Equity Research Report: Credo Technology Group Holding Ltd ($CRDO)
Current Price: $168.785 | Rating: BUY | Fair Value: $190 | 12-Mo Target: $230
One-Page Summary
Credo just delivered its best quarter ever — $479.0M revenue, +114.7% YoY — and the stock fell 20% in a day. The sell-off was not a demand problem: it was GAAP gross margin cracking from 68.2% to 64.5% and opex roughly doubling, both mechanically driven by the $1.3B DustPhotonics acquisition that closed May 28, 2026. At $168.79 the stock is the cheapest AI-connectivity pure-play on trailing P/E, forward P/E and EV/EBITDA, with zero debt and $764.3M of liquidity — but the chart is broken and the entry deserves staging, not chasing.
Top 5 Reasons
- Growth is intact and accelerating in absolute dollars. FQ1 FY27 revenue $479.0M (+114.7% YoY, +9.6% QoQ); FY2026 revenue $1,335.1M vs $436.8M (+205.7%). The September crash followed a beat, not a miss.
- Valuation has genuinely reset. 33.6x forward GAAP EPS (FY Apr-2027 consensus $5.02) and 20.4x FY Apr-2028 ($8.27) — versus Astera Labs at 125.3x forward and a 53.4x peer-set average.
- A real, enforceable moat. Credo created the Active Electrical Cable category and won patent-infringement settlements against TE Connectivity, Molex, Amphenol and Volex in March 2026 — a rare defensible IP position in semis.
- Fortress balance sheet. Zero long-term debt, $764.3M cash + short-term investments, $438.6M TTM free cash flow, 10.15x current ratio. The DustPhotonics deal was funded with cash and equity, not leverage.
- Estimate revisions are still going up. Over the trailing four weeks, 5 of 6 FY27 estimates and 5 of 5 FY28 estimates were revised upward — analysts cut price targets on multiple compression, not numbers on demand.
The offsetting concern: the top three hyperscaler customers are 34%, 27% and 16% of revenue (~77% combined). This is a concentrated bet on a handful of capex budgets.
1. Company Overview
Credo Technology Group Holding Ltd (NASDAQ-GS: CRDO) is a Technology-sector semiconductor company supplying high-speed connectivity for AI and hyperscale data centers: SerDes IP, retimer ICs, optical DSPs, and the Active Electrical Cables (AECs) it pioneered. Market cap $31.69B at $168.785. The company operates on an April/May fiscal year — FY2026 ended May 2, 2026, and FQ1 FY2027 ended August 1, 2026.
The defining corporate event of the past year is the DustPhotonics acquisition (agreed April 13, 2026; closed May 28, 2026): ~$750M cash plus $123M in stock upfront, with up to 3.21M contingent shares tied to milestones, for a total potential value near $1.3B. It vertically integrates Credo into silicon photonics for 800G/1.6T/3.2T optical and co-packaged optics. Two smaller tuck-ins preceded it: Hyperlume ($92M cash, announced September 2025, microLED optical interconnect) and CoMira Solutions (announced March 2, 2026, link-layer/ECC/security IP).
Data vintage note: Nasdaq's annual financial tables are stale, ending 5/2/2026 (~4 months old). All quarterly figures in this report come from the fresh table through 8/1/2026 and are cross-checked against the company's FQ1 FY27 release and SEC filings. One Scout-level correction was made in analysis: the year-over-year revenue comparator is the quarter ended ~8/2/2025 ($223.1M), giving +114.7% growth — not the +78.7% implied by mistakenly using the November 2025 quarter.
2. Price Chart
3. Technical Setup
Trend
Strong downtrend with accelerating weakness; a -20% earnings gap on 9/2 shattered intermediate support and price now sits 3.85% below the critical 200-day MA.
Wyckoff Phase
Secondary Test / Early Markdown Phase — The gap-down on 9/2 with 4.4x volume (30.1M shares) is classic capitulation selling. The stock is testing the 200-day MA ($175.34) as a potential Accumulation base or reversal point; if this breaks, institutional weakness likely extends to S1 ($87.81).
