Viavi Solutions Inc. (NASDAQ: VIAV) — A Banknote-Pigment Annuity Bolted to a Cyclical Test House, Repriced as a Pure-Play AI-Optics Bet

Independent fundamental research — evidence-driven and skeptical. Report date: 2026-06-21. Fiscal year ends ~June 30.

⚡ Claude’s Take

This block is the author’s own subjective opinion and general information only — not investment advice. The analysis that follows takes no position, names no price target, and carries no buy/sell.

Verdict: AVOID at ~$47 / accumulate only on a deep washout into the low-to-mid $20s / NOT a short. Conviction: medium. Fair-value zone ~$26–$36 (≈ a sum-of-the-parts of a 14–18× franchise multiple on the OSP pigment annuity plus a 2.5–4× cyclical-T&M sales multiple on NSE, net of ~$0.58B net debt and the convert overhang). At $47 the equity is at its richest valuation in its entire post-2015 history (composite 98.6th own-history percentile; ~8.5× TTM sales; ~68× TTM EBITDA), and even a genuine bull case — revenue +50% to ~$2.4B by FY29 and operating margin doubling back to the FY22 all-time peak of ~20%, exited at a premium 20× EBITDA — barely matches today’s enterprise value. You are not being paid to underwrite that.

The framing is a crowded one-way momentum trade in a structurally mixed business. Two-thirds of revenue (NSE — network/service test) is a contested, cyclical, sub-WACC-through-cycle oligopoly that the market round-tripped to $6.66 just twenty-four months ago; it is now riding a real but extrapolation-prone AI-datacenter-optical capex wave that the tape is pricing as permanent. The genuinely good business — OSP, the near-monopoly anti-counterfeit currency-pigment franchise (~36% operating margins, ~73% of segment profit on 28% of revenue) — is the smallest part and grows only mid-single-digits. The factor model confirms VIAV now trades as an AI-optics basket member (peers LITE/CIEN/GLW/FN), not a test-and-measurement name: Momentum loading +0.73, Value −0.52, +405% trailing year, y1 Sharpe 6.4. Management and the board agree it is fully valued — they sold $500M of stock into the all-time high at ~$45 and insiders have logged ~21 sells and zero open-market buys. Why not short it? Because the operating reacceleration is real (Q3 FY26 +42.8%), the converts/short interest make squeezes violent, and VIAV is a plausible takeout — momentum + improving fundamentals + strategic optionality is a dangerous short. Tag: “the best business in the house is the smallest, and the whole house is on sale at full price.” Flip bullish: two-plus more quarters of 30%+ growth with NSE operating margin durably crossing 15%. Flip bearish (more so): a single sub-12% NSE-margin quarter or visible AI-capex digestion.

📈 Stock Price Action — Five-Year Event Map

VIAV has round-tripped and then some: from ~$17.5 in mid-2021, down a brutal ~62% to a $6.66 closing trough (26-Jun-2024) as the telecom/service-provider capex cycle collapsed, then a ~7–8× moonshot to a $55.33 closing all-time high (1-May-2026) ($60.43 intraday on 30-Apr) on the AI-datacenter optical-test narrative. It now sits at $47.17 (18-Jun-2026), ~15% below the closing high (~22% below the intraday high), with a 52-week range of $6.66–$55.33. The explosive leg was compressed into ~5 months (Dec-2025 → May-2026).

Cycle narrative. (1–2) VIAV is a textbook capital-cycle casualty: the post-COVID, 5G-fiber capex surge (FY22 record $1,292M revenue) inverted into a savage two-year service-provider spending winter (FY24 $1,000M, GAAP loss), and the stock lost roughly three-quarters of its value into June-2024. (3) The recovery began on stabilization plus the announced Spirent high-speed-Ethernet carve-out and the first signs that hyperscaler/AI-datacenter buildout would pull a wholly different demand stream through the NSE lab/production-test business. (4–6) Three consecutive blow-out prints (Oct-25, Jan-26, Apr-26), each with a violent single-day gap up, compounded into a 4–5× re-rate as the Street re-categorized VIAV from “cyclical telecom test” to “AI-optics.” (7–8) The melt-up topped on 1-May; the subsequent ~15% fade is well explained by a $500M primary equity offering priced near $45 (dilutive, repaying acquisition debt) and a mechanical 18-Jun session (49M shares, ~8× normal volume, coincident with the 19-Jun quarterly index reconstitution / quad-witch) with no corresponding 8-K — i.e., flow, not fundamentals. Price moves are facts; the attributed drivers are interpretation.

1. Executive Summary

Viavi Solutions is two very different businesses under one ticker. Network & Service Enablement (NSE) — test, monitoring and assurance instruments and software for communications service providers, network-equipment makers, hyperscalers, government and avionics customers — is ~72% of revenue ($776.6M FY25) but earns a thin, cyclical operating margin (5.4% FY25; 1.1% at the FY24 trough). Optical Security & Performance Products (OSP) — optically variable pigments and magnetic materials that secure the world’s banknotes, plus optical filters/coatings for 3D-sensing and anti-counterfeiting — is only ~28% of revenue ($307.7M FY25) but throws off a ~36.5% operating margin and ~73% of total segment operating profit. OSP is the genuine franchise; NSE is the cyclical engine that is, for now, levered to the AI-datacenter optical buildout.

