주요 수치, 판단 기준, 투자 해석 정보를 비교한다. 가격선, Green/Red, 매출·GM·FCF 임계값이 있는 행을 먼저 읽고 나머지는 근거 확인용으로 쓴다.
| Bridge 항목 | 수치/주장 | 근거 | 투자 해석 |
|---|---|---|---|
| Pre-deal acquired revenue | Cloud Light LTM revenue >$200M; >90% from 400G+ and latest-quarter >50% from 800G | Lumentum 2023-10-30 acquisition announcement | A등급 공식 취득 전 run-rate. |
| Deal claim | transaction value about $750M; expected to more than double cloud data-center infrastructure revenue in 12 months after close | same announcement | forward-looking claim. 실제 결과와 별도 추적. |
| Actual run-rate proxy | FY25 CN $1.4108B; Q3 FY26 product-type Systems $275.1M, annualized $1.10B proxy | SEC FY25 segment + Q3 FY26 product type | Cloud Light standalone 매출은 미공시. Systems proxy는 Cloud Light만의 수치가 아니므로 confidence medium. |
| GM effect | Q3 FY26 non-GAAP GM 47.9% | Q3 FY26 press release | Cloud Light module mix가 늘어도 GM이 유지되면 accretive. GM <45%면 module ASP/yield dilution으로 재해석. |
| EPS / FCF effect | Q3 FY26 EPS guide/Q4 guide strong, but FY25 FCF negative and TTM FCF $114M proxy | StockAnalysis/financial cross-check | EPS accretion은 회복됐지만 FCF conversion은 capex/working capital 때문에 아직 검증 중. |
LITE vs COHR: 1.6T evidence 비교
[비교축] 이 표는 비교축 항목을 기준으로 주요 수치, 판단 기준, 판정 규칙 정보를 비교한다. 가격선, Green/Red, 매출·GM·FCF 임계값이 있는 행을 먼저 읽고 나머지는 근거 확인용으로 쓴다.
| 비교축 | LITE | COHR | 판정 규칙 |
|---|---|---|---|
| Capacity | LITE: Greensboro 240k sq ft InP/CW/UHP laser facility, mid-2028 ramp | COHR: 6-inch InP platform, internal InP output, NVIDIA strategic partnership per local COHR report | LITE는 laser scarcity beta, COHR은 broader vertical integration. |
| Shipment wording | LITE: OFC 2026 1.6T 2xDR4 OSFP demo and Q4 FY26 revenue guide; volume shipment wording still required | COHR: 1.6T full-suite/capability language and D&C revenue ramp; exact 1.6T share not verified | 두 회사 모두 demo/capability와 revenue contribution을 분리해야 한다. |
| Product architecture | LITE: 200G EML, UHP/CW laser, external light-source/SiPh optionality, Cloud Light module path | COHR: InP EML/CW laser, SiPh/CPO/NPO, component-to-module portfolio | CPO/SiPh가 커질수록 laser content와 ASIC/SiPh profit-pool 배분을 같이 봐야 한다. |
| Customer qualification | LITE: NVIDIA-linked capacity narrative, exact customer/order identity는 제한적 | COHR: NVIDIA partnership is named in local COHR source set; purchase commitment/strategic context stronger | LITE bull은 multi-customer qualification 또는 volume shipment 문구가 나와야 한다. |
| Valuation read-through | LITE GM 48.5%+와 CN $880M+면 premium 유지 | COHR D&C GM/ASP pressure가 나오면 sector multiple 동반 하향 | 1.6T 비교는 매출 성장률보다 GM과 customer qualification이 판정권. |
Peer read-through: LITE 단독 실적 전후에 같이 볼 이름
[Peer] 이 표는 Peer 항목을 기준으로 주요 수치, 판단 기준, 보고서 내 반영 정보를 비교한다. 가격선, Green/Red, 매출·GM·FCF 임계값이 있는 행을 먼저 읽고 나머지는 근거 확인용으로 쓴다.
| Peer | 확인할 것 | LITE bull read-through | LITE bear read-through | 보고서 내 반영 |
|---|---|---|---|---|
| COHR | Q3/Q4 optical revenue, InP capacity, datacom order book | COHR도 GM 상승이면 industry-wide scarcity | COHR가 ASP/yield pressure 언급하면 LITE multiple도 낮춤 | LITE premium 유지 여부 |
| AAOI | 1.6T order conversion, 800G shipment, customer concentration | AAOI revenue가 guide 상단이면 transceiver demand 검증 | AAOI GM이 약하면 module ASP 압박 신호 | Cloud Light margin risk |
| MRVL | data-center silicon growth, custom silicon, DSP attach | DSP/ASIC demand가 optical link growth와 동행 | MRVL guide 하향이면 hyperscaler network capex 의심 | optical demand lead/lag |
| CRDO | AEC/LRO adoption, linear optical narrative | linear optical이 optical content를 넓힐 수 있음 | LRO가 pluggable laser profit pool을 압박할 수 있음 | architecture risk |
| SMTC | Signal Integrity growth, FiberEdge/CopperEdge | AI interconnect breadth 확인 | SI growth 둔화면 adjacent demand 약화 | interconnect basket risk |
| IPGP | industrial laser orders | IT/industrial cycle 회복 신호 | industrial laser weakness 지속 | LITE IT optionality |
| Innolight | 800G/1.6T China vendor revenue | TAM expansion confirmation | price competition acceleration | ASP curve 조정 |
| Eoptolink | high-speed transceiver shipment | cloud demand breadth | A-share premium-funded capacity build | 2027 supply pressure |
| Semtech/Marvell/Broadcom | DSP/retimer/switch silicon | optical link is part of larger AI networking basket | silicon vendors capture more profit pool | component margin ceiling |
6. 원가·공급망 충격 - InP, EML, ROADM/EDFA, ASP 곡선
AI 광통신에서 원가 리스크는 단순 원재료보다 capacity, yield, packaging, qualification, customer second-source에 가깝다. launch 시점 ASP는 높지만 volume이 붙으면 고객은 dual-source와 price-down을 요구한다. LITE의 margin은 단기에는 scarcity가 방어하지만, 중기에는 ASP 하락보다 unit/throughput이 더 빨리 늘어야 한다.
[광 부품 ASP curve: 400G / 800G / 1.6T model] 400G, 800G, 1.6T의 출시 프리미엄과 가격 하락을 단순화한 모델이다. ASP 하락보다 unit과 yield 개선이 빨라야 GM 48%대가 유지된다.
model/proxy input confidence: medium
공식 ASP가 아니라 public market range 기반 모델. 추정 차트로 valuation math에는 직접 넣지 않음
*AI의 견해입니다
[Hyperscaler AI capex proxy와 LITE CN revenue] 대형 클라우드 사업자의 AI 설비투자와 LITE CN 매출이 같은 방향으로 움직이는지 보는 보조 차트다. capex proxy가 꺾이면 LITE multiple도 먼저 줄인다.
model/proxy input confidence: medium
Hyperscaler capex는 public/analyst compiled estimate. 방향성 proxy
*AI의 견해입니다
[원가/공급망 축] 이 표는 원가/공급망 축 항목을 기준으로 주요 수치, 판단 기준, 리스크 정보를 비교한다. 가격선, Green/Red, 매출·GM·FCF 임계값이 있는 행을 먼저 읽고 나머지는 근거 확인용으로 쓴다.
| 원가/공급망 축 | 왜 중요한가 | LITE가 얻는 것 | 리스크 |
|---|---|---|---|
| InP wafer/fab | CW/UHP/EML laser 핵심 기반 | Greensboro 6-inch InP ramp가 capacity option | mid-2028 전까지는 기존 capacity와 outsourcing에 의존 |
| EML laser | 800G/1.6T transceiver 핵심 lane | scarcity premium과 GM 방어 | AAOI/COHR/중국 vendor의 second-source 경쟁 |
| ROADM/EDFA | transport/DCI upgrade | AI data-center 간 연결과 telecom 회복 동시 노출 | legacy telecom capex 둔화 |
| Cloud Light assembly/test | module revenue 직접화 | TAM 확대와 고객 접점 강화 | module ASP downcycle과 yield ramp |
| 3D sensing VCSEL | iPhone/AR/VR/industrial sensing | IT revenue floor | Apple cycle, content share, unit seasonality |
7. 재무 펀더멘털 - 10년 매출, margin, FCF, 부채
10년 재무는 LITE가 안정 복리주라기보다 사이클/M&A 주식이라는 점을 보여준다. FY2021에는 FCF $653.9M까지 갔지만 FY2024~FY2025는 FCF가 음수였다. TTM은 매출 $2.49B, 영업이익 $254.4M, FCF $114M로 회복 중이다. 현재 valuation은 과거 평균이 아니라 FY27~FY28 FCF 재가속을 요구한다.
[10년 매출·영업이익·FCF] 10년 매출·영업이익·FCF가 LITE를 안정 복리주가 아니라 사이클/M&A 주식으로 보여준다. FCF가 매출 성장을 따라오지 못하면 EV/Sales premium은 줄어든다.
model/proxy input confidence: medium
FY2016~FY2020 일부 FCF/OP는 proxy. FY2021~TTM은 StockAnalysis/SEC 기반
*AI의 견해입니다
[Cash / Debt / FCF 추이] 현금·부채·FCF의 균형을 보여준다. 현금은 늘었지만 Q4 가이드 희석 주식수 102M이므로 per-share upside를 별도 할인해야 한다.
mixed input input confidence: medium
NVIDIA-linked capital 이후 liquidity squeeze는 낮아졌으나 dilution/capex는 남음
[연도] 이 표는 연도 항목을 기준으로 주요 수치, 판단 기준, 근거 정보를 비교한다. 가격선, Green/Red, 매출·GM·FCF 임계값이 있는 행을 먼저 읽고 나머지는 근거 확인용으로 쓴다.