Key Levels
| Level | Price | Significance |
|---|---|---|
| R2 | $302.52 | All-time high; reversal would need to close -43.5% gap |
| R1 / 50-day SMA | $230.22 | Prior support, now overhead resistance |
| 200-day SMA | $175.34 | Critical support — price 3.85% below |
| Current | $168.785 | Vulnerable; must hold above S1 or accelerate lower |
| S1 | $87.81 | 200-day low; next major level if breakdown occurs |
| S2 | $7.35 | IPO floor (extreme scenario) |
Entry Strategy
- Bullish: Buy zone $160—$175 (at/below 200-day MA support), stop loss $150 (breakdown signal), target $185—$210
- Bearish: Short zone $170—$185 (fade bounce at 200-day MA), stop loss $190 (above 50-day), target $87.81
Ant's Verdict
WAIT FOR PULLBACK — The 200-day MA at $175.34 is the fulcrum; confirm a close above $180 and bounce toward $190 before risking long entries, or SHORT the failure below $170 targeting $120—$145 as institutional selling accelerates.
4. Financial Health
Business Quality
Moat width: Narrow, strengthening toward Wide in the AEC/optical-DSP niche.
- Patent-protected Active Electrical Cable (AEC) category: Credo pioneered AECs (copper cables with embedded DSP retimers) and won patent-infringement settlements against TE Connectivity, Molex, Amphenol, and Volex in March 2026 — a rare enforceable IP moat in semis.
- Vertical integration across the full connectivity stack: SerDes IP → Retimer ICs → DSP → now Silicon Photonics (via the DustPhotonics acquisition, closed May 28, 2026), letting Credo offer copper-to-optical AI-cluster interconnect that few competitors match end-to-end.
- Sticky hyperscaler design-ins: AI cluster interconnect requires long qualification cycles, creating switching costs — but this cuts both ways (see concentration risk below).
Financial Snapshot
| Metric | Value | Source | Assessment |
|---|---|---|---|
| Revenue TTM | $1,591.0M | Scout/Nasdaq (sum of FQ1-FQ4 trailing) | Exceptional — but decelerating off 205.7% FY26 base |
| Revenue Prior Q (FQ4 FY26, 5/2/26) | $437.0M | Scout | +9.6% QoQ into FQ1 FY27 ($479.0M) |
| Free Cash Flow (TTM) | $438.6M | Scout (OCF $500.4M − CapEx $61.8M) | Strong; FCF margin ~27.6% of TTM revenue |
| Cash on Hand | $466.9M cash + $297.4M ST investments = $764.3M | Scout (8/1/26 balance sheet) | Down $698M QoQ — funded DustPhotonics cash portion |
| EBITDA (TTM, derived) | ~$540M | Derived from EV ($30.9B, ex-cash) ÷ stockanalysis.com EV/EBITDA (55.44x) | Not directly disclosed — cross-checked estimate, treat as approximate |
| Profit Margin (TTM) | 33.8% | Scout/Nasdaq | High for semis, but net margin inflated by tax items — watch GAAP gross margin compression |
| Book Value | $2,728.5M total / $14.52 per share (187.95M shares) | Scout balance sheet ÷ stockanalysis share count | P/B ≈ 11.6x — rich, but supported by zero debt |
| EPS TTM | $2.87 | Scout | — |
| Debt/Equity | 0.00 (zero LT debt) | Scout | Pristine balance sheet; DustPhotonics funded via cash+equity, not debt |
| P/E (Trailing) | 58.8x | Scout; confirmed ~58.0x by stockanalysis.com | Consistent across sources |
| P/E (Forward, FY Apr-2027) | 33.6x (GAAP consensus EPS $5.02) | Scout/Zacks | Disagrees with stockanalysis.com's 22.49x. The gap is basis, not error: stockanalysis appears to use non-GAAP (SBC-adjusted) forward EPS consensus (~$7.50), which is materially higher than GAAP. Jeb trusts the GAAP-consensus 33.6x as the more conservative, comparable figure — both flagged for the client. |
Quarterly Trend (Nasdaq, $000s)
| Quarter Ending | 8/1/2026 | 5/2/2026 | 1/31/2026 | 11/1/2025 |
|---|---|---|---|---|
| Total Revenue | 479,003 | 437,003 | 407,012 | 268,027 |
| Gross Profit | 309,080 | 298,067 | 278,868 | 181,046 |
| R&D | 114,524 | 90,534 | 78,483 | 57,916 |
| SG&A | 73,856 | 51,688 | 50,763 | 44,334 |
| Operating Income | 120,700 | 155,845 | 149,622 | 78,796 |
| Net Income | 129,425 | 169,102 | 157,142 | 82,636 |
The FQ1 story in one line: revenue grew 9.6% sequentially while R&D grew 26.5% and SG&A grew 42.9% — the DustPhotonics cost base arrived before its revenue did.