The investable event is a sharp FY26 reacceleration: revenue grew from $299.1M in Q1 to $406.8M in Q3 (+42.8% YoY), NSE +54% in Q3, on AI/hyperscaler demand for 800G/1.6T optical lab and production test plus two acquisitions (Spirent’s high-speed-Ethernet assets, bought from Keysight; and Inertial Labs in positioning/navigation/defense). Roughly half the YoY growth is acquired; organic NSE is still a strong ~+28–30%. Reported GAAP remains barely profitable-to-negative (TTM GAAP EPS −$0.25), but this is a non-operating artifact — a $46.2M loss on a convertible-debt exchange, a $24.3M earnout mark-up (the Inertial deal beating plan), and a non-economic $36.1M tax provision booked against a pretax loss. The clean non-GAAP run-rate is ~$0.27/quarter (~$1.10–1.20 annualized) at ~21% operating margin, with attractive 40–45% incremental margins. Quality genuinely improves with scale.

The problem is price. At ~$47 the equity (~234M shares, ~$11.0B market cap; ~$11.6B EV including ~$1.08B debt and the convert structure, less ~$0.50B cash) trades at ~8.5× TTM sales and ~68× TTM EBITDA — its richest multiple ever and above the EV/sales of higher-margin, elite-ROIC peer Keysight. A sum-of-the-parts (OSP as a franchise annuity; NSE at a cyclical-test multiple) clears in the ~$18–25/share range; a base-case three-year DCF/scenario clears near ~$26; only an aggressive bull case justifies today’s price. Insiders are unanimous net sellers and the company itself sold $500M of stock into the high. This is a structurally better business than it was two years ago — but it is priced as if the AI-optics capex wave is both permanent and margin-transforming, in a name that demonstrably round-trips with the telecom cycle. The analysis below takes no position and names no price target; it stands on the evidence.

2. Business Overview

What VIAV does. Viavi (formerly JDS Uniphase / JDSU, renamed in 2015 after spinning out Lumentum) sells (a) test, monitoring and assurance hardware and software for communications networks, and (b) specialty optical materials — most importantly the pigments that authenticate paper currency. After a March-2025 reorganization the company reports two segments:

• Network & Service Enablement (NSE) — $776.6M FY25 (71.6% of revenue). Instruments, software and services to design, build, activate, certify, troubleshoot and optimize fiber, wireless, datacenter and access networks; field test handhelds; lab and production test for high-speed optical (400G/800G/1.6T) and Ethernet; network/service assurance and monitoring software; and test-and-measurement instrumentation for aviation, aerospace, government and defense. NSE folds together the legacy Network Enablement (NE) instrument business and the smaller Service Enablement (SE) assurance-software business. Customers: communications service providers (CSPs), network-equipment manufacturers (NEMs), hyperscalers/cloud operators, governments, avionics/defense. NSE gross margin is high (~63–65%) but operating margin is thin and volatile (5.4% FY25; ~1% FY24 trough) because of heavy R&D and SG&A spread over a cyclical revenue base.

• Optical Security and Performance Products (OSP) — $307.7M FY25 (28.4% of revenue), ~36.5% operating margin. Two pieces: (1) anti-counterfeiting — optically variable pigments (OVP) and optically variable magnetic pigments (OVMP) that produce the color-shifting security features on banknotes in >100 countries, sold primarily through a single ink-formulation partner, SICPA; and (2) performance products — thin-film optical coatings and filters for 3D-sensing (consumer devices), anti-counterfeiting documents, automotive, and government/defense. OSP is the profit engine: ~28% of revenue, ~73% of segment operating profit, with margins that have held ~36% for 3+ years on comparatively little reinvestment.

Revenue model & recurring mix. The bulk of revenue is product (instruments, pigments, coatings) — transactional and, for NSE, cyclically tied to customer capex budgets. Software/assurance, support, calibration and professional services add a recurring layer, but VIAV is not a predominantly-recurring software business; it is a hardware/materials company with a services tail. OSP’s banknote pigment is the most annuity-like revenue: central-bank denominations are re-designed and re-issued on multi-year cycles, and once a country’s currency specifies a Viavi/SICPA security feature, the relationship persists for years.

Customer concentration flag. OSP’s banknote pigments flow through SICPA, which represented ~$166.7M, or ~15.4% of total FY25 revenue (FACT, FY25 10-K segment/customer disclosure). That is a meaningful single-customer dependency, mitigated by the entrenched, certified, hard-to-displace nature of currency-security supply (a switch would require re-qualifying a new pigment chemistry across many sovereign mints).

Segment economics at a glance (FY25). The disclosure consolidation masks how lopsided the business is. On 28.4% of revenue, OSP produces roughly three-quarters of segment operating profit; NSE, with 71.6% of revenue, produces about one-quarter:

(Segment operating income sums above the unallocated corporate line; consolidated GAAP operating income was $58.2M after corporate/amortization.) The asymmetry is the single most important fact about VIAV: a sophisticated investor should value the two streams separately, because they have different growth rates, different cyclicality, different moats and — crucially — different deserved multiples.

Verdict. A two-engine business: a small, high-quality, capital-cycle-insulated franchise (OSP) cross-subsidizing a large, cyclical, capital-intensive, lower-return test business (NSE) that is currently enjoying an AI-datacenter demand windfall. The reporting consolidation masks how lopsided the economics are.

3. Industry Dynamics

Network test & measurement (NSE). A mature, cyclical, consolidating oligopoly. The end-market is communications-network capex — CSPs (telecoms, cable), NEMs, and increasingly hyperscalers building AI datacenters. The cycle is brutal on the downswing: when service providers froze spending in FY22–FY24, VIAV’s revenue fell ~$292M (−23%) peak-to-trough and operating margin collapsed from 14.3% to 3.4%. Profit pools are split among a handful of players — Keysight (KEYS) (much larger, ~$5B+ revenue, and now owner of the bulk of Spirent following its 2024–25 acquisition), EXFO (taken private), Anritsu, Rohde & Schwarz (private), and NetScout in assurance. It is a commoditizing field where differentiation is in software, breadth, and installed-base relationships rather than fundamental technology lock-in. Structural verdict: mediocre — cyclical, capital-hungry, fragmented enough to keep through-cycle returns near or below cost of capital.