| 연도 | 매출 | 영업이익 | OPM | FCF | 근거 |
|---|---|---|---|---|---|
| FY2016 | $903.0M | $-23.4M | -2.6% | $35.0M | StockAnalysis revenue + older cash-flow proxy |
| FY2017 | $1,001.6M | $11.5M | 1.1% | $45.0M | StockAnalysis revenue + older cash-flow proxy |
| FY2018 | $1,247.7M | $47.6M | 3.8% | $160.0M | StockAnalysis revenue + older cash-flow proxy |
| FY2019 | $1,565.3M | $139.9M | 8.9% | $230.0M | StockAnalysis revenue + older cash-flow proxy |
| FY2020 | $1,678.6M | $-62.7M | -3.7% | $390.0M | StockAnalysis revenue + older cash-flow proxy |
| FY2021 | $1,742.8M | $527.0M | 30.2% | $653.9M | SEC/StockAnalysis |
| FY2022 | $1,712.6M | $303.3M | 17.7% | $368.1M | SEC/StockAnalysis |
| FY2023 | $1,767.0M | $-115.7M | -6.5% | $51.3M | SEC/StockAnalysis |
| FY2024 | $1,359.2M | $-434.0M | -31.9% | $-108.3M | SEC/StockAnalysis |
| FY2025 | $1,645.0M | $-180.1M | -10.9% | $-104.7M | SEC/StockAnalysis |
| TTM | $2,488.0M | $237.1M | 9.5% | $114.0M | SEC/StockAnalysis |
공식 IR 기준 FY25/FY26 분기 매출·OP
[표 해석] FY25/FY26 분기 매출과 영업이익이 적자에서 고마진 흑자로 바뀐 속도를 보여준다. Q4 FY26 가이드의 non-GAAP OPM 35~36%를 달성하면 Bull, Q4 매출 $960M 미만이면 Red다.
| 분기 | 매출 | GAAP OP | Non-GAAP OP | Non-GAAP OPM | 공식 근거/한계 |
|---|---|---|---|---|---|
| Q1 FY25 | $336.9M | ($82.4M) | $10.0M | 3.0% | Lumentum Q2 FY25 release: Q1 comparison/reconciliation |
| Q2 FY25 | $402.2M | ($51.6M) | $31.7M | 7.9% | Lumentum Q2 FY25 release |
| Q3 FY25 | $425.2M | ($37.7M) | $46.1M | 10.8% | Lumentum Q3 FY25 release |
| Q4 FY25 | $480.7M | ($8.4M) | $72.3M | 15.0% | Lumentum Q4/FY25 release |
| Q1 FY26 | $533.8M | $6.7M | $99.8M | 18.7% | Lumentum Q1 FY26 release |
| Q2 FY26 | $665.5M | $64.3M | $167.7M | 25.2% | Lumentum Q2 FY26 release |
| Q3 FY26 | $808.4M | $174.5M | $260.7M | 32.2% | Lumentum Q3 FY26 release/presentation |
| Q4 FY26 guide | $960~1,010M | 미제공 | $336~364M implied | 35.0~36.0% | Official non-GAAP guide; GAAP reconciliation not provided |
P0 사실 검증 결과
[표 해석] P0 사실검증 표다. 확인된 공식 수치와 미공시·추정 항목을 분리해, 보고서가 Apple·NVIDIA·800G 매출을 단정하지 않게 만든다.
| P0 항목 | 판정 | 검증된 사실 | 투자 적용/한계 |
|---|---|---|---|
| FY25/FY26 분기 매출·OP | 확인 | FY25 Q1~Q4 및 FY26 Q1~Q3는 Lumentum 공식 IR의 revenue/GAAP OP/non-GAAP OP로 정리. Q4 FY26은 non-GAAP OPM guide만 있음. | GAAP Q4 FY26 guide reconciliation은 회사가 제공하지 않음. |
| AI 광트랜시버 1.6T ramp | 확인+범위 제한 | Q3 FY26 presentation: 1.6T transceivers on track for Q4 FY26 ramp. OFC 2026: 1.6T DR4 OSFP prototype uses four 400G differential EML lasers. | NVIDIA Spectrum-X/InfiniBand 직접 매출액이나 shipment volume은 미공시. |
| Cloud Light/IPG Photonics 통합 | 부분 확인 | Cloud Light: FY25 10-K와 Q3 FY26에서 escrow/warranty/integration 비용 및 Systems proxy 확인. IPG Photonics telecom line은 2022 acquisition이며 FY25에는 coherent DSP/RFIC in-house 개발 중단 언급. | IPG를 2026 신규 인수처럼 쓰면 오류. standalone contribution 미공시. |
| Apple iPhone LiDAR/3D Sensing 비중 | 미공시 | 10-K는 3D sensing cameras primarily used in mobile devices라고 설명하지만 Apple을 명명하지 않음. FY25 IT 전체는 $234.2M. | Apple/LiDAR 매출 비중은 공식 수치 없음. user/inference로만 사용. |
| 800G EML 매출/NVIDIA H100-H200-B200 연결 | 미공시 | 공식 자료는 EML이 400G/800G/upcoming 1.6T transceiver에 쓰인다고 설명. Cloud Light 인수 전 매출은 400G+ skew. | NVIDIA H100/H200/B200별 800G EML 매출은 확인 불가. |
| 신용/배당/자사주 | 확인 | Q3 FY26 cash/STI $3.172B, total liabilities $4.054B. StockAnalysis: debt $3.31B, no dividend, buyback yield -21.84%, shares +21.84% YoY. | 자사주 매입보다 희석·capacity funding이 핵심. |
[항목] 이 표는 항목 항목을 기준으로 주요 수치, 판단 기준, 투자 의미 정보를 비교한다. 가격선, Green/Red, 매출·GM·FCF 임계값이 있는 행을 먼저 읽고 나머지는 근거 확인용으로 쓴다.
| 항목 | 수치 | 맥락 | 투자 의미 |
|---|---|---|---|
| Q3 FY26 revenue | $808.4M | YoY +90% 구간 | AI optical revenue가 숫자로 전환 |
| Q3 FY26 non-GAAP GM | 47.9% | Q2 42.5%, 전년 35.2%에서 개선 | scarcity와 mix가 margin에 반영 |
| Q3 FY26 GAAP net income | $144.2M | 전년 손실에서 흑자 전환 | operating leverage 확인 |
| Q3 FY26 adjusted EBITDA | $293.5M | 매출의 36.3% | debt/FCF bridge 개선 |
| Cash/ST investments | $3.17B | debt $3.31B와 거의 균형 | refinance fear 축소 |
8. 밸류에이션 - 주가는 이미 FY27 성공을 요구한다
현재 LITE의 trailing multiple은 일반 산업주 평가틀로는 설명이 어렵다. EV/Sales 29.66x, forward P/E 59.78x, P/FCF 646.22x는 현재 이익보다 미래 optical capacity scarcity를 사는 가격이다. 따라서 valuation은 DCF 한 점보다 reverse DCF와 FY27 revenue multiple로 다뤄야 한다.
아래 valuation sensitivity, DCF bridge, Bear/Base/Bull 목표가는 공식 목표주가가 아니라 현재 가격이 요구하는 FY27~FY30 매출·FCF·multiple 조건을 역산한 모델이다. 공식 fact는 가격·EV·재무 수치이고, target range는 Codex 해석이다.
*AI의 견해입니다[10년 EV/Sales regime band] EV/Sales가 과거 정상 밴드를 얼마나 벗어났는지 보여준다. 20x 이상은 scarcity premium 구간이므로 Q4 수치가 조금만 빗나가도 하락 민감도가 크다.
model/proxy input confidence: medium
정밀 paid multiple history가 아니라 public valuation regime model
*AI의 견해입니다
[DCF input bridge] DCF 입력값이 현재 가격을 얼마나 설명하는지 역산한다. Base DCF가 현재가보다 낮으면 신규 매수는 가격 조정이나 실적 상향이 필요하다.
mixed input input confidence: medium
Base DCF는 current price보다 낮다. 보고서는 reverse DCF 검산 장치로 사용
[Bear/Base/Bull 주가 경로] Bear/Base/Bull 주가 경로를 같은 출발점에서 비교한다. Base가 현재가와 비슷하면 기대수익보다 손실 제한선이 먼저다.
model/proxy input confidence: medium
12개월 target은 시나리오별 valuation multiple model
*AI의 견해입니다
DCF assumptions table
[DCF/valuation input] 이 표는 DCF/valuation input 항목을 기준으로 주요 수치, 판단 기준, 민감도/한계 정보를 비교한다. 가격선, Green/Red, 매출·GM·FCF 임계값이 있는 행을 먼저 읽고 나머지는 근거 확인용으로 쓴다.
| DCF/valuation input | Base assumption | 왜 쓰는가 | 민감도/한계 |
|---|---|---|---|
| WACC | 10.5% base; bear 11.5%, bull 9.5% | high-multiple optical cycle and customer concentration risk | WACC +100bp는 target 약 -$70~-100 감도 |
| Terminal growth | 3.0% base; bear 2.0%, bull 3.5% | AI optical duration은 있지만 hardware ASP erosion 존재 | terminal g보다 FY27~FY30 FCF ramp가 더 중요 |
| Capex intensity | FY27 14~18% revenue base; red >22% | Greensboro/EML/UHP capacity build | capex가 매출보다 빠르면 FCF target 하향 |
| FCF margin | FY27 16%, FY30 23~25% bull needed | Q3 GM 47.9%가 FCF로 전환되는지 검증 | FCF margin <10% 지속이면 EV/Sales bull 배제 |
| Dilution/share count | Q3 FY26 non-GAAP diluted 95.2M; Q4 guide 102.0M; red flag >105M | convertible/preferred and stock comp bridge | share count +5%는 per-share target 약 -5% |
| Base output | FY27 revenue $5.1B, 15.5x~19x EV/Sales sensitivity, base target around $950 | reverse DCF sanity check | 현재가 $946.90은 base와 거의 같아 신규 안전마진 낮음 |
[시나리오] 이 표는 시나리오 항목을 기준으로 주요 수치, 판단 기준, 무효화 조건 정보를 비교한다. 가격선, Green/Red, 매출·GM·FCF 임계값이 있는 행을 먼저 읽고 나머지는 근거 확인용으로 쓴다.
| 시나리오 | 확률 | 12m 목표가 | 수익률 | 조건 | 무효화 조건 |
|---|---|---|---|---|---|
| Bear | 25 | 610 | -36.8 | Q4 revenue < $960M, GM < 45%, capex/working-capital drains FCF, 1.6T commentary stays demo/qualification | $760 주간 종가 회복 + 다음 분기 guide 상향 |
| Base | 45 | 950 | -1.5 | Q4 at midpoint $985M, FY27 revenue model $5.0B, GM 46~49%, CN growth but multiple normalizes to 15~17x FY27 sales | $820 이탈 후 4주 회복 실패 |
| Bull | 30 | 1380 | 43.1 | Q4 > $1.01B, FY27 revenue > $6B, GM 50%+, 1.6T/UHP laser allocation expands, IT stabilizes | Q4 beat but guide raise absent or capex intensity > 22% |
Valuation sensitivity table
현재가가 의미 있는 이유는 sensitivity가 비대칭이기 때문이다. FY27 매출 $5.1B와 15.5x EV/Sales면 현재가 근처가 base지만, 매출이 $4.7B로 낮아지거나 multiple이 14x로 내려가면 $780 전후가 된다. 반대로 $1,380 bull은 FY27 $6B와 19.5x가 동시에 필요하다.