Margin & Return History (Nasdaq annual ratios)
| Period Ending | 5/2/2026 | 5/3/2025 | 4/27/2024 | 4/29/2023 |
|---|---|---|---|---|
| Gross Margin | 68.04% | 64.77% | 61.89% | 57.65% |
| Operating Margin | 33.33% | 8.50% | -19.20% | -11.53% |
| Profit Margin | 35.37% | 11.95% | -14.70% | -8.98% |
| After-Tax ROE | 22.89% | 7.66% | -5.25% | -4.76% |
| Current Ratio | 10.15x | 6.62x | 11.88x | 10.58x |
Consensus EPS Forecast (Nasdaq/Zacks)
| Fiscal End | Consensus | High | Low | # Est | Revisions Up/Down (4wk) |
|---|---|---|---|---|---|
| Oct 2026 (FQ2) | $1.03 | $1.06 | $1.00 | 5 | 4 up / 0 down |
| Jan 2027 (FQ3) | $1.34 | $1.43 | $1.26 | 6 | 4 up / 0 down |
| Apr 2027 (FQ4) | $1.88 | $2.00 | $1.61 | 5 | 4 up / 0 down |
| FY Apr-2027 | $5.02 | $5.19 | $4.85 | 6 | 5 up / 1 down |
| FY Apr-2028 | $8.27 | $9.17 | $7.67 | 5 | 5 up / 0 down |
Valuation vs Peers
| Metric | CRDO | ALAB (Astera Labs) | MRVL (Marvell) | Sector Avg* |
|---|---|---|---|---|
| P/E (Trailing) | 58.8x | 283.5x | 79.2x | 117.6x |
| Forward P/E | 33.6x | 125.3x | 50.2x | 53.4x |
| P/B | 11.6x | 32.0x | 9.9x | 15.9x |
| EV/EBITDA | ~55.4x | 270.3x | 74.7x | 100.3x |
| Revenue Growth (latest qtr YoY) | +114.7% | +104.2% | +28.0% | n/m — mixed periods, see note |
| Profit Margin (TTM net) | 33.8% | 30.7% | 32.6% (derived) | 27.3% |
*Sector Avg computed across CRDO, ALAB, MRVL, AVGO (61.8x/23.6x/20.7x P/B/43.1x EV-EBITDA/38.85% margin), COHR (66.0x/28.9x/4.7x/41.2x/11.3%), MTSI (156.2x/58.7x/16.4x/117.3x/16.5%). Revenue-growth row mixes quarterly and annual reporting cadences across peers (AVGO's figure is a 3-yr forward consensus CAGR, not a trailing quarter actual) — directional only, not apples-to-apples.
Read: CRDO is actually the cheapest of the AI-connectivity pure-plays on trailing and forward P/E and EV/EBITDA (ALAB and MTSI both trade far richer), but still expensive on an absolute basis and against broad-market semis (AVGO, COHR).
5. Ownership & Capital Structure
Active/hedge-fund holders only (BlackRock, Vanguard, State Street, Geode Capital excluded as passive index managers):
| Holder | Shares | Est. Value | Type |
|---|---|---|---|
| Point72 Asset Management, L.P. | 5.6M (as of YE2025, most recent found) | ~$945M at current price | Hedge fund |
| Wellington Management Group LLP | 1.037M | ~$41.6M | Active asset manager |
| T. Rowe Price Associates | 1.028M | ~$41.3M | Active mutual fund |
| AllianceBernstein L.P. | Confirmed top-10 holder; exact share count not found this session | — | Active asset manager |
Notable: Lip-Bu Tan — now Intel's CEO (appointed March 2025) — is CRDO's single largest individual shareholder at 14.29M shares (7.75%) and reportedly still sits on Credo's board. This is an unusual governance overlap worth watching (Intel and Credo do not directly compete, but it's a related-party dynamic to flag).