The new demand stream — AI-datacenter optical test. This is the genuine change. Hyperscaler AI buildouts require validation and production test of 800G → 1.6T (→ 3.2T) optical interconnects, co-packaged optics, and high-speed Ethernet fabrics, plus fiber monitoring inside and between datacenters. This pulls a different, faster-growing, less-telecom-cyclical demand curve through NSE’s lab/production-test and high-speed-Ethernet (Spirent-acquired) lines — now estimated at roughly high-40s% of NSE revenue and growing fast. In Marathon capital-cycle terms, capital is flooding into this end-market (every optics and test vendor is chasing AI datacenter), which is precisely the condition that historically precedes margin mean-reversion — the demand is real, but so is the eventual supply response and the cyclicality of hyperscaler capex.

Banknote security & specialty optics (OSP). A tiny, opaque, structurally attractive niche. Currency anti-counterfeiting is a high-barrier, certification-gated, slow-moving business: central banks change security features rarely and deliberately, the qualification process is long, and the incumbent pigment chemistry is deeply embedded across sovereign mints. It is essentially insulated from the technology and capital cycle that whipsaws NSE. 3D-sensing/coatings is more competitive and consumer-cyclical but smaller. Structural verdict: good — narrow, defensible, annuity-like.

Overall industry verdict: MIXED. OSP sits in a genuinely good niche; NSE sits in a structurally average-to-poor one currently flattered by an AI-capex up-cycle. The blended business is better than NSE alone but should not be valued as if the entire enterprise shared OSP’s economics or the AI cycle’s growth rate.

4. Competitive Position

OSP — a genuine, durable moat (the crown jewel). In Greenwald’s taxonomy this is a combination of intangibles (proprietary pigment chemistry, decades of trust), scale economies in a tiny niche, and customer captivity via certification. The financial proof is unambiguous: ~36.5% operating margins sustained for years on modest R&D, in a business where the customer (a sovereign central bank, via SICPA) cannot casually switch suppliers without re-qualifying currency-grade security across an entire denomination set. Counterfeiting-resistance is a “must-not-fail” specification; incumbency is worth far more than a price concession. This is the cleanest moat in the company and would clearly deteriorate financially (margins would compress toward commodity coatings) if the franchise eroded — it passes the “moat must tie to a financial outcome” test. Verdict: durable advantage. The risk is not competition but slow growth and the single-partner (SICPA) channel.

NSE — a narrow, contested moat. VIAV has real assets here: the legacy JDSU optical heritage, a large installed base of field instruments, brand recognition with CSPs, and now Spirent’s high-speed-Ethernet software/hardware. But it competes against Keysight — several times its size and, post-Spirent, the scale leader — plus Anritsu, Rohde & Schwarz, EXFO and NetScout. The tell that the moat is thin is the through-cycle return profile: NSE earned ~1% operating margin at the FY24 trough and sub-WACC returns across the cycle; a true moat would have defended profitability in the downturn. Its differentiation in AI-datacenter optical test rests partly on bought capability (Spirent) and on optical heritage rather than on switching costs customers cannot escape. Verdict: crowded oligopoly with modest, partly-acquired differentiation — not a durable standalone moat.

Direct comparison. Versus Keysight (the obvious peer and prior report subject), VIAV is smaller, lower-margin, lower-ROIC, more cyclical, and now more expensive on EV/sales — an inversion of the quality-vs-price ordering that should make a quality-focused investor cautious. Versus the OSP “comp set,” there effectively isn’t a public pure-play; the franchise is rare, which is part of why it is undervalued within VIAV’s consolidated multiple while the whole trades rich.

Verdict. One genuine moat (OSP) that is small and slow; one contested position (NSE) that is large and cyclical and currently the entire growth story. The market is paying a franchise-and-growth multiple for a business whose franchise and growth live in different segments.

5. Growth History and Forward Opportunities

History. Revenue: FY20 $1,136M → FY21 $1,199M → FY22 $1,292M (peak) → FY23 $1,106M → FY24 $1,000M (trough) → FY25 $1,084M. The arc is the telecom capex cycle: COVID/5G-fiber surge, then a two-year service-provider spending freeze. OSP was the stabilizer through the bust; NSE drove both the boom and the collapse.

The FY26 reacceleration. Quarterly revenue: Q1 $299.1M → Q2 $369.3M → Q3 $406.8M, with Q3 +42.8% YoY and NSE +54% YoY. Nine-month FY26 revenue ~$1,075M vs ~$794M prior-year (+35.5%). This is the sharpest growth in VIAV’s history as a standalone — but its composition matters:

• ~Half is acquired. Spirent’s high-speed-Ethernet assets (bought from Keysight; ~$200M annualized, ~$54M/quarter) and Inertial Labs (positioning/navigation/timing for defense; ~$23M/quarter) together add ~$77M/quarter that wasn’t in the base a year ago.
• Organic NSE is still strong — roughly +28–30% — and genuinely AI-datacenter-led, with attractive 40–45% incremental operating margins and near-zero US cash tax (large NOLs from the JDSU era).

Forward opportunities. (1) 800G → 1.6T → 3.2T optical and co-packaged-optics test for AI datacenters — the headline driver; (2) high-speed Ethernet/data-center fabric test (Spirent); (3) fiber/5G/O-RAN field and assurance as the telecom cycle eventually normalizes off a ~45%-down wireless/core trough; (4) military/avionics and PNT (Inertial Labs); (5) OSP design wins in new banknote series and 3D-sensing/anti-counterfeiting. The optionality is real and multi-pronged.