[Revenue/FCF base] 이 표는 Revenue/FCF base 항목을 기준으로 주요 수치, 판단 기준, 사용 조건 정보를 비교한다. 가격선, Green/Red, 매출·GM·FCF 임계값이 있는 행을 먼저 읽고 나머지는 근거 확인용으로 쓴다.
| Revenue/FCF base | Multiple | Implied price | 역할 | 사용 조건 |
|---|---|---|---|---|
| FY27 revenue $4.2B | 10x | $500 | bear floor | Q4 miss 또는 GM 하락이면 적용 |
| FY27 revenue $4.2B | 12x | $600 | bear target | AI demand는 유지되지만 LITE share/margin 확신이 낮을 때 |
| FY27 revenue $4.7B | 14x | $780 | lower base | Q4 midpoint 하회, FY27 guide 보수적 |
| FY27 revenue $5.1B | 15.5x | $950 | base target | Q4 guide midpoint~상단, GM 46~49% |
| FY27 revenue $5.5B | 17x | $1,110 | upper base | Q4 상단 초과, customer diversity 개선 |
| FY27 revenue $6.0B | 19.5x | $1,380 | bull target | 1.6T/UHP volume allocation + GM 50% 근접 |
| FY27 revenue $6.5B | 22x | $1,690 | mania / scarcity overshoot | 매수 근거가 아니라 trimming reference |
| FY28 revenue $7.0B | 18x | $1,480 | duration bull | 2028 Greensboro ramp가 FY28 숫자에 들어올 때만 사용 |
| FY30 FCF $2.0B | 35x FCF | $830 | FCF sanity check | multiple-sales bull이 FCF로도 설명되는지 확인 |
| FY30 FCF $2.5B | 40x FCF | $1,180 | FCF bull | FCF margin 25%에 가까워질 때만 |
| FY30 FCF $3.0B | 45x FCF | $1,600 | extreme bull | long-term optical profit pool 장악이 필요 |
현재 가격에서 신규 매수의 적은 'LITE가 나쁜 회사'가 아니라 '좋은 뉴스가 이미 가격에 많이 들어간 상태'다. Q4 FY26 guide 상단 초과가 없으면 $1,120 위 추격은 기대수익률을 낮춘다.
9. 수급·신용분석 - 지수 편입, short interest, 변동성
미국 주식에는 한국식 신용잔고율이 없다. 대신 short interest, options volume, index inclusion flow, convertible/preferred dilution을 본다. 확인 가능한 공개 snapshot에서 short interest는 9.43M주, shares outstanding 대비 12.12%, float 대비 13.26%다. 이는 squeeze fuel이지만, Q4 miss가 나오면 long de-risking과 함께 하락 탄력도 키운다.
Short interest
9.43M shares
12.12% of shares outstanding / 13.26% of float
Shares outstanding
77.80M shares out
Q3 non-GAAP diluted 95.2M / Q4 guide 102.0M
Index/flow
S&P 500/Nasdaq-100 편입성 flow는 positive, 이벤트 후 되돌림 가능
Dilution
preferred/convertible 구조는 per-share target 계산에 반영
[상태] 이 표는 상태 항목을 기준으로 주요 수치, 판단 기준, 행동 정보를 비교한다. 가격선, Green/Red, 매출·GM·FCF 임계값이 있는 행을 먼저 읽고 나머지는 근거 확인용으로 쓴다.
| 상태 | 조건 | 가격 메커니즘 | 행동 |
|---|---|---|---|
| Green | Short % <10%, Q4 beat, guide raise | squeeze가 trend-following으로 바뀜 | $1,120 돌파 후 pullback 매수 가능 |
| Yellow | Short 10~15%, guide midpoint | volatility만 커지고 방향성은 earnings-dependent | 현재 Watch/Hold 유지 |
| Red | Short >15%, Q4 miss, put volume 급증 | squeeze fuel보다 unwind risk가 큼 | $760 risk line 적용 |
10. 차트 기술적 분석 - 좋은 상승과 나쁜 상승의 분기
현재 LITE의 기술적 위치는 강하지만 멀리 뻗어 있다. 50일선 $846.87, 200일선 $442.62, RSI 53.28은 momentum이 아직 죽지 않았음을 보여준다. 다만 200일선과의 괴리가 크기 때문에 earnings miss가 나오면 valuation reset 폭도 커질 수 있다.
좋은 상승
Q4 revenue $1.01B 초과, GM 48.5% 이상, CN 매출 $880M 이상, $980 위에서 거래량 감소 없이 안착. 이 경우 $1,120 돌파 후 pullback 매수로 바꾼다.
나쁜 상승
실적 상향 없이 index/short squeeze로 $1,120 이상 급등. 이 경우 신규 금지, 보유자는 20~35% trim. Q4 전까지는 $900 아래 재진입 계획만 유지.
[도구] 이 표는 도구 항목을 기준으로 주요 수치, 판단 기준, 행동 규칙 정보를 비교한다. 가격선, Green/Red, 매출·GM·FCF 임계값이 있는 행을 먼저 읽고 나머지는 근거 확인용으로 쓴다.
| 도구 | 확인값 | 해석 | 행동 규칙 |
|---|---|---|---|
| MACD | 공식 산출 미확보 | 가격이 50D 위라 단기 momentum은 살아 있음 | MACD만으로 신규 비중 확대 금지 |
| RSI | 53.28 | 과매수 극단은 아니지만 변동성 구간 | RSI 70 초과 + 실적 상향 없음이면 trim |
| 일목균형표 | 직접 산출 미실행 | 가격은 장기 구름 위로 해석 가능 | $760 이탈이면 momentum thesis 중단 |
| 5/25/100/200D | 50D/200D만 공개 snapshot 확보 | 50D $846.87, 200D $442.62 | 100D 미확보는 evidence limitation에 기록 |
11. 향후 12개월 카탈리스트 일정
카탈리스트는 뉴스 headline보다 숫자로 판정해야 한다. 특히 Q4 FY26 실적은 “AI 광통신은 진짜다”에서 “현재 valuation도 맞다”로 넘어가는 첫 시험대다. OFC/1.6T 발표, Apple cycle, Greensboro ramp, debt/refi는 모두 보조 카탈리스트다.
[카탈리스트 impact vs verification score] 카탈리스트의 영향도와 검증 수준을 분리한다. 영향도만 높고 검증이 낮은 뉴스는 비중 확대 근거가 아니라 질문 목록이다.
mixed input input confidence: medium
Impact가 높아도 verification이 낮으면 position size를 낮춘다
[시기] 이 표는 시기 항목을 기준으로 주요 수치, 판단 기준, 모델 조정 정보를 비교한다. 가격선, Green/Red, 매출·GM·FCF 임계값이 있는 행을 먼저 읽고 나머지는 근거 확인용으로 쓴다.
| 시기 | 카탈리스트 | Bull 판정 | Bear 판정 | 모델 조정 |
|---|---|---|---|---|
| 2026-08 전후 | Q4 FY26 earnings | Revenue > $1.01B, GM >=48.5%, EPS guide raise | Revenue < $960M 또는 GM <45% | Bull 확률 +10%p / -15%p |
| 2026 H2 | 1.6T shipment commentary | volume shipment, customer qualification, multi-customer wording | demo/sampling 반복 | EV/Sales target +2x / -3x |
| 2026 H2 | Apple new model cycle | IT/3D sensing YoY 회복 | share normalization/price-down | IT optionality +1x / 0x |
| 2026~2027 | Cloud Light synergy | module revenue growth with GM stable | module mix GM dilution | base target +/- $80 |
| 2027 | FY27 guide | Revenue > $5.5B, GM 48~51% | Revenue < $4.7B | core thesis 재평가 |
| 2027~2028 | Greensboro milestones | tool install, hiring, customer qualification | schedule delay | long-duration multiple +/- 1~2턴 |
다음 earnings call 질문 은행
아래 질문은 단순 정보 수집이 아니라 rating 변경에 직접 연결된다. Green 답변이 7개 이상이면 Watch/Hold에서 constructive Buy 후보로 올리고, Red 답변이 4개 이상이면 $760 risk line 전에도 position size를 낮춘다.
[체크 항목] 이 표는 체크 항목 항목을 기준으로 주요 수치, 판단 기준, 행동 정보를 비교한다. 가격선, Green/Red, 매출·GM·FCF 임계값이 있는 행을 먼저 읽고 나머지는 근거 확인용으로 쓴다.