Capital structure: Zero long-term debt. Total liabilities $284.2M against $2,728.5M equity (0.10x). Total assets $3,012.7M as of 8/1/2026, of which $986.4M is goodwill and $378.8M intangibles — i.e. 45% of book equity is now acquisition-related intangible value, a material change in balance-sheet quality from a quarter ago.
Recent Funding:
- Oct 2025: $750M follow-on equity offering + broad shelf registration (shares, debt, warrants) filed.
- May 28, 2026: Completed acquisition of DustPhotonics — upfront $750M cash +
0.92M CRDO shares, plus up to 3.21M additional contingent shares tied to milestones, total deal value up to $1.3B. This explains the FQ1 FY27 balance-sheet jump: goodwill +$893.6M, intangibles +$349.6M ($1.24B combined), cash down $698M. Strategic rationale: vertical integration into silicon photonics for 800G/1.6T/3.2T optical interconnect (NPO/CPO); combined optical portfolio targeted at >$500M revenue in FY2027, with company reiterating >$600M optical revenue guidance for FY27.
6. Competition & Sector
| Competitor | Ticker | Overlap |
|---|---|---|
| Marvell Technology | MRVL | Custom silicon, DSP/optical interconnect, direct hyperscaler competitor |
| Astera Labs | ALAB | PCIe/CXL retimers, closest pure-play "AI connectivity" comp |
| Broadcom | AVGO | Optical DSPs, custom ASICs, much larger scale |
| Coherent | COHR | Optical components/transceivers, silicon photonics |
| MACOM | MTSI | RF/optical semis, data-center interconnect |
Tailwinds:
- AI/hyperscaler data-center capex supercycle; Marvell guided Q2 FY27 revenue +35% YoY, corroborating sector-wide demand strength.
- Transition to 200G-per-lane SerDes and 800G/1.6T/3.2T optical connectivity plays directly to Credo's roadmap.
- Newly acquired silicon-photonics IP (DustPhotonics) opens a vertically-integrated optical stack competitors largely lack.
- Enforceable AEC patent portfolio (March 2026 litigation wins) protects a fast-growing, high-margin niche.
Headwinds:
- Extreme customer concentration: per FY2026 10-K, the top 3 hyperscaler customers represented 34%, 27%, and 16% of revenue in the last reported quarter (~77% combined); four-to-five hyperscalers each contribute >10%. Any single customer's capex pause or dual-sourcing decision is a large single-name risk.
- Gross margin compression: 68.2% (FQ4 FY26) → 64.5% (FQ1 FY27), driven by acquisition dilution/mix shift — flagged by multiple analysts as the key reason the Sep 2, 2026 earnings beat still triggered a ~20% stock decline.
- Valuation risk: 58.8x trailing P/E leaves no room for a hyperscaler capex digestion cycle.
- DustPhotonics integration risk — silicon photonics at CRDO's target scale (800G+) is not yet proven in high-volume production.
- High beta (~3.2 reported post-selloff) — the stock trades with extreme volatility around prints.