Quality caveats. Visibility is short (1–3 quarters; book-to-bill driven by hyperscaler order timing); AI-datacenter capex is itself cyclical and concentrated in a few buyers; the telecom/wireless core remains depressed (so part of the “growth” is also a future cyclical-recovery call, not pure secular); and the highest-quality segment (OSP) grows only mid-single-digits, so the mix is shifting toward the lower-quality, more-cyclical engine. Verdict: high-rate but mixed-quality, extrapolation-prone growth — strong and partly secular, but partly acquired and partly cyclical, and emphatically not OSP-quality across the whole.

6. Financial Quality

Margins and the operating trajectory. Gross margin is stable and healthy at ~57–60% (blended; NSE ~63–65%, OSP higher still). Operating margin is the cyclical variable: FY22 14.3% peak → FY24 3.4% trough → FY25 5.4% → Q3 FY26 10.3% GAAP / ~21% non-GAAP. EBITDA: FY22 $260.3M (20.1% margin) → FY24 $93.1M → FY25 $120.9M, and rising sharply in FY26 on the revenue ramp. Incremental operating margins of 40–45% on NSE mean the model has real operating leverage if the revenue holds.

Quality of earnings — the central question, resolved. GAAP looks ugly (nine-month FY26 net loss −$63.1M on −$27.2M pretax; TTM GAAP EPS −$0.25), but the gap to economic earnings is fully explained by non-operating and non-economic items:

Normalizing, the clean run-rate is ~$0.27 non-GAAP EPS/quarter (~$1.10–1.20 annualized) at ~21% operating margin. GAAP understates earnings power here. Verdict on QoE direction: GAAP-flattering-to-the-bears, real economics better.

But the bull QoE case has a real catch — cash and dilution. (1) Stock-based comp is ~$53M/year (≈ GAAP net income) — a large wedge between non-GAAP and owner economics; net of SBC, “owner FCF” is thin. (2) Operating cash flow is lumpy and was a use of ~−$26M in Q3 FY26 (earnout payment + working capital build on the revenue ramp). FY25 OCF was $89.8M against $121.6M of acquisition spend. (3) Dilution is structural: the convertible notes are deep in-the-money, pushing diluted share count to 249.5M in Q3 FY26 (from ~223M) with guidance to ~256M — a ~22.7M-share overhang baked in.

Six-year financial summary (the cycle in numbers). The volatility is the story — a peak-to-trough revenue swing of ~$292M and an operating margin that quartered, then a violent FY26 recovery:

*FY26E annualizes the Q1–Q3 run-rate; not a forecast. The takeaway: VIAV has never sustained the ~20% operating margin the bull case requires except at the FY22 cyclical peak, and SBC (~$50–53M/yr) has exceeded GAAP net income in most years — a persistent wedge between “adjusted” earnings and owner economics.

Returns and balance sheet. Reported ROIC/ROE are distorted by negative tangible common equity (goodwill $701.8M + intangibles $398M exceed total equity of $846.5M) and by the legacy −$69.7B accumulated deficit (a JDSU-era write-off artifact, not operating). On an ex-goodwill, normalized-EBIT basis NSE earns sub-to-low-teens returns and OSP earns very high returns; the blend is unimpressive and below where the multiple implies. Liquidity is adequate: ~$0.5B cash, ~$1.08B gross debt, net debt ~$0.58B, current ratio ~1.6×. Verdict: economics improve with scale (good incrementals), but owner-FCF is thin after SBC and the capital structure dilutes — quality is improving, not yet proven across a full cycle.

7. Capital Allocation

M&A-led strategy. Under CEO Oleg Khaykin, capital allocation has pivoted decisively to acquisitions: Inertial Labs (FY25; ~$121.6M cash paid, total consideration including contingent earnout materially higher; PNT/defense; described by management as among the best-performing deals of Khaykin’s tenure, now beating plan — hence the earnout mark-up) and the Spirent high-speed-Ethernet / lab assets (~$400M cash, carved out of the Keysight–Spirent transaction, closed ~Sep-2025, ~$200M annualized revenue), funded by a ~$600M Term Loan B. These are strategically coherent — they buy directly into the AI-datacenter and defense growth vectors — and Inertial in particular looks like a good deal so far. The Spirent price (~2× sales for high-growth datacenter test) is defensible if the AI cycle persists, rich if it doesn’t.

Financing and the dilution scorecard. VIAV then de-levered opportunistically by selling $500M of equity at ~$45 — essentially the all-time high (11.1M shares, ~May-2026) — to repay the Term Loan B. Issuing stock at a 98th-percentile valuation to retire debt is, viewed coldly, good capital-markets timing for the company (and a tell about how management views the price). It is dilutive to holders but accretive to the balance sheet. Combined with the convert overhang and ongoing SBC, net share count is rising — VIAV did buy back ~$236M of stock in FY22 near the prior peak, but buybacks are now dormant and there is no dividend.

Capital structure. ~$1.08B gross debt: 0.625% convertible notes due 2031 (~$648M, ~$13.79 conversion price — now deep in-the-money), 3.75% notes due 2029 (~$371M), plus the remaining term debt. The low-coupon converts were cheap money but the equity-dilution cost is now large given the share-price move.

Incentives — a governance demerit. The proxy compensation plan keys on revenue, non-GAAP operating income, and relative TSR versus the Nasdaq Telecom index — with no ROIC or capital-efficiency metric. For a company whose whole value question is whether the returns on a lower-quality, M&A-built revenue base justify the multiple, the absence of a capital-return governor is a real weakness (consistent with a pattern flagged across this coverage universe). Insider behavior reinforces caution: over the trailing period there were ~21 insider sales and zero open-market purchases; CEO Khaykin sold $22M+ in early May at $51–55 under a 10b5-1 plan. Routine and planned, but unanimous distribution into strength.