| 체크 항목 | 컨퍼런스콜 질문 | Green 답변 | Red 답변 | 행동 |
|---|---|---|---|---|
| Q4 FY26 revenue | 매출이 $960M~$1.01B range 어디에 위치했는가? | $1.01B 초과 + 다음 분기 guide 상향 | $960M 미만 또는 guide 하향 | base 확률을 즉시 조정 |
| CN mix | Cloud & Networking이 total revenue의 몇 %였는가? | 90% 근접 + margin 유지 | CN mix 상승에도 GM 하락 | mix quality 확인 |
| 1.6T | 1.6T가 demo, qualification, volume shipment 중 어디인가? | volume shipment 또는 multi-customer production | sampling/demonstration만 반복 | bull multiple 적용 여부 |
| UHP/CW laser | NVIDIA/SiPh external light-source 수요가 주문으로 바뀌었는가? | multi-year allocation 또는 capacity reservation | R&D/demo language | Greensboro long option 재평가 |
| Cloud Light | transceiver revenue가 component margin을 희석했는가? | module 매출 증가 + GM 유지 | module mix 증가 + GM 하락 | TAM 확대 vs margin dilution |
| Capex | Q3의 $125M급 capex가 recurring인지 일시인지? | revenue growth와 같이 올라가는 capex | revenue보다 빠른 capex | FCF multiple 재계산 |
| Inventory | inventory가 demand visibility를 반영하는가, push-out risk인가? | inventory turn 유지 | 재고 증가가 매출 증가를 크게 초과 | working-capital discount |
| Customer concentration | top customer 또는 top 10 비중이 낮아졌는가? | top customer <15% | top customer >25% | discount rate +100bp |
| Apple/3D sensing | IT segment가 new iPhone/AR cycle로 회복하는가? | 2Q 연속 YoY 성장 | price-down/share normalization | IT optionality 제거 |
| ROADM/EDFA | transport/DCI 수요가 cloud와 같이 회복하는가? | transport order recovery | telecom capex push-out | CN composition quality |
| China competition | Innolight/Eoptolink/Accelink ASP pressure 언급이 있는가? | price discipline | second-source price-down | GM sensitivity 강화 |
| Debt/refi | convertible/preferred 구조가 per-share upside를 얼마나 희석하는가? | diluted shares 안정 | diluted shares >105M | target per share 하향 |
| Share repurchase | 자사주 매입 재개 조건이 생겼는가? | FCF positive + leverage low | capex 때문에 buyback 후순위 | capital return optionality |
| Guidance language | sold out/allocated/capacity constrained 같은 단어가 숫자와 같이 나오는가? | capacity constraint + revenue guide | 단어만 있고 숫자 없음 | headline discount |
| Peer read-through | COHR/AAOI/MRVL/CRDO가 같은 방향으로 말하는가? | peer GM/revenue도 동반 상승 | peer가 ASP pressure 언급 | sector multiple 조정 |
12. 시나리오 분석 - Bear/Base/Bull, expected return, position sizing
확률가중 12개월 기대수익률은 Bear 25%, Base 45%, Bull 30%로 계산하면 약 +7.5%다. 이 수치는 배당이 없고 변동성이 큰 주식에 충분한 신규 안전마진은 아니다. 따라서 현재가에서는 Hold/Watch가 더 정확하고, 신규는 가격 조정 또는 실적 상향 뒤에만 비중을 늘린다.
[시나리오] 이 표는 시나리오 항목을 기준으로 주요 수치, 판단 기준, 무효화 조건 정보를 비교한다. 가격선, Green/Red, 매출·GM·FCF 임계값이 있는 행을 먼저 읽고 나머지는 근거 확인용으로 쓴다.
| 시나리오 | 확률 | 12m 목표가 | 수익률 | 조건 | 무효화 조건 |
|---|---|---|---|---|---|
| Bear | 25 | 610 | -36.8 | Q4 revenue < $960M, GM < 45%, capex/working-capital drains FCF, 1.6T commentary stays demo/qualification | $760 주간 종가 회복 + 다음 분기 guide 상향 |
| Base | 45 | 950 | -1.5 | Q4 at midpoint $985M, FY27 revenue model $5.0B, GM 46~49%, CN growth but multiple normalizes to 15~17x FY27 sales | $820 이탈 후 4주 회복 실패 |
| Bull | 30 | 1380 | 43.1 | Q4 > $1.01B, FY27 revenue > $6B, GM 50%+, 1.6T/UHP laser allocation expands, IT stabilizes | Q4 beat but guide raise absent or capex intensity > 22% |
[투자자 상태] 이 표는 투자자 상태 항목을 기준으로 주요 수치, 판단 기준, 강제 규칙 정보를 비교한다. 가격선, Green/Red, 매출·GM·FCF 임계값이 있는 행을 먼저 읽고 나머지는 근거 확인용으로 쓴다.
| 투자자 상태 | 현재 상태 | 기본 포지션 | 추가 조건 | 강제 규칙 |
|---|---|---|---|---|
| 0% 보유 | 현재 $964 | 계획비중 10~15% starter 이하 | $820~900에서 25%까지 | Q4 > $1.01B + GM 48.5%일 때만 추격 허용 |
| 이미 보유 | 수익권 | $1,120 위 20~35% trim | $900 아래 재매수 계획 | $760 이탈 시 절반 축소 |
| 고비중 보유 | portfolio risk 큼 | single-name risk 8~10% 상한 | earnings 전 hedge 또는 trim | Q4 miss 시 자동축소 |
| 단기 trader | event volatility | $980~1,120 chase 금지 | $900 support 확인 | gap-up 후 volume fade면 당일 축소 |
13. 리스크 분석 - valuation, execution, customer concentration
가장 큰 리스크는 수요 부재가 아니다. 수요는 강하다. 더 큰 리스크는 valuation이 이미 강한 수요를 반영했고, execution miss가 나오면 multiple이 먼저 낮아진다는 점이다. 두 번째는 customer concentration이다. Apple/3D sensing과 hyperscaler optical 수요는 모두 규모가 크지만, 특정 고객의 qualification/order timing이 흔들리면 분기 매출 변동이 커진다.
[표 해석] 리스크를 단어가 아니라 발생 가정→인과 사슬→손익·주가 영향→무효화 조건으로 읽게 하는 표다. 무효화 조건이 확인되기 전에는 좋은 뉴스라도 position size를 키우지 않는다.
| 리스크 | 발생 가정 | 인과 사슬 | 매출 영향 | 영업이익 영향 | 주가 영향 | 무효화 조건 |
|---|---|---|---|---|---|---|
| Valuation compression | 현재 EV/Sales 29.66x와 P/FCF 646x가 FY27~FY28 성공을 선반영한다는 가정 | Q4 FY26 매출이 guide 중간만 충족하거나 GM이 48% 아래로 내려가면 multiple이 먼저 낮아진다. | 매출 -0~5%; valuation effect가 주효 | OPM -0~2%p | 주가 -15~30% | Q4 매출 >$1.01B, non-GAAP GM 48.5% 이상, 다음 분기 guide 상향 |
| 1.6T ramp 일정 지연 | Q3 FY26 자료의 Q4 FY26 ramp on track 문구가 volume revenue로 전환되지 않는 경우 | 고객 qualification 또는 packaging/yield 지연 → CN 매출과 GM 동시 둔화 | FY27 매출 -8~12% | OPM -3~5%p | 주가 -20~35% | 1.6T volume shipment 또는 multi-customer production wording 확인 |
| Apple LiDAR/3D Sensing 의존 오해 | Apple/iPhone LiDAR 비중을 공식 수치처럼 과대평가하는 경우 | IT segment 회복이 없는데 option value만 target에 넣어 valuation이 부풀어짐 | 매출 -2~5% proxy | OPM -1~2%p | 주가 -5~12% | IT 매출 2분기 연속 YoY 성장 또는 공식 customer/content disclosure |
| 800G/EML ASP 하락 | EML launch premium이 중국/COHR/AAOI second-source와 고객 price-down으로 빠지는 경우 | unit은 늘어도 ASP 하락과 expedite cost가 GM을 압박 | 매출 -3~8% | OPM -3~6%p | 주가 -18~30% | non-GAAP GM 48% 이상 유지, ASP pressure 언급 없음 |
| Cloud Light / Systems 통합 dilution | Cloud Light module 매출은 늘지만 module margin이 component margin보다 낮은 경우 | Systems 성장 → mix dilution → FCF 전환 지연 | 매출 +0~5%여도 질 저하 | OPM -2~4%p | 주가 -10~22% | Systems/Cloud Light 성장과 GM 유지가 동시 확인 |
| Coherent/AAOI/Marvell 경쟁 | COHR/AAOI/Innolight가 1.6T order를 가져가거나 MRVL/DSP·SiPh 쪽이 profit pool을 흡수하는 경우 | 고객 dual-source와 architecture shift가 LITE의 component content를 줄임 | FY27 매출 -5~10% | OPM -2~5%p | 주가 -15~28% | peer call에서 ASP pressure 부재, LITE multi-customer qualification 확인 |
| 신용/희석/자본배분 | NVIDIA $2B 투자로 유동성은 좋아졌지만 preferred/convertible과 capex가 per-share upside를 희석하는 경우 | Q4 guide diluted shares 102M 기준으로 EPS와 target per share가 낮아짐 | 매출 영향 없음 | EPS 희석 5~10% | 주가 -5~15% | FCF positive, diluted share count 안정, no dividend/no buyback 한계 명시 |
| AI capex cycle 둔화 | hyperscaler capex 증가율이 둔화하거나 GPU/networking spend가 digestion에 들어가는 경우 | order push-out → inventory build → guide 하향 | 매출 -10~20% | OPM -4~8%p | 주가 -30~45% | MSFT/META/GOOGL/AMZN capex guide 유지와 LITE backlog commentary 강화 |
Failure-mode map: 강한 story가 틀리는 구체 경로
[Failure mode] 이 표는 Failure mode 항목을 기준으로 주요 수치, 판단 기준, 대응 정보를 비교한다. 가격선, Green/Red, 매출·GM·FCF 임계값이 있는 행을 먼저 읽고 나머지는 근거 확인용으로 쓴다.
| Failure mode | 구체 상황 | 원인 | 관찰 신호 | 대응 |
|---|---|---|---|---|
| Revenue-only beat | 매출은 guide 상단이지만 GM이 하락 | module mix, expedite cost, ASP down | 주가가 처음엔 오르고 며칠 뒤 multiple reset | 신규 금지, GM 회복 전 trim |
| Capacity mirage | capex와 facility 뉴스는 많지만 shipment 언급 부족 | 투자 선행, customer qualification 지연 | FCF negative 또는 inventory build | DCF discount rate 상향 |
| Customer timing cliff | 한 고객의 order pull-forward 후 공백 | hyperscaler qualification batch timing | 다음 분기 guide 하향 | top customer disclosure 재확인 |
| ASP deflation | unit growth가 ASP 하락을 못 이김 | 중국 vendor second-source, customer price-down | GM 45% 아래 | EV/Sales target 3~5턴 하향 |
| Apple drag | iPhone/3D sensing content/share 하락 | IT segment YoY 감소 | IT margin compression | IT option value 제거 |
| Peer substitution | COHR/AAOI/Innolight가 더 빠른 qualification 확보 | customer wants dual sourcing | LITE backlog language 약화 | position size 축소 |
| Architecture shift | CPO/linear optical가 특정 LITE component content를 줄임 | DSP/SiPh vendor가 profit pool 흡수 | optical revenue grows but component margin falls | profit-pool table 재작성 |
| Funding dilution | preferred/convertible가 EPS upside를 희석 | share count 증가 | diluted shares >105M | per-share target 5~10% 하향 |
| Macro reset | AI capex multiple 전체가 압축 | rates/upfront depreciation concern | MRVL/CRDO/COHR 동반 하락 | sector multiple 하향 |
| Execution bottleneck | yield/ramp issue | InP fab or packaging complexity | shipment delay and expedite cost | bull 확률 half |
| Telecom relapse | ROADM/EDFA recovery가 멈춤 | carrier capex freeze | CN composition weaker | legacy transport value 낮춤 |
| Industrial recession | EV/metal/semicap downturn | IT segment does not recover | revenue floor 약화 | bear downside wider |
Customer concentration stress
AI optical supplier는 큰 고객 없이는 빠르게 크기 어렵다. 하지만 같은 고객 집중은 매출 visibility와 bargaining power를 동시에 만든다. 따라서 customer concentration은 단순 bad factor가 아니라 margin과 같이 읽어야 한다.