7. 12-Month Catalyst Timeline
| Date/Quarter | Catalyst | Type | Impact |
|---|---|---|---|
| ~Oct 2026 | FQ2 FY27 earnings (guided revenue $525–535M, non-GAAP net margin ~50%) | Earnings | High |
| Oct/Jan/Apr/Jul 2026-27 | Hyperscaler capex guidance (AMZN/GOOGL/META/MSFT) | Macro read-through | High |
| H2 2026 – 2027 | DustPhotonics integration milestones / contingent share triggers (up to 3.21M shares) | M&A | Medium |
| ~Jan 2027 | FQ3 FY27 earnings (consensus EPS $1.34) | Earnings | High |
| ~March 2027 | OFC 2027 (optical networking industry conference) — product/design-win visibility | Conference | Medium |
| ~May 2027 | FQ4 FY27 & full-year FY27 earnings (consensus EPS $1.88 Q4 / $5.02 FY) | Earnings | High |
| Ongoing | Gross-margin trajectory post-DustPhotonics (watch for stabilization vs. further compression) | Fundamental | High |
8. Recent News
| Date | Outlet | Headline | Material? |
|---|---|---|---|
| 2026-09-03 | The Motley Fool | "Why Credo Technology Sank This Week" — shares down 29.5% cumulative through Thursday despite beat | Yes |
| 2026-09-02 | The Motley Fool / Yahoo Finance | "Why Credo Technology Stock Plunged 20% Today" — margin compression, opex doubling, customer concentration cited | Yes |
| 2026-09-02 | MarketBeat / GuruFocus | JPMorgan cuts CRDO PT to $310 from $335 (Overweight maintained); Evercore ISI cuts to $292 from $325; Bank of America cuts to $275 from $340 | Yes |
| 2026-08-20 | 24/7 Wall St. | "Credo Is Up 71% This Year and Wall Street Just Raised Its Targets Again" — Barclays to $300 (from $260), Susquehanna to $250 (from $235), ahead of the print | Yes |
| 2026-08-04/06 | Company/Trade press | Credo showcased AI memory/storage interconnect products at FMS 2026 conference, Santa Clara | No |
| 2026-07-16 | The Motley Fool | COO sold 55,998 shares for $12.6 million (Form 4 disclosure) | Somewhat |
| 2026-07-01 (approx) | Company press release | Q4/FY2026 results: Q4 revenue $437.0M (+157% YoY), GAAP gross margin 68.2% | Yes |
| 2026-05-28 | Investor relations / DataCenterDynamics | Credo completes acquisition of DustPhotonics (silicon photonics PIC maker) | Yes |
| 2026-04-13 | DataCenterDynamics / Times of Israel | Credo agrees to acquire DustPhotonics for ~$750M cash + ~$123M stock, up to ~$1.3B with earnout milestones | Yes |
| 2026-03-02 | Company press release | Credo acquires CoMira Solutions (link-layer/error-correction/security IP) | No |
| 2025-09-29 | BusinessWire / BetaKit | Credo acquires Hyperlume for ~$92M cash (microLED optical interconnect tech) | No |
9. M&A & Rumor Watch
- THE ACQUISITION IS CONFIRMED AND SOURCED: DustPhotonics. Credo agreed on April 13, 2026 to acquire the Israeli silicon-photonics company DustPhotonics for approximately $750M cash plus
$123M in stock upfront, with an additional earnout of up to 3.21M shares ($430M at deal-time prices) tied to performance milestones — a total potential value of$1.3B. The deal closed May 28, 2026. This lines up exactly with the balance-sheet jump: goodwill $92.8M→$986.4M and intangibles $29.3M→$378.8M ($1.24B combined increase) in the quarter, and cash falling from $1.165B to $467M (funded largely with cash on hand). Rationale: DustPhotonics brings in-house Silicon Photonics PIC technology, making Credo a vertically integrated supplier across 800G/1.6T/3.2T optical and co-packaged optics — reducing reliance on third-party optical engine suppliers as it chases the $600M+ FY27 optical revenue target. Market reception at announcement was reportedly positive (one outlet: "CRDO Rises on its $750M DustPhotonics Deal"), but by the September print the added stock-based comp, opex, and integration costs tied to this deal were cited as contributors to the gross-margin/opex deterioration that spooked investors. - Two smaller tuck-ins preceded it: Hyperlume (announced Sept 29, 2025, ~$92M cash, microLED optical interconnect IP) and CoMira Solutions (announced March 2, 2026, link-layer/ECC/security semiconductor IP) — both smaller and not material to the goodwill jump on their own, but part of the same "buy vs. build" optical strategy narrative.
- Insider activity: CEO Bill Brennan and other executives (including the COO, who sold 55,998 shares for $12.6M on ~July 16, 2026) have been steady sellers under Rule 10b5-1 plans — 43 sales, 0 buys over 5 years per GuruFocus/insider trackers. This is routine executive diversification, not a new red flag, but worth noting there is no insider buying to signal confidence after the crash (no post-selloff insider purchases could be sourced).
- No activist investor filings or hostile-stake rumors were found. No credible acquisition-target rumors about CRDO itself (i.e., Credo being acquired) were found.