Verdict: mixed, leaning opportunistic-but-not-shareholder-friendly. Smart deal-buying and superb equity-issuance timing, undercut by structural dilution, a comp plan with no ROIC discipline, and insiders uniformly selling.

8. Changes and Headwinds — Last Two Years

• Segment reorganization (Mar-2025): collapsed NE + SE into NSE alongside OSP — reducing disclosure granularity right as the AI-datacenter mix became the key question.
• Two major acquisitions: Inertial Labs (defense/PNT) and Spirent’s high-speed-Ethernet assets (datacenter test) — transforming NSE’s exposure and adding ~$77M/quarter of acquired revenue, ~$600M of term debt (since repaid via equity), and integration risk.
• The cyclical inflection itself: from the FY22–24 telecom-capex bust (revenue −23%, GAAP loss, stock −75% to $6.66) to the FY26 AI-datacenter boom (Q3 +42.8%) — the single biggest change, and the entire basis of the re-rate.
• Convertible-debt exchange (Aug-2025): rolled 2026 notes into the 0.625% 2031 converts, generating the $46.2M GAAP extinguishment loss; cheap financing but now a large dilution source.
• $500M equity raise (May-2026) into the all-time high to repay the Term Loan B.
• Leadership continuity: Oleg Khaykin (President & CEO) and Ilan Daskal (CFO) in place; no destabilizing C-suite change, but no obvious succession either.
• Headwinds: AI-capex cyclicality and customer concentration among a few hyperscalers; a still-depressed telecom/wireless core; SICPA single-channel dependency in OSP; a deep-ITM convert overhang; and the simple fact of a 98th-percentile valuation that prices in continuation.

Verdict: the changes have genuinely improved the business (better mix, faster growth, AI exposure, repaired balance sheet) — but they have also increased acquisition/integration and dilution risk, and the valuation has more than fully captured the improvement.

9. Risk Analysis (Risk Matrix)

The dominant risks are valuation and cyclicality, not solvency. A total loss is unlikely; a 40–60% drawdown on a demand/margin disappointment from a 98th-percentile multiple is very plausible.

10. Valuation Discussion (Embedded Expectations)

Where it trades. ~234M shares × $47.17 ≈ $11.0B market cap; adding ~$1.08B debt and the convert structure and netting ~$0.50B cash gives ~$11.6B EV. That is ~8.5× TTM sales (~$1.37B), ~68× TTM EBITDA (~$121M trailing, though run-rating higher), and ~150×+ TTM EBIT — and on AZI’s own-history percentile lens, composite 98.6th, P/S 98.4th, P/B 98.7th: the richest VIAV has ever been. The P/E is null only because GAAP TTM EPS is negative for the non-operating reasons discussed under Financial Quality; on the clean ~$1.10–1.20 non-GAAP run-rate the stock is ~40× forward non-GAAP earnings — also rich for a hardware/materials business. Notably, VIAV’s EV/sales now exceeds Keysight’s, despite KEYS being larger, higher-margin and higher-ROIC — an inversion of quality-vs-price.

Reverse-engineering the price (what must be true). To merely hold ~$11.6B EV at a sane exit multiple, the market is underwriting (a) the AI-datacenter optical-test demand as durable and secularly growing, (b) NSE operating margin expanding from ~5% toward the mid-teens-plus and holding there, and © the whole enterprise compounding at a rate that re-rates an average-quality test business to franchise economics. The scenario math:

Even the bull — revenue +50%, operating margin doubling to the all-time FY22 peak, exited at a premium 20× — only ~matches today’s EV. The base case implies meaningful downside, and the bear (a plausible AI-capex digestion + margin reversion) implies a ~70% decline, which is roughly the magnitude VIAV actually delivered in FY22–24.

Peer comparison (the quality-vs-price inversion). VIAV now carries a richer sales multiple than larger, higher-margin, higher-return peers — the opposite of how a quality-disciplined investor would expect the market to rank them:

VIAV trading above Keysight on EV/sales — despite Keysight’s superior margins, returns and scale — is the clearest single tell that the multiple is a momentum/AI-thematic phenomenon rather than a fundamentals-justified re-rating. (Peer figures are approximate, for ordinal comparison; see Keysight’s public filings for detail.)

Sum-of-the-parts cross-check. Valuing OSP as a franchise annuity (14–18× its ~$112M segment operating income → ~$1.6–2.0B) plus NSE at a cyclical-test 2.5–4× sales (~$3.1–4.4B on a ~$1.25B run-rate) gives ~$4.7–6.4B EV, or ~$18–25/share after net debt — well below the tape. A richer growth multiple on the AI-exposed NSE lifts the top of the range toward the high-$20s/low-$30s, but it takes heroic assumptions to reach $47 on parts.

What the market is right and wrong about. Right: the demand inflection is real, the QoE is better than GAAP suggests, and OSP is a genuine franchise. Wrong (likely): extrapolating a concentrated, cyclical AI-capex surge into a permanent ~20%-margin franchise, and paying the richest multiple in the company’s history for a name that round-tripped to $6.66 twenty-four months ago. No price target; no recommendation.