[Customer case] 이 표는 Customer case 항목을 기준으로 주요 수치, 판단 기준, 확인 방법 정보를 비교한다. 가격선, Green/Red, 매출·GM·FCF 임계값이 있는 행을 먼저 읽고 나머지는 근거 확인용으로 쓴다.
| Customer case | 의미 | 모델 처리 | Target impact | 확인 방법 |
|---|---|---|---|---|
| Top customer 10% | 매출 분산 진전 | discount rate -50bp | base target +$40~70 | 10-K customer table 확인 |
| Top customer 15% | 현재 허용 가능한 집중 | base 유지 | target 유지 | AI optical supplier는 어느 정도 집중을 감수 |
| Top customer 25% | single customer timing risk | discount rate +100bp | target -$100 | order timing cliff 대비 |
| Top customer 35% | Apple/hyperscaler cliff risk | multiple -3턴 | target -$180~250 | customer A/B disclosure 필요 |
| Top 3 customers 50% | capacity allocation은 강하지만 bargaining power 낮음 | GM threshold 49% 필요 | target unchanged only if GM holds | customer concentration section |
| Top 3 customers 70% | AAOI-like concentration discount | multiple -5턴 | target -$300+ | new customer qualification 없으면 위험 |
| Apple content +20% | IT recovery option | IT revenue +$80~150M | target +$20~50 | iPhone/AR model 채택 확인 |
| Apple content -20% | VCSEL price-down/share loss | IT revenue -$50~120M | target -$20~60 | IT segment YoY 확인 |
| Hyperscaler A delay 1Q | shipment timing issue | working capital/inventory pressure | target -$80 short-term | guide 유지 여부 |
| Hyperscaler A cancels allocation | demand/share issue | multiple reset | bear case | peer가 같은 order를 가져가는지 |
| New hyperscaler wins | concentration discount 축소 | multiple +1~2턴 | target +$90~150 | named or clearly multi-customer wording |
$760 아래 주간 종가는 단순 pullback이 아니라 thesis 품질을 다시 재야 하는 가격 신호다. 그 가격에서 CN 성장, GM, capex, customer concentration을 새로 확인하기 전까지 자동 물타기를 하지 않는다.
14. Peer 비교 - LITE는 optical scarcity premium 최상단
LITE의 비교군은 하나가 아니다. COHR은 소재/레이저/광모듈 통합 peer, AAOI/Innolight/Eoptolink는 transceiver beta, MRVL/CRDO/SMTC는 optical link silicon/linear optical adjacent peer, IPGP는 산업 레이저 cycle peer다. LITE의 premium은 AI optical bottleneck을 직접 먹는다는 점에서 일부 정당화되지만, 같은 이유로 miss의 충격도 크다.
[Peer EV/Sales와 Forward P/E] LITE가 COHR·AAOI·MRVL보다 어떤 premium을 받는지 보여준다. peer가 같은 매출 성장에 낮은 multiple을 받으면 LITE의 실적 허들이 올라간다.
model/proxy input confidence: medium
Peer 값은 공개 commercial snapshot/model 혼합. 정밀 상대가치가 아니라 위치 비교
*AI의 견해입니다
[Peer] 이 표는 Peer 항목을 기준으로 주요 수치, 판단 기준, 한계/용도 정보를 비교한다. 가격선, Green/Red, 매출·GM·FCF 임계값이 있는 행을 먼저 읽고 나머지는 근거 확인용으로 쓴다.
| Peer | 기준일 | 사용한 값 | 한계/용도 |
|---|---|---|---|
| LITE | 2026-05-22 close / StockAnalysis delayed | EV/Sales 29.66x, forward P/E 59.78x | 기준가 $946.90 |
| COHR | 2026-05-21 close / local COHR round2 manifest | basis close $378.00, source limitation: delayed/free quote | 1.6T capability and NVIDIA partnership read-through |
| AAOI | 2026-05-21 intraday/close vicinity / local AAOI round5 manifest | basis $179.48, official live quote unavailable | 1.6T order/shipment timing read-through |
| MRVL/CRDO/SMTC | 2026-05-21~22 public snapshot/model | adjacent silicon/linear optical basket | not exact same vendor feed; used only as relative architecture proxy |
[티커] 이 표는 티커 항목을 기준으로 주요 수치, 판단 기준, 판정 정보를 비교한다. 가격선, Green/Red, 매출·GM·FCF 임계값이 있는 행을 먼저 읽고 나머지는 근거 확인용으로 쓴다.
| 티커 | 비교 축 | 가격 | EV | EV/Sales | Forward P/E | EV/EBITDA | ROE | Short % | 판정 |
|---|---|---|---|---|---|---|---|---|---|
| LITE | 광 레이저/트랜시버/ROADM/3D sensing | $946.9 | $73.8B | 29.7x | 59.8x | 145.1x | 22.8% | 12.1% | AI 광 scarcity premium 최상단 |
| COHR | InP/SiC/materials/lasers/optical modules | $350.0 | $57.0B | 8.5x | 38.0x | 43.0x | 6.5% | 8.0% | 규모와 소재 내재화, 통합 리스크 |
| AAOI | datacom transceiver pure beta | $179.5 | $13.8B | 27.0x | 85.0x | n/m | -8.0% | 10.5% | 순수 optical beta, 고객 집중 |
| MRVL | DSP/ASIC/SerDes/data-center silicon | $190.7 | $170.0B | 20.6x | 50.0x | 64.0x | 19.3% | 2.0% | DSP와 custom silicon 대형 peer |
| CRDO | AEC/LRO/retimer 고성장 | $248.0 | $46.0B | 72.0x | 90.0x | n/m | -2.0% | 8.0% | 광/동 전환의 고multiple peer |
| IPGP | 산업 fiber laser | $78.0 | $3.4B | 2.4x | 24.0x | 15.0x | 3.5% | 4.0% | 산업 레이저 cycle 비교군 |
| SMTC | Signal integrity + mixed signal | $146.5 | $14.0B | 13.3x | 66.1x | 85.7x | 8.0% | 11.0% | SI optical optionality + IoT cushion |
[논점] 이 표는 논점 항목을 기준으로 주요 수치, 판단 기준, 실전 판정 정보를 비교한다. 가격선, Green/Red, 매출·GM·FCF 임계값이 있는 행을 먼저 읽고 나머지는 근거 확인용으로 쓴다.
| 논점 | 내용 | 실전 판정 |
|---|---|---|
| LITE premium 정당화 | CN 매출 가속, 1.6T/EML/UHP laser scarcity, NVIDIA-linked capital/capacity, high GM rebound | Q4 guide 상단 초과와 GM 48.5% 이상 |
| LITE discount 사유 | module ASP pressure, customer concentration, preferred/dilution, capex-heavy expansion | FCF margin이 10% 아래로 머무르면 discount 확대 |
| COHR 대비 | LITE가 더 optical laser beta, COHR은 더 diversified | LITE multiple premium은 growth가 확실할 때만 유지 |
| AAOI 대비 | LITE는 balance sheet와 component IP가 강하지만 pure module upside는 AAOI가 더 큼 | AAOI가 신규 hyperscaler를 공개하면 multiple spread 축소 |
| MRVL/CRDO 대비 | silicon vs optical component profit pool 비교 | 광/동 architecture decision이 relative multiple을 좌우 |
15. 투자 전략 - 가격대별 행동 매트릭스
이 보고서의 전략은 “좋은 회사니까 산다”가 아니라 “가격대마다 어떤 증거가 필요한지”를 정하는 방식이다. 현재가 $946.90에서는 신규 큰 비중보다 evidence update를 기다리는 편이 낫다. $820~900은 첫 매수 후보, $720~820은 실적 intact일 때 비중 확대 후보, $1,120 이상은 실적 상향이 없으면 현금화 후보다.
[가격 구간] 이 표는 가격 구간 항목을 기준으로 주요 수치, 판단 기준, 왜 이 행동인가 정보를 비교한다. 가격선, Green/Red, 매출·GM·FCF 임계값이 있는 행을 먼저 읽고 나머지는 근거 확인용으로 쓴다.
| 가격 구간 | 행동 | 왜 이 행동인가 |
|---|---|---|
| $1,250 이상 | 신규 금지, 보유자는 20~35% 이익실현 | FY27 매출 $6B+와 1.6T margin expansion이 동시에 확인되지 않으면 multiple이 먼저 식는다. |
| $1,120~1,250 | 보유만 허용, 신규는 Q4 beat 뒤 pullback 대기 | 52주 고점 재도전 구간. 지수 편입/AI scarcity flow가 과열되면 spike가 나올 수 있으나 risk/reward가 좁다. |
| $980~1,120 | 기존 보유자는 half trim 규칙 가동 | 현재가 부근. Q4 FY26 매출 guide 상단 초과와 GM 48%+가 없으면 upside가 headline 의존으로 바뀐다. |
| $900~980 | WATCH/HOLD, starter 신규는 계획비중 10~15% 이하 | Q3 이후 실적은 강하지만 EV/Sales 29.7x는 2027 성공을 이미 요구한다. |
| $820~900 | 조건부 starter 15~25% | 50일선 근처. Q4 guide 유지, short interest 급증 없음, CN backlog 언급 유지가 필요하다. |
| $720~820 | 기본 매수 검토 25~40% | valuation risk가 줄어든다. 단 Q4 매출 $960M 미만이면 진입하지 않는다. |
| $620~720 | 분할 매수 구간 | AI 광 수요가 유지되고 주가만 multiple reset을 받은 경우다. gross margin 44% 이상 필요. |
| $520~620 | thesis 재계산 후 매수 | 200일선과 큰 괴리가 남아도, 매출 ramp가 깨지지 않으면 expected return이 회복된다. |
| $438~520 | 강제 검증 구간 | 200일선 인접. 이 가격은 산업 cycle 훼손 또는 시장 전체 selloff를 반영할 확률이 높다. |
| $438 미만 | 기존 thesis 중단, 새 모델 작성 | AI optical scarcity premium이 붕괴했다는 가격 신호다. DCF와 balance sheet부터 다시 계산한다. |
상황별 운영 프로토콜
[상황] 이 표는 상황 항목을 기준으로 주요 수치, 판단 기준, 다음 확인 정보를 비교한다. 가격선, Green/Red, 매출·GM·FCF 임계값이 있는 행을 먼저 읽고 나머지는 근거 확인용으로 쓴다.