10. Social Sentiment
- Overall mood: Mixed-to-Bullish on retail platforms, but increasingly cautious among sell-side analysts. Retail (Stocktwits) swung to "Extremely Bullish" (~85/100) immediately after the crash, viewing it as a buying opportunity; broader Reddit/StockTwits aggregate trackers (AltIndex) show a more tempered ~58/100 "Neutral" reading over the trailing period.
- Retail vs Institutional: Retail traders on Stocktwits/Reddit are framing the sell-off as a valuation overreaction to a "beat and raise" quarter — sample quotes: "Nothing but great news... bright future ahead" and "Price action is ridiculous and will reverse strongly." Message volume spiked over 680% in 24 hours post-print. Institutions/sell-side are more sober: JPMorgan, Evercore ISI, and Bank of America all cut price targets (to $310, $292, and $275 respectively) even while several kept positive ratings, explicitly citing gross-margin compression (68.2%→64.5% GAAP), opex doubling ($89.6M→$188.4M GAAP), rising stock-based comp, and customer concentration (five hyperscalers, top three at 34%/27%/16% of revenue) as the reasons for trimming numbers rather than a loss of conviction in the growth story.
- Trending narratives: (1) "Great quarter, ugly optics" — the beat-and-raise vs. margin-compression tension dominates coverage; (2) hyperscaler concentration risk (Amazon, Microsoft, plus 3 other unnamed hyperscalers) as the structural bear case; (3) the DustPhotonics-driven vertical integration story as the bull case for why margins may recover once integration costs normalize and optical DSP/silicon-photonics revenue scales past $600M in FY27.
- Notable voices: JPMorgan analyst Joseph Cardoso (Overweight maintained, PT cut to $310) — bullish long-term but flags near-term margin risk; Bank of America (PT cut to $275 from $340) — more cautious tone on opex trajectory; Motley Fool contributor commentary framed the pullback as a potential buying opportunity if triple-digit growth persists, while flagging that on an SBC-adjusted basis the "true" earnings multiple is closer to 50x than the headline 26x, i.e., not obviously cheap.
- Sentiment shift: Sharp divergence emerged in the past 30 days: pre-print (mid-August) sentiment was strongly bullish with sell-side hiking targets (Barclays to $300, Susquehanna to $250) ahead of what was expected to be another clean beat; post-print (since Sept 2) sell-side consensus targets have been cut roughly 8-19%, while retail sentiment actually intensified bullish, creating a retail-vs-institutional gap. Driver: the market re-rated the stock from a "flawless hypergrowth" multiple to one that discounts margin risk and customer concentration, even though no revenue thesis was broken.
- Sentiment score: 3.5/5
Pete's Read
The fundamentals were never the problem — revenue +114.7% YoY, a genuine capability-broadening acquisition (DustPhotonics, ~$1.3B all-in) that vertically integrates Credo into silicon photonics, and guidance that still calls for 85%+ FY27 growth. What news flow and sentiment reveal that the income statement alone doesn't is a market that had priced CRDO for perfection: the moment GAAP gross margin cracked below 65% and opex growth guidance was revised up (to ~55% from ~50%), sell-side desks moved in lockstep to cut targets 8-19% even while retaining bullish ratings, while retail piled in on the dip, creating an unusual bull/bear split by cohort rather than by fact pattern. The DustPhotonics deal is being underappreciated in the panic — it's the actual explanation for the goodwill/intangibles/cash swing on the balance sheet, and its integration costs (stock comp, opex) are the mechanical drivers of the margin miss, not a demand problem. Watch for two things next: whether Q2 (guided $525-535M revenue, 67-69% non-GAAP gross margin) confirms margins stabilizing post-deal-integration, and whether the hyperscaler concentration (5 customers, top 3 at 34%/27%/16%) shows any signs of order lumpiness — that's the real tail risk sentiment keeps flagging that the "record revenue" headline obscures.