11. Variant Perception

Consensus belief. VIAV is an AI-datacenter optical-test winner with a reaccelerating top line, improving margins, and a hidden-gem currency-pigment franchise — a “re-rating to growth” story, justifying a premium multiple. The factor model confirms the market has literally re-categorized it: it now trades as an AI-optics basket member (factor peers LITE/CIEN/GLW/FN), Momentum loading +0.73, Value −0.52, +405% trailing year, y1 Sharpe 6.4 — a crowded, one-way momentum trade, not a falling knife (it is ~15% off its high, not in a downtrend), though relative-strength is cooling (rs_peak −70.6).

Strongest bull case. AI-datacenter optical test is a multi-year secular wave; VIAV has the optical heritage (JDSU) plus bought capability (Spirent) to ride it; organic NSE +28–30% with 40–45% incremental margins drives operating margin durably above the mid-teens; OSP provides a stable, high-margin annuity floor; NOLs shield cash taxes; and a fully-priced equity used to repay debt leaves a clean balance sheet. If margins approach the FY22 peak and growth persists, the bull scenario (≈ today’s EV) is achievable, and a strategic acquirer could pay a premium.

Strongest bear case. Two-thirds of revenue is a cyclical, sub-WACC-through-cycle test oligopoly currently flattered by a concentrated AI-capex up-cycle; ~half the growth is acquired; owner-FCF is thin after ~$53M SBC; dilution is structural (249.5M → ~256M shares); GAAP is unprofitable; the comp plan has no ROIC discipline; insiders sold ~21× with zero buys and the company itself issued $500M at the high — and all of this is priced at the richest multiple in company history, above a higher-quality Keysight. A single AI-capex digestion quarter re-rates it hard, exactly as FY22–24 did.

The 3–5 assumptions that matter most: (1) durability and growth rate of AI-datacenter optical-test demand; (2) whether NSE operating margin can cross and hold ~15%+; (3) how much of FY26 growth is secular vs cyclical vs acquired; (4) the exit multiple a cyclical-test business can sustain; (5) dilution trajectory from converts + SBC.

Falsification. Bull breaks on one sub-12% NSE-margin quarter or visible order air-pocket. Bear breaks on two-plus more 30%+ growth quarters with NSE margin durably above 15% and book-to-bill > 1. The factor-positioning read (crowded momentum, anti-value, cooling RS, into a fresh equity supply and index-flow event) tilts the risk/reward toward the bear over the next several quarters.

12. Fact vs. Interpretation Table

13. Open Questions

1. Precise FY26 segment split and the organic-only NSE growth and margin (ex-Spirent, ex-Inertial) by quarter — disclosure narrowed post-reorg.
2. Exact OSP sub-mix (banknote pigment vs 3D-sensing/coatings) and the SICPA contract structure/renewal terms.
3. AI-datacenter optical-test backlog/book-to-bill and customer concentration among hyperscalers.
4. Total Inertial Labs consideration including the full earnout, and the cash-flow drag from contingent payments.
5. Spirent-HSE standalone margin and whether it is accretive or dilutive to NSE.
6. Normalized cash tax rate as NOLs deplete, and the true owner-FCF after SBC over a full year.
7. Management’s actual through-cycle NSE operating-margin target and the path to it.

14. What Must Be True

Bull thesis — what must be true: AI-datacenter optical-test demand is a durable multi-year secular wave (not a 2026 capex spike); VIAV converts its optical heritage + Spirent capability into share gains; organic NSE compounds ~20%+ with 40–45% incrementals; NSE operating margin crosses and holds mid-teens-plus, lifting blended op margin toward the high-teens/20%; OSP holds its annuity; and the market sustains a premium multiple on the result. Falsification test: a single quarter of sub-12% NSE operating margin, a sub-1.0 book-to-bill, or two consecutive quarters of decelerating organic NSE growth toward the high-teens would break the “durable franchise-economics” claim and the multiple with it.

Bear thesis — what must be true: the AI-capex surge is concentrated and cyclical; ~half of growth was bought; NSE reverts toward its through-cycle sub-WACC economics once the wave digests; dilution (converts + SBC) and thin owner-FCF cap per-share value; and a 98th-percentile multiple compresses toward a normal cyclical-test multiple. Falsification test: two-plus more quarters of 30%+ total growth with NSE operating margin durably above 15% and positive, growing owner-FCF (OCF less capex less SBC) would refute the “cyclical, over-extrapolated, fully-priced” claim and validate the re-rate.

The variant view: the market is underwriting the bull falsification not happening, at a price where even bull success barely earns a positive return — an asymmetric setup tilted to the downside over the next several quarters.

15. Source Appendix

See the separate Source Appendix (below) for the full citation list. Primary sources: VIAV FY2025 Form 10-K and FY2026 Q1–Q3 Form 10-Qs (SEC EDGAR, CIK 0000912093); FY2026 earnings releases and call transcripts (Q1 29-Oct-2025, Q2 28-Jan-2026, Q3 29-Apr-2026); the DEF 14A proxy; Form 4 insider filings; the May-2026 equity-offering filing (S-3ASR/424B/8-K); the convertible-note exchange 8-K (Aug-2025). Quantitative cross-checks: ROIC.ai (statements, ratios, multiples), AZI valuation_index (own-history percentiles) and AZI 5-year price CSV, and the FactorsToday factor model (loadings, leaderboard, factor-peers). Peer cross-read: Keysight Technologies (KEYS) public disclosures. All third-party aggregated data reconciled to filings; the filing governs where they disagree.

APPENDIX A — Standard Diligence Questionnaire

Supplemental to the main analysis. Fact/Interpretation/Assumption labels applied where material. Report date: 2026-06-21.

General

What thoughtful questions have other investors asked about this company?