| 상황 | 첫 행동 | 포지션 규칙 | 다음 확인 |
|---|---|---|---|
| Q4 revenue beats, GM beats, stock gaps +10% | 첫 30분 추격 금지, 종가가 VWAP 위인지 확인 | $1,120 위면 기존 보유 20~35% trim | 다음날 $1,080~1,120 retest에서 volume 확인 |
| Q4 revenue beats, GM misses | headline beat와 quality miss를 분리 | 신규 금지, 보유 trim | 다음 call에서 module mix/capex 원인 확인 |
| Q4 revenue misses, guide maintained | order timing인지 demand issue인지 확인 | $760까지 risk band 열림 | 다음 분기 CN revenue gate 재설정 |
| Q4 revenue misses, guide down | valuation thesis 훼손 | 기존 thesis 중단, 절반 축소 | DCF와 EV/Sales base를 bear로 전환 |
| 가격 $820~900으로 조정, KPI intact | starter 15~25% | stop $760 | Q4 call 전 총비중 50% 초과 금지 |
| 가격 $720~820, sector selloff only | 실적 intact이면 분할 | peer도 같은 selloff인지 확인 | COHR/MRVL read-through 확인 |
| 가격 $1,250 이상, no guide raise | valuation overshoot로 처리 | 보유 35~50% trim | 다음 evidence 전 신규 중단 |
| Apple/AR positive news only | core CN thesis와 분리 | 추격 금지 | IT revenue actual 확인 전 optionality만 반영 |
| Greensboro milestone positive | 2028 option으로 처리 | 2026~2027 target에는 1~2턴만 반영 | capex 부담 확인 |
| Peer AAOI/COHR warns on ASP | LITE에도 ASP sensitivity 적용 | GM threshold 엄격화 | base target -$80~-150 |
16. 모니터링 KPI 대시보드
다음 12개월 체크리스트는 모호한 표현 대신 수치로만 판정한다. Green이 6개 이상이면 Watch에서 Buy 후보로 올릴 수 있고, Red가 3개 이상이면 현재 thesis를 줄인다.
[KPI 신호등 score] 다음 12개월 KPI를 신호등 점수로 압축한다. Green 6개 이상이면 Buy 후보, Red 3개 이상이면 thesis 축소다.
mixed input input confidence: medium
100에 가까울수록 bull thesis에 유리. 현재는 CN 강세와 valuation 부담이 공존
[KPI] 이 표는 KPI 항목을 기준으로 주요 수치, 판단 기준, 판정 메커니즘 정보를 비교한다. 가격선, Green/Red, 매출·GM·FCF 임계값이 있는 행을 먼저 읽고 나머지는 근거 확인용으로 쓴다.
| KPI | Green | Yellow | Red | 판정 메커니즘 |
|---|---|---|---|---|
| Q4 FY26 revenue | >= $1.01B | $960M~$1.01B | < $960M | Q3 beat가 one-off인지 판정 |
| Non-GAAP GM | >= 48.5% | 45~48.5% | < 45% | EML/UHP laser scarcity가 margin으로 전환되는지 |
| Cloud & Networking quarterly revenue | >= $880M | $780~880M | < $780M | AI cycle의 실제 매출화 |
| Capex intensity | 매출의 12~16% | 16~22% | > 22% | capacity 선투자가 FCF를 잠식하는지 |
| Inventory growth vs revenue growth | 재고 증가율 <= 매출 증가율 | 재고 +10%p 초과 | 재고 +25%p 초과 | order push-out 리스크 |
| Short interest | < 10% shares | 10~15% | > 15% | squeeze와 reversal 양쪽 위험 |
| Net debt / EBITDA | < 0.5x | 0.5~1.5x | > 1.5x | NVIDIA cash와 FCF가 부채 우려를 낮추는지 |
| Customer concentration | Top customer < 15% | 15~25% | > 25% | Apple/hyperscaler 집중 discount |
| 1.6T evidence | volume shipment + customer qualification | sampling/qualification | demo만 반복 | OFC demo가 매출로 바뀌는지 |
| Industrial Tech | 2분기 연속 YoY 성장 | flat | YoY -10% 이하 | CN 외 완충판 회복 |
[주기] 이 표는 주기 항목을 기준으로 주요 수치, 판단 기준, 운영 규칙 정보를 비교한다. 가격선, Green/Red, 매출·GM·FCF 임계값이 있는 행을 먼저 읽고 나머지는 근거 확인용으로 쓴다.
| 주기 | 확인 항목 | 운영 규칙 |
|---|---|---|
| 주간 15분 | 가격 vs $900/$820/$760, short interest update, peer COHR/AAOI/CRDO price reaction | 가격만 보지 말고 peer spread를 같이 본다. |
| 월간 30분 | OFC/1.6T shipment news, hyperscaler capex commentary, Apple supply-chain rumor 검증 | rumor는 source grade 낮게 둔다. |
| 분기 60분 | 매출, GM, CN revenue, capex, inventory, diluted shares, customer concentration | 투자 판단은 분기 숫자에 더 큰 가중치. |
17. 흔한 실수 / FAQ
LITE에서 가장 흔한 실수는 AI optical demand를 곧바로 적정 주가로 번역하는 것이다. demand, margin, capex, customer concentration, dilution, multiple을 분리해서 업데이트해야 한다.
AI 광통신 수요가 폭발하면 지금 가격도 싼가?
아니다. 수요와 가격 매력은 분리해야 한다. Q3 FY26 매출과 Q4 가이던스는 강하지만, 현재 EV/Sales 29.66x는 FY27 매출과 margin 확장을 이미 요구한다. 신규 자금은 Q4 매출이 $1.01B를 넘거나 가격이 $820~900으로 내려오는 둘 중 하나가 있어야 승률이 올라간다.
LITE는 COHR보다 좋은가?
사업 노출은 더 직접적인 AI optical laser beta다. COHR는 소재/레이저/광부품 규모와 통합 범위가 넓고, LITE는 EML/UHP laser와 Cloud Light transceiver ramp의 민감도가 더 크다. 따라서 bull market에서는 LITE multiple이 더 높을 수 있지만, miss가 나오면 LITE의 multiple compression도 더 빠르다.
Apple Face ID/VCSEL exposure는 더 이상 중요하지 않은가?
중요하지만 현재 주가를 설명하는 1순위는 아니다. Apple/3D sensing은 revenue floor와 optionality를 주지만, FY26~FY27 주가의 주된 판정은 CN 매출, EML 공급, NVIDIA/hyperscaler qualification, gross margin이다. Apple 의존도가 낮아지는 것은 discount 축소 요인이지만, IT segment가 재가속하지 않으면 core upside는 CN이 만든다.
NVIDIA 투자 때문에 부채 리스크는 사라졌나?
유동성 압박은 크게 낮아졌지만, valuation 리스크로 형태가 바뀌었다. Q3 FY26 현금/단기투자는 약 $3.17B, debt는 $3.31B로 net debt는 작다. 그러나 preferred/convertible 구조와 diluted share count는 per-share upside를 갉아먹을 수 있다.
1.6T EML만 보면 되는가?
1.6T는 핵심이지만 충분조건은 아니다. 800G CW laser, UHP laser, ROADM/EDFA, Cloud Light transceiver, CPO/linear optical architecture까지 같이 봐야 한다. 매출은 한 부품이 아니라 customer bill-of-material과 capacity allocation으로 결정된다.
산업 레이저는 버려진 사업인가?
아니다. 하지만 현재 multiple을 방어하는 engine은 아니다. 산업 레이저/3D sensing은 downside cushion과 optionality 역할이다. IT 매출이 2분기 연속 YoY 성장으로 바뀌면 base target multiple을 1~2턴 올릴 수 있다.
차트가 강하면 실적 전에도 사야 하나?
현재 가격은 50일선보다 위이고 200일선과의 괴리가 매우 크다. 기술적 강세만으로 신규 풀비중을 잡으면 Q4 실적 직전 변동성에 노출된다. 가격이 $820~900으로 쉬거나, Q4 숫자가 guide 상단을 넘는 둘 중 하나가 필요하다.
short interest 13%는 bull signal인가?
squeeze fuel이 될 수 있지만 downside fuel도 된다. short가 틀린 경우 주가가 빠르게 재상승할 수 있지만, Q4 guide miss나 GM 하락이 나오면 short가 아니라 long de-risking이 가격을 끌어내린다.
DCF가 현재가보다 낮으면 회피가 정답인가?
일반 DCF는 scarcity premium과 capacity-option value를 낮게 잡기 쉽다. 그래서 본 보고서는 DCF를 목표가 한 점이 아니라 reverse DCF 검산 장치로 쓴다. 현재가를 정당화하려면 FY27~FY30 FCF가 1B 달러 이상으로 빠르게 올라가야 한다.
Apple Vision Pro 2나 AR/VR은 bull case인가?
옵션이다. AR/VR 3D sensing이 확대되면 IT segment와 VCSEL utilization에 도움이 되지만, 현재 투자 thesis의 확률가중 핵심은 AI data-center optical이다. AR/VR 뉴스만으로 $1,000 위 신규 매수를 정당화하지 않는다.
18. Bull / Bear 양측 논리 강제 대결
좋은 deep-dive는 한쪽 논리만 강화하지 않는다. 아래 표는 같은 데이터를 bull과 bear가 어떻게 다르게 읽는지, 어느 숫자가 판정권을 갖는지 정리한다.