11. Bull vs Bear Case
| Bull Case | Bear Case | |
|---|---|---|
| Growth | FQ1 revenue +114.7% YoY, +9.6% QoQ; FY27 guided to 85%+ growth; FQ2 guided $525–535M | Growth is decelerating off a 205.7% FY26 base; a single hyperscaler pause resets the trajectory |
| Margins | 64.5% GM reflects one-time DustPhotonics integration; FQ2 guided 67–69% non-GAAP — recovery already signposted | GM fell 370bp in one quarter; opex doubled to $188.4M GAAP and guidance for opex growth was raised to ~55% from ~50% |
| Valuation | 33.6x forward GAAP EPS, 20.4x FY28 — cheapest AI-connectivity pure-play vs ALAB at 125.3x forward | 58.8x trailing, 11.6x book; SBC-adjusted "true" multiple nearer 50x — no margin for error |
| Balance sheet | Zero debt, $764.3M liquid, $438.6M TTM FCF, 10.15x current ratio | Cash halved in one quarter; 45% of book equity is now goodwill + intangibles |
| Moat | AEC category creator with March 2026 patent wins vs TE, Molex, Amphenol, Volex; now vertically integrated into silicon photonics | Silicon photonics at 800G+ volume is unproven for Credo; MRVL, AVGO and ALAB all attack the same sockets |
| Customers | Multi-year hyperscaler design-ins with long qualification cycles create switching costs | Top 3 customers = ~77% of revenue (34/27/16); concentration is the defining structural risk |
| Sentiment/Positioning | 5 of 6 FY27 and 5 of 5 FY28 estimates revised UP in the past 4 weeks; consensus target $287.50 implies +70% | Every major desk cut its target 8–19% on Sep 2; price is below both the 50-day ($230.22) and 200-day ($175.34) MA |
| Insiders | Point72 holds ~5.6M shares; Lip-Bu Tan holds 7.75% | 43 insider sales, 0 buys over 5 years; no post-crash insider buying |
12. The Verdict
| Item | Call |
|---|---|
| Rating | BUY (staged entry — do not chase) |
| Current Price | $168.785 |
| Fair Value | $190 (+12.6%) |
| 12-Month Price Target | $230 (+36.3%) |
| Street Consensus Target | $287.50 (+70.3%) — we are deliberately below the Street |
| Entry Zone | $160 – $175 (at/below the 200-day MA at $175.34) |
| Confirmation Signal | Daily close above $180 |
| Stop Loss | $150 (a close below invalidates the 200-day-MA base thesis) |
| Upside Trigger | FQ2 print (~Oct 2026) confirming 67–69% non-GAAP gross margin and $525–535M revenue |
| Conviction | Medium |
| Time Horizon | 12–18 months |
| Key Risk | Customer concentration — top 3 hyperscalers ≈ 77% of quarterly revenue; a single capex pause or in-house ASIC shift impairs revenue faster than the diversification narrative can offset |
Why BUY and not HOLD, given a broken chart: the September de-rating was a multiple event, not an earnings event. Revenue beat, forward estimates went up (5 of 6 revisions higher for FY27, 5 of 5 for FY28), and the margin compression has an identifiable, dated, non-recurring cause in the DustPhotonics close. At 33.6x forward GAAP earnings with zero debt and $438.6M of trailing free cash flow, CRDO now trades at a discount to every AI-connectivity pure-play comp. Why staged and not aggressive: price sits 3.85% below the 200-day MA on 4.4x volume, the technical structure is in early markdown, and management needs one clean quarter to prove margins normalize. Buy into the $160–175 zone, add on a close above $180, and cut below $150.
Fair value derivation: $190 ≈ 37.8x FY Apr-2027 consensus GAAP EPS of $5.02 — a discount to the 53.4x peer forward average to reflect margin compression and concentration risk, but a premium to Broadcom/Coherent given the growth differential. The $230 target ≈ 28x FY Apr-2028 consensus EPS of $8.27.
Data caveats: Nasdaq's annual financial tables end 5/2/2026 and are ~4 months stale; all current-quarter figures are drawn from the fresh quarterly table through 8/1/2026 and cross-checked against the FQ1 FY27 release and SEC filings. Forward P/E is quoted on a GAAP basis (33.6x); widely-cited non-GAAP forward multiples run materially lower (~22x) — the difference is stock-based compensation, not a data error. The $287.50 consensus target could not be independently confirmed as a post-September-cut figure and may be a stale pre-cut blend. Point72's 5.6M-share position is as of year-end 2025, the most recent 13-F located.
Disclaimer: For informational purposes only. Not investment advice.