• How much of the FY26 reacceleration is secular AI-datacenter demand versus cyclical telecom recovery versus acquired (Spirent HSE + Inertial)? (Interpretation: roughly one-half acquired, ~half organic, and the organic half is mostly AI-datacenter — but telecom/wireless is still trough-depressed, so part of the bull case is a future cyclical recovery, not pure secular growth.)
• Why is GAAP unprofitable while the stock 7×'d? (Fact: non-operating items — $46.2M debt-extinguishment loss, $24.3M earnout mark-up, $36.1M non-economic tax — mask a ~$0.27/qtr non-GAAP run-rate.)
• Should VIAV be valued as one company or as a sum-of-the-parts (franchise OSP + cyclical NSE)?
• Is OSP a stranded crown jewel that should be spun or sold?

Cyclicality & Earnings Nature

Are earnings at a cyclical high or low? NSE is emerging from a cyclical trough (FY24) into an AI-driven up-cycle; OSP is mid-cycle/stable. Blended, margins are well below the FY22 peak (op margin 10.3% GAAP Q3 vs 14.3% FY22) but rising fast — so neither extreme, with the risk skewed to the AI-capex side being nearer a high than a low. (Interpretation.)

Driven by external environment or internal actions? Both: external AI-datacenter capex wave + internal M&A (Spirent, Inertial) and margin/operating leverage.

How stable are revenues? Low stability in NSE (capex-cyclical, ±20%+ peak-to-trough demonstrated FY22–24); high stability in OSP (currency-security annuity). Blended volatility is high — the stock has a 52-week range of $6.66–$55.33.

Outlook for products/services; how big is the market? AI-datacenter optical/Ethernet test is a multi-year, fast-growing but concentrated and cyclical market; telecom test is large, mature, low-growth; banknote security is small, slow, defensible. International and US; hyperscaler-concentrated on the growth edge.

Business Quality & Competitive Moat

Is the industry getting more or less competitive? NSE: consolidating (Keysight absorbed most of Spirent) but still a crowded oligopoly — effectively more concentrated at the top, with Keysight the scale leader. OSP: stable, high-barrier, low-competition.

How profitable is the business (ROIC, ROE)? Distorted by negative tangible common equity (goodwill+intangibles > equity) and the −$69.7B legacy accumulated deficit. Ex-goodwill, OSP earns very high returns (~36% op margin on little capital); NSE earns sub-WACC through-cycle. Blended returns are unimpressive relative to the multiple. (Interpretation.)

How profitable is the industry; barriers to entry? NSE: moderate margins, moderate barriers (software/installed base), commoditizing. OSP: high margins, very high barriers (certification, sovereign trust, chemistry).

Can the business be easily understood? Moderately — two segments with opposite economics; the consolidated GAAP and capital structure (converts, earnouts, NOLs) require normalization.

Undermined by foreign low-cost labor? Limited direct risk (IP/precision-instrument and certified-materials businesses), but NSE faces low-cost competition in commodity field instruments.

Do brands matter? Yes in NSE (Viavi/JDSU heritage, installed base) and decisively in OSP (currency-grade trust).

Nature of competition / switching costs? NSE: moderate switching costs (calibration, software, training) but real alternatives exist. OSP: very high — re-qualifying a currency security feature across sovereign mints is slow and risky.

Financial Condition & Balance Sheet

Assets not fully on the balance sheet? The OSP franchise value and the NOL tax shield are economically large but not capitalized; conversely, goodwill/intangibles ($1.1B) overstate tangible asset backing (tangible common equity is negative).

Off-balance-sheet liabilities? Operating leases (modest), pension (~$51M), and contingent earnout consideration (Inertial) that has been marked up. Convertible dilution is on-balance-sheet but economically a forward share-issuance liability.

How conservative is the accounting? Mixed: GAAP is conservative-looking (unprofitable on real charges), but heavy reliance on non-GAAP adjustments and large SBC (~$53M/yr ≈ GAAP NI) argues for skepticism on “adjusted” headline figures.

How CapEx-hungry? Moderate — instruments/materials manufacturing; capex is a few percent of revenue, lighter than a fab but heavier than pure software. Working-capital builds on the revenue ramp consume cash (Q3 FY26 OCF was negative).

Capital Allocation & Management

FCF generation and use; philosophy? FCF is real but thin after SBC; recent philosophy is M&A-led growth funded by debt, then de-levered via opportunistically-timed equity. (Fact: $500M equity issued at ~$45 ATH to repay the Spirent Term Loan B.)

Significant recent acquisitions? Yes — Spirent HSE (~$400M cash) and Inertial Labs (~$121.6M cash + earnout). Both target the AI-datacenter/defense growth vectors.

Buying back shares? Not currently (buybacks dormant since the ~$236M FY22 program); net share count is rising via converts + SBC.

Issuing shares to insiders? SBC ~$53M/yr; plus the $500M public raise. Dilution is a live concern.

Compensation policy? Proxy keys on revenue, non-GAAP operating income, and relative TSR vs Nasdaq Telecom — no ROIC/capital-efficiency metric (governance demerit).

Motivations of management? CEO Khaykin/CFO Daskal have executed a credible strategic pivot, but insider selling is unanimous (~21 sells, 0 buys; CEO sold $22M+ at $51–55) — consistent with a “fully valued” internal read.

Valuation & Market Data

ADR, MLP, or K-1? No — ordinary US common stock, single class, NASDAQ-listed; no K-1.

Dividend policy? No dividend.

How profitable; is net income diverging from cash from operations? Yes, materially — GAAP NI is negative (non-operating items) while non-GAAP operating income is solidly positive; OCF is lumpy (negative in Q3 FY26 on earnout/WC). The divergence is mostly explainable but warrants normalization.

Risks & Downside

What would cause the stock to decline? A valuation de-rate from the 98th percentile; an AI-capex digestion/order air-pocket; NSE margin reversion; renewed telecom weakness; dilution; disappointing integration.