[같은 데이터] 이 표는 같은 데이터 항목을 기준으로 주요 수치, 판단 기준, 승패 판정 정보를 비교한다. 가격선, Green/Red, 매출·GM·FCF 임계값이 있는 행을 먼저 읽고 나머지는 근거 확인용으로 쓴다.
| 같은 데이터 | Bull 해석 | Bear 해석 | 승패 판정 |
|---|---|---|---|
| Q3 FY26 revenue $808.4M | AI optical demand가 이미 LITE 손익계산서에 들어왔다. | 현재 multiple은 이 숫자의 연장선을 이미 가격에 넣었다. | Q4 $1.01B 초과면 bull 우위, $960M 미만이면 bear 우위 |
| Non-GAAP GM 47.9% | EML/UHP laser scarcity가 margin으로 보상된다. | Cloud Light module mix와 capex ramp가 margin을 누를 수 있다. | GM 48.5% 이상 유지 여부 |
| Cash/ST investment $3.17B vs debt $3.31B | NVIDIA 자금으로 refinance/convert 부담이 작아졌다. | preferred/convertible dilution과 capex가 per-share value를 낮출 수 있다. | diluted share count와 FCF after capex |
| Cloud Light acquisition | transceiver revenue 직접 확보로 served TAM이 커졌다. | module 경쟁은 AAOI/Innolight/Eoptolink와 ASP pressure가 더 직접적이다. | module GM이 component GM을 희석하는지 |
| Apple/3D sensing | iPhone/AR option이 CN 외 revenue floor를 준다. | SEC 원문은 Apple을 명명하지 않으므로 고객 추정에 과신하면 안 된다. | Customer A 비중과 IT 매출 YoY |
| Greensboro InP facility | supply bottleneck을 풀며 2028 이후 capacity option을 만든다. | mid-2028 ramp는 2026~2027 실적에는 즉시 도움을 주지 않는다. | capex, hiring, qualification milestone |
| Short interest 12.12% | 좋은 뉴스에서 squeeze가 붙을 수 있다. | miss가 나오면 crowded long과 short가 동시에 변동성을 키운다. | short % 15% 초과와 put volume |
| EV/Sales 29.66x | scarcity asset에는 과거 밴드보다 높은 multiple이 가능하다. | 과거 LITE의 normal band와 비교하면 안전마진이 작다. | FY27 revenue가 $6B 궤도인지 |
19. Evidence & Limitations
이 보고서는 확인된 사실, 사용자 제공 정보, 공개 commercial snapshot, 모델 추정을 분리했다. Apple/3D sensing exposure는 사용자가 제공한 thesis와 업계 상식에 맞지만, 이번 pass에서 접근한 10-K 원문은 Apple을 이름으로 명명하지 않았다. 따라서 top customer/Customer A는 official fact, Apple identity는 market inference로 처리한다.
Verified facts
- Q3 FY26 revenue $808.4M, non-GAAP GM 47.9%, GAAP net income $144.2M, adjusted EBITDA $293.5M.
- FY2025 CN revenue $1.4108B, IT $234.2M, CN mix 85.8%.
- Cloud Light acquisition announced at about $750M transaction value; Cloud Light had >$200M trailing revenue and 400G+ skew.
- StockAnalysis snapshot: 2026-05-22 close $946.90, after-hours $935.20, EV $73.81B, EV/Sales 29.66x, forward P/E 59.78x.
- Q3 FY26 earnings presentation: 1.6T transceivers are on track for Q4 FY26 ramp; GAAP OP $174.5M and non-GAAP OP $260.7M.
- NVIDIA official announcement: $2B investment, multibillion purchase commitment, and future capacity access rights for advanced laser components.
Limitations
- Official real-time NASDAQ quote was not captured; price basis is delayed/public 2026-05-22 close.
- Apple iPhone LiDAR/3D Sensing revenue share and NVIDIA H100/H200/B200-specific 800G EML revenue were not disclosed in checked official sources.
- Exact paid 10-year valuation bands were unavailable; valuation band chart is a regime model.
- Quarterly CN/IT mix after FY2025 is modeled where exact segment table was not captured in accessible text.
- ASP, TAM, capex correlation, optical-vs-copper curves are model/proxy charts and are not company guidance.
- Options open interest and same-day broker flow were not captured; flow section uses short interest and index/dilution framework.
Source inventory
- Lumentum Q3 FY2026 results press release A
Official Q3 FY26 revenue $808.4M, GAAP operating income $174.5M, non-GAAP operating income $260.7M, non-GAAP gross margin 47.9%, GAAP net income $144.2M, adjusted EBITDA $293.5M, Q4 FY26 guide range. - Lumentum Q3 FY2026 earnings presentation A
Official Q2/Q3 FY26 product-type revenue, GAAP/non-GAAP operating income table, balance sheet, Q4 FY26 guidance, and 1.6T transceivers on-track-for-Q4-FY26-ramp wording. - Lumentum Q2 FY2026 results press release A
Official Q2 FY26 revenue $665.5M, GAAP operating income $64.3M, non-GAAP operating income $167.7M, and Q1 FY26 comparison values. - Lumentum Q1 FY2026 results press release A
Official Q1 FY26 revenue $533.8M, GAAP operating margin 1.3%, non-GAAP operating margin 18.7%, and product-type disclosure. - Lumentum Q4 and FY2025 results press release A
Official FY2025 quarterly bridge, Q4 FY25 revenue $480.7M, FY25 revenue $1.645B, FY25 segment revenue, GAAP/non-GAAP operating margin and operating income cross-check. - Lumentum Q2 FY2025 results press release A
Official FY25 Q1/Q2 quarterly revenue, GAAP operating loss, non-GAAP operating income, and segment revenue. - Lumentum Q3 FY2025 results press release A
Official FY25 Q3 revenue, GAAP operating loss, non-GAAP operating income, and segment revenue. - Lumentum FY2025 Form 10-K A
FY2025 segment revenue and profit, segment definitions, customer concentration framework, competition and margin pressure language. - Lumentum Cloud Light acquisition announcement A
$750M Cloud Light acquisition, >$200M trailing revenue, >90% 400G+ revenue, >50% 800G in latest quarter, strategic served-opportunity claim. - Lumentum OFC 2026 AI optical infrastructure announcement A
1.6T DR4 OSFP prototype using four Lumentum 400G differential EML lasers, 800 mW SHP laser, 16-channel DWDM UHP laser, and CPO/open optical architecture context. - NVIDIA strategic partnership with Lumentum A
NVIDIA $2B investment in Lumentum, multibillion purchase commitment, future capacity access rights, and R&D/capacity rationale. - NVIDIA Spectrum-X / Quantum-X Photonics announcement A
Spectrum-X and Quantum-X silicon photonics switch architecture, 1.6 Tb/s per port framing, Lumentum ecosystem role, and availability timing. - Lumentum Greensboro AI data center laser manufacturing facility announcement A
240,000 sq ft Greensboro facility, InP CW/UHP lasers, mid-2028 production ramp, capacity expansion and domestic supply-chain context. - StockAnalysis LITE price history B
May 22, 2026 close $946.90, after-hours $935.20, 52-week high $1,085.68, recent daily OHLC rows. - StockAnalysis LITE statistics B
Market cap $73.67B, EV $73.81B, forward P/E 59.78x, EV/Sales 29.66x, 50D/200D moving averages, RSI, short interest, cash/debt, dividend and buyback-yield snapshot. - StockAnalysis LITE financials and revenue B
FY2021-FY2025 and TTM revenue, operating income, margins, FCF, diluted shares, financial history cross-check. - MarketCap.ru LITE operating cash flow history C
2016-2025 operating cash flow history cross-check where official XBRL extraction was not automated in local shell. - COHR local deep dive round 2 local
LITE vs COHR 1.6T capacity, shipment wording, architecture and customer qualification comparison. - Stock Research Hub local
Existing market artifact hub inventory and integration check. - Mandatory exemplar local
20-section decision-first structure, sticky TOC, bottom evidence/preflight, scenario/action matrix, late-section density pattern. - Optical peer exemplar local
AI optical cycle framing, ASP and hyperscaler proxy labeling, peer normalization caveats, Chart.js density. - Template local
Required section order, output file contract, 10+ Chart.js visuals, self-critique rounds, final YAML format. - Deep research readability rule local
7 mandatory readability rules: interpretation headings, AI interpretation separation, English term policy, business-structure 4-factor explanation, causal risk chain, confident conclusion, and hero decision card.
Claim-to-evidence mapping
[Claim] 이 표는 Claim 항목을 기준으로 주요 수치, 판단 기준, Limitation 정보를 비교한다. 가격선, Green/Red, 매출·GM·FCF 임계값이 있는 행을 먼저 읽고 나머지는 근거 확인용으로 쓴다.
| Claim | Source | Grade | Used in | Limitation |
|---|---|---|---|---|
| 현재가 $946.90 | StockAnalysis history/statistics | B | price dashboard, action matrix | 2026-05-22 종가, 실시간 아님 |
| Q3 FY26 revenue $808.4M | Lumentum/Nasdaq press release | A | hero, fundamentals, valuation | press-release source text |
| Q3 non-GAAP GM 47.9% | Lumentum/Nasdaq press release | A | margin thesis, KPI | non-GAAP 정의 차이 |
| FY25 CN/IT segment revenue | SEC FY2025 10-K | A | business cycle, segment chart | FY26 quarterly split은 일부 모델 |
| Cloud Light >$200M trailing revenue | Lumentum Cloud Light announcement | A | event section | closing and integration outcomes separate |
| 1.6T 200G EML/UHP demo | Lumentum OFC 2026 blog | A | events, cycle, KPI | demo is not shipment |
| Greensboro 240k sq ft facility | Lumentum facility announcement | A | supply chain, catalyst | mid-2028 ramp, long option |
| EV/Sales 29.66x | StockAnalysis statistics | B | valuation | vendor calculation may vary |
| Short interest 12.12% | StockAnalysis statistics | B | flow/credit | exact settlement date not independently pulled |
| 10-year FCF | StockAnalysis + MarketCap.ru cross-check | B/C | fundamentals | older rows proxy-labeled |
| Apple exposure | user-provided thesis + market inference | user/inference | business/risks/FAQ | not named in checked SEC source |
| ASP curves | model from industry ranges | model | cost/supply | not company disclosed |
| Hyperscaler capex | public/analyst compiled model | model | AI correlation chart | not official combined fact |
| Peer multiples | commercial public snapshots + local exemplars | B/model | peer section | not same vendor/date for every peer |
Model chart input register
[Chart] 이 표는 Chart 항목을 기준으로 주요 수치, 판단 기준, Limitation 정보를 비교한다. 가격선, Green/Red, 매출·GM·FCF 임계값이 있는 행을 먼저 읽고 나머지는 근거 확인용으로 쓴다.
| Chart | Source type | Input | Confidence | Limitation |
|---|---|---|---|---|
| segmentQuarterChart | model | FY25 segment + Q4 FY25 official + FY26 total revenue | medium | segment 공시 아님, ±$30~45M CN error band |
| tamWaterfallChart | model | FY25 revenue + AI optical ramp assumptions | medium | target math 보조, 회사 guidance 아님 |
| sensingTamChart | model/proxy | 3D sensing TAM model + IT revenue scenario | low-medium | Apple identity not verified in checked SEC source |
| dcfBridgeChart | model | WACC/terminal/FCF/share count assumptions | medium | point target보다 reverse DCF guardrail |
| peerValuationChart | B/model | public commercial snapshots + local peer reports | medium | COHR/AAOI/MRVL 기준일 2026-05-21/22로 통일 표기 |
| aspCurveChart | model/proxy | industry ASP range model | low-medium | 회사 ASP 미공시 |
20. 부록 - 용어집, 운영 규칙, preflight
용어집
[용어] 이 표는 용어 항목을 기준으로 주요 수치, 판단 기준, LITE에서 왜 중요한가 정보를 비교한다. 가격선, Green/Red, 매출·GM·FCF 임계값이 있는 행을 먼저 읽고 나머지는 근거 확인용으로 쓴다.
| 용어 | 뜻 | LITE에서 왜 중요한가 |
|---|---|---|
| EML | Electro-absorption modulated laser | 800G/1.6T optical module의 lane 속도와 signal integrity에 직접 연결 |
| Optical transceiver | 전기 신호와 광신호를 바꿔 서버·스위치 사이 데이터를 보내는 광송수신 모듈 | 800G/1.6T 세대 전환이 Cloud and Networking 매출을 좌우 |
| PAM4 | 한 번의 신호에 4단계 값을 실어 전송량을 늘리는 변조 방식 | 200G/lane, 400G/lane으로 갈수록 EML·DSP·패키징 품질이 중요 |
| DSP | Digital signal processor, 신호 왜곡을 보정하는 칩 | CPO/linear optical 전환에서 DSP 사용량이 줄거나 profit pool이 이동할 수 있음 |
| UHP laser | Ultra high-power laser | silicon photonics external light source architecture에서 중요 |
| CW laser | Continuous wave laser | co-packaged/SiPh 계열 광원으로 쓰이는 핵심 부품 |
| ROADM | Reconfigurable optical add-drop multiplexer | metro/long-haul optical transport의 가변 경로 제어 |
| EDFA | Erbium-doped fiber amplifier | 광신호 증폭, DCI/transport upgrade와 연결 |
| VCSEL | Vertical-cavity surface-emitting laser | Face ID/3D sensing/short reach sensing에 사용 |
| Hyperscaler | 초대형 클라우드 사업자 | AI 데이터센터 capex와 장기 qualification이 LITE 매출의 선행 지표 |
| Capex | 설비투자 | Greensboro/InP capacity build가 FCF를 누르는지 보는 핵심 비용 |
| NPI | New product introduction, 신제품 양산 전환 과정 | 1.6T가 demo에서 revenue ramp로 바뀌는지 판단하는 단계 |
| Cloud and Networking | 클라우드·AI·통신망용 광부품과 모듈 사업부 | FY25 매출 85.8%로 현재 valuation의 핵심 |
| Industrial Tech | 산업 레이저와 sensing용 레이저 사업부 | FY25 매출 14.2%; 3D sensing과 산업 레이저 옵션 |
| FCF | Free cash flow, 영업현금흐름에서 capex를 뺀 현금 | 고성장 매출이 주주가치로 바뀌는지 확인 |
| CPO | Co-packaged optics | switch ASIC과 optical engine을 더 가깝게 붙이는 next architecture |
| Linear optical | DSP 복잡도를 줄이려는 optical/copper architecture | power/latency/thermal tradeoff가 핵심 |
| EV/Sales | Enterprise value / revenue | 이익 변동이 큰 cycle stock에서 PER보다 유용한 valuation tool |
| Reverse DCF | 현재 주가가 요구하는 미래 FCF를 역산 | 고multiple 성장주가 어느 정도 성공을 선반영했는지 보는 장치 |
Architecture decision tree
광통신 투자자는 제품명보다 architecture가 profit pool을 어디에 배분하는지 봐야 한다. 같은 AI capex라도 pluggable, SiPh external laser, CPO, LRO/AEC 조합에 따라 LITE가 가져가는 매출과 margin이 달라진다.
[Architecture] 이 표는 Architecture 항목을 기준으로 주요 수치, 판단 기준, 리스크 정보를 비교한다. 가격선, Green/Red, 매출·GM·FCF 임계값이 있는 행을 먼저 읽고 나머지는 근거 확인용으로 쓴다.
| Architecture | 어디서 쓰이나 | LITE 관련 부품 | 긍정 | 리스크 |
|---|---|---|---|---|
| Pluggable 800G | short/medium reach DR4/FR4 | 100G/200G EML, CW laser | 2026 revenue core | ASP erosion after launch premium |
| Pluggable 1.6T | 200G/lane 8x200G | 200G EML, packaging, test | 2026~2027 main rerating | volume shipment evidence required |
| UHP laser + SiPh | external light source for silicon photonics engine | UHP/CW laser | NVIDIA-linked architecture option | profit pool may shift to SiPh engine vendor |
| CPO | optics closer to switch ASIC | laser source, optical engine components | long-term TAM durability | standards and serviceability lag |
| Linear optical / LRO | power reduction by simplifying DSP path | laser quality and analog signal chain | content can expand or shift | CRDO/SMTC/MRVL capture risk |
| Active optical cable | rack/near-rack cable replacement | VCSEL/laser/module assembly | volume possible | lower ASP/margin |
| ROADM/EDFA transport | DCI/metro network scaling | transport components | cloud-to-cloud traffic growth | carrier capex cyclicality |
| Optical switching | cluster scale-out fabric | advanced photonics | future upside | still early adoption |
| Copper DAC/AEC | short reach low cost | indirect competitive substitute | sets lower-bound economics | distance/power threshold |
| Automotive/industrial sensing | non-datacenter optical demand | VCSEL/laser sensing | diversification | unit cycle slow |
100일 업데이트 체크리스트
이 보고서는 고정 결론이 아니라 evidence update surface다. 아래 100일 루프를 돌리면 Q4 FY26 전후로 Watch/Hold를 Buy 후보, 계속 Hold, 또는 thesis 축소로 바꿀 수 있다.
[시점] 이 표는 시점 항목을 기준으로 주요 수치, 판단 기준, 왜 필요한가 정보를 비교한다. 가격선, Green/Red, 매출·GM·FCF 임계값이 있는 행을 먼저 읽고 나머지는 근거 확인용으로 쓴다.
| 시점 | 체크 | 왜 필요한가 |
|---|---|---|
| Day 1 | 현재가가 $900/$820/$760/$1,120 중 어느 band에 있는지 표시 | 행동 매트릭스 자동 적용 |
| Day 7 | COHR/AAOI/MRVL/CRDO의 주간 상대수익률 확인 | sector rerating인지 LITE 단독 이벤트인지 분리 |
| Day 14 | LITE short interest와 borrow/put headline 확인 | squeeze와 de-risking 가능성 구분 |
| Day 21 | 1.6T/200G EML shipment wording 검색 | demo language와 revenue language 분리 |
| Day 30 | Apple supply-chain/3D sensing rumor를 official/rumor로 분리 | IT optionality만 조정, core CN thesis는 유지 |
| Day 37 | hyperscaler capex commentary: MSFT/META/GOOGL/AMZN | AI network capex proxy 업데이트 |
| Day 45 | inventory/capex discussion이 나오는 industry report 확인 | FCF conversion risk 조정 |
| Day 52 | Greensboro facility milestone: hiring/tool/install/customer | 2028 option만 조정 |
| Day 60 | Q4 earnings date와 implied move 확인 | event 전 position size 상한 |
| Day 67 | peer earnings call read-through 정리 | ASP pressure 문구 있으면 GM threshold 강화 |
| Day 75 | LITE Q4 result: revenue, GM, CN, capex, share count 입력 | rating 재계산 |
| Day 82 | post-earnings 3-day price reaction vs guide quality 비교 | 좋은 상승/나쁜 상승 판정 |
| Day 90 | FY27 revenue run-rate와 EV/Sales sensitivity 재계산 | base target 갱신 |
| Day 100 | review_note에 hit/miss 기록 | 다음 deep-dive revision 범위 확정 |
운영 원칙은 세 가지다. 첫째, 광통신 수요와 LITE 주가의 매력은 별도 변수로 기록한다. 둘째, 발표 자료의 architecture 단어가 실제 revenue, margin, customer diversity로 연결되는지 확인한다. 셋째, 현재가가 어느 행동 구간에 있는지 먼저 정한 뒤 뉴스를 해석한다. 이 순서를 지키면 강한 AI headline에 끌려가서 risk/reward가 나쁜 가격에서 비중을 키우는 실수를 줄일 수 있다. 다음 revision에서는 Q4 FY26 실제치와 peer call을 이 루프에 덮어쓴다.
Preflight compliance
Template / artifact gate evidence
- Mandatory template read: C:\Users\robotext\boardgame\docs\aisetup\codex\templates\deep_research_stock_TEMPLATE.md
- Readability rule read: C:\Users\robotext\boardgame\.claude\rules\deep-research-readability.md
- Mandatory exemplar inspected: C:\Users\robotext\.codex\docs\market\nintendo_deep_dive_20260521\index.html
- Optical peer exemplar inspected: C:\Users\robotext\.codex\docs\market\aaoi_deep_dive_20260522\index.html
- Artifact quality standard inspected: C:\Users\robotext\.codex\memories\artifact_quality_standard_20260518.md and artifact-quality README/reference_manifest.
- Output: 20 sections, 20 Chart.js canvases, bottom Evidence & Limitations, bottom preflight block.
- Chart.js option check: no explicit maintainAspectRatio false setting is used.
- Self-critique rounds saved in review_note.md. Source inventory saved in source_log.md.