Catastrophic loss risk? Low — net debt is modest (~0.58B) against rising EBITDA; the converts are deep-ITM (an equity-dilution, not default, risk).

Chance of total loss? Very low. The dominant risk is a large drawdown (a 40–70% de-rate is plausible from here, matching the FY22–24 precedent), not impairment of the enterprise.

Recent News & Events

Has the business environment changed recently? Dramatically — from telecom-capex bust to AI-datacenter boom; the entire re-rate rests on this.

Significant acquisitions? Spirent HSE and Inertial Labs (above).

Change in accounting policies? Segment reorganization (Mar-2025) consolidated NE+SE into NSE.

Recent changes — markets, facilities, management, capital? New AI-datacenter end-market emphasis; $500M equity raise (May-2026); Aug-2025 convertible-note exchange; ~$60 intraday all-time high (Apr-2026) then a ~15–22%-off-high fade on the offering and index/quad-witch flow (18-Jun-2026, 49M shares).

APPENDIX B — Source Appendix

Report date: 2026-06-21. Primary sources first. Third-party aggregated data is labeled and was reconciled to filings; where they disagree, the filing governs.

Primary — SEC filings (EDGAR, CIK 0000912093)

• Form 10-K, FY2025 (fiscal year ended ~28-Jun-2025) — segment revenue and operating income (NSE $776.6M / OSP $307.7M; OSP ~36.5% op margin); customer concentration (SICPA ~15.4% of revenue); risk factors; capital structure. https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0000912093&type=10-K — accessed 2026-06-21.
• Form 10-Q, FY2026 Q1 (Sep-2025), Q2 (Dec-2025), Q3 (Mar-2026) — quarterly revenue ($299.1M / $369.3M / $406.8M), GAAP loss bridge (debt-extinguishment loss, earnout mark-up, tax), balance sheet (cash, debt, converts, equity), diluted share counts. EDGAR, accessed 2026-06-21.
• Form 8-K filings (2024–2026) — earnings releases (FY25 Q4 7-Aug-2025; FY26 Q1 29-Oct-2025; Q2 28-Jan-2026; Q3 29-Apr-2026); convertible-note exchange (Aug-2025); acquisition announcements (Spirent HSE; Inertial Labs); May-2026 equity offering. EDGAR, accessed 2026-06-21.
• S-3ASR / 424B prospectus (May-2026) — $500M primary equity offering (~11.1M shares at ~$45), use of proceeds (repay Term Loan B). EDGAR, accessed 2026-06-21.
• DEF 14A proxy statement — executive compensation metrics (revenue, non-GAAP operating income, relative TSR vs Nasdaq Telecom; no ROIC metric); board. EDGAR, accessed 2026-06-21.
• Form 4 insider filings (trailing period) — ~21 sales, zero open-market purchases; CEO Oleg Khaykin sales ($22M+ at $51–55, 10b5-1) early May-2026. EDGAR, accessed 2026-06-21.

Primary — earnings-call transcripts

• VIAV FY2026 Q3 earnings call (29-Apr-2026), Q2 (28-Jan-2026), Q1 (29-Oct-2025) — segment commentary, AI-datacenter optical demand, guidance, capital structure. Accessed via ROIC.ai transcript tools, 2026-06-21.

Quantitative cross-checks (third-party; reconciled to filings)

• ROIC.ai — income statement, balance sheet, cash flow, profitability/credit/liquidity ratios, per-share data, valuation multiples (FY2020–FY2026 annual + quarterly). NOTE: ROIC’s enterprise value ($267M) and market cap are computed on a stale share count and are unusable; EV was recomputed (~$11.6B). Accessed 2026-06-21.
• AZI valuation_index (azitrading.com) — own-history valuation percentiles as of 2026-06-18: price $47.17, P/B 13.9× (98.7th pctile), P/S 8.0× (98.4th pctile), composite 98.6th pctile; P/E null (negative TTM GAAP EPS −$0.25). Accessed 2026-06-21.
• AZI price CSV (azitrading.com) — 5-year daily OHLCV, EMAs, beta; used for the price-action event map (5yr low $6.66 26-Jun-2024; ATH close $55.33 1-May-2026; $47.17 18-Jun-2026 on 49.0M shares). Accessed 2026-06-21.
• AZI news feed — recent headlines/sentiment (through 15-Jun-2026). Accessed 2026-06-21.
• FactorsToday factor model (factorstoday.com) — stock loadings (Momentum +0.73, Value −0.52, Market 0.88, beta ~1.19), leaderboard (y1 +405%, y1 Sharpe 6.4; raw ~+157% 6m, ~+50% 3m de-annualized), factor-similar peers (LITE/CIEN/GLW/FN). Third-party statistical estimates. Accessed 2026-06-21.

Peer / cross-read

• Keysight Technologies (NYSE: KEYS) public filings and disclosures — test & measurement industry framing and valuation comparison (VIAV’s EV/sales now exceeds KEYS).

Notes on reconciliation

• Computed enterprise value: ~234M shares × $47.17 ≈ $11.0B market cap + ~$1.08B gross debt − ~$0.50B cash ≈ $11.6B EV (ROIC’s figure discarded).
• GAAP-to-normalized bridge (9-mo FY26) reconstructed from the 10-Q line items (loss on debt extinguishment $46.2M; contingent-consideration mark-up $24.3M; tax provision $36.1M against a pretax loss; intangible amortization $47.6M).
• Acquisition figures (Spirent HSE ~$400M cash; Inertial Labs ~$121.6M cash + earnout) per 8-K/10-K; total Inertial consideration including the full earnout is an open question.