KEYS — Keysight Technologies, Inc.

Sentiment is driven by the Q3 FY26 (quarter ended 2026-07-31, reported 2026-08-18) beat: revenue of $1.846B vs ~$1.75B expected (+36.5% YoY) and non-GAAP EPS of $3.07 vs $2.48 consensus, with management flagging a second consecutive record orders quarter (>$2B, $2.91B of total demand across segments) and raising the full-year outlook. Analysts are firmly constructive - 13 Buy / 3 Hold / 0 Sell, Buy consensus - with price targets clustered at $397-$418 and highs around $440-$452. Notably, the stock's reaction has been muted rather than euphoric: even after the 2026-09-13 gain of 3.99%, KEYS is still down 4.47% over one month and 3.43% over three months versus a 21.0% six-month gain, which tells you the beat was largely anticipated after a run to a 52-week high of $374.96. This is bullish-but-digesting sentiment, not a melt-up.

The three-year quarterly record shows a clear inflection, not a blip. Revenue was essentially flat for nine quarters - $1.311B (Oct-23), $1.259B, $1.216B, $1.217B, $1.287B (Oct-24), $1.298B, $1.306B, $1.352B, $1.419B (Oct-25) - then broke out: $1.600B (Jan-26), $1.717B (Apr-26), $1.846B (Jul-26). YoY growth has accelerated three straight quarters: +23.3%, then +31.5%, then +36.5%, off a base that had been growing ~10-11% in FY25. Operating margin followed the same path: after grinding between 14.6% and 17.9% from Jan-24 through Jan-26 (15.5% in the Jan-26 quarter, 15.3% in Oct-25), it jumped to 23.7% in Apr-26 and 25.0% in Jul-26 - roughly 800bp of expansion in two quarters - with gross margin moving from 61.7% (Jul-25) to 65.9% (Jul-26). Diluted EPS went $1.32, $1.63, $2.02, $2.29 over the last four quarters, and R&D rose 24.8% YoY ($312M vs $250M) while falling as a share of revenue (18.5% to 16.9%), which is genuine operating leverage rather than cost-cutting. Two caveats: GAAP net income is noisy (the Oct-24 quarter posted a -$73M net loss on $230M of operating income, while Jul-24 showed $389M of net income on only $205M of operating income - both one-off, below-the-line items), so GAAP EPS overstates the volatility in the underlying business. Cash generation is clean: FCF of $403M in the Jul-26 quarter, $1.469B TTM, with a -0.34% accruals ratio (cash flow slightly ahead of reported earnings). Bottom line: a flat-to-modest-growth business through FY25 has converted into an accelerating upcycle, most plausibly tied to AI/networking test demand.

Expensive on trailing numbers, fair-to-full on forward numbers - this is not a cheap stock. As of 2026-09-13, TTM P/E was 46.6x on $7.26 of GAAP EPS versus a 5-year average P/E of 33.1x; P/S was 8.79x versus a 5-year average of 5.53x; EV/EBITDA was 34.3x; P/B was 8.9x. The trailing GAAP figures understate earning power (amortization, stock comp and one-off items), so the more relevant lens is forward: at $338.64, the stock trades at 29.5x the $11.485 consensus next-FY EPS, close to its own long-run trailing average and nowhere near bubble territory on that basis, with a reported PEG of 0.36. The problem is the interaction of multiple and cycle: you are paying ~29-30x forward for earnings that sit on a 25% operating margin peak (versus a 15-17% norm for the prior two years) and 8.8x sales versus a 5.5x historical average. ROIC of 8% also doesn't scream franchise economics at a 34x EV/EBITDA. Verdict: priced for the upcycle to continue - reasonably valued if FY27 revenue/EPS estimates get revised up, expensive if order growth normalizes.

The setup as of 2026-09-13 is 'estimates need to keep rising to justify the price.' A 29.5x forward multiple on consensus EPS of $11.49 is defensible, but that consensus was set around the report and likely embeds only ~8% revenue growth to ~$7.11B next FY - far below the +36.5% just delivered and below what a book-to-bill above 1.0 (two consecutive quarters of >$2B orders) would imply. That creates an asymmetric-but-real upside path: if FY27 guidance on ~2026-11-23 comes in above roughly $12.00 of EPS, the same 30x multiple produces a $360+ stock without any re-rating, and analysts likely chase targets toward the $416-$418 cluster. The flip side is that the stock already rallied ~21% in six months and the Q3 beat did not produce a durable breakout - the market has pre-paid for good news. Practically: this is a 'buy on estimate revisions, not on multiple expansion' name. If orders decelerate or guidance merely meets the current conservative bar, there is limited valuation support - the 5-year average P/S of 5.5x would imply a share price in the low $200s on today's sales, so the downside case is a de-rating, not just a flat tape.

As of 2026-09-13, RSI(14) stood at 56.77 - neutral, with no overbought or oversold signal - and the stock closed at $338.64, which is 3.36% above its 20-day SMA, 4.0% above its 50-day, and 16.09% above its 200-day, so the primary uptrend is intact. However, the intermediate picture is consolidation, not momentum: KEYS was -4.47% over one month and -3.43% over three months, having peaked at a 52-week high of $374.96 (52-week low $158.79). The 3.99% one-day pop on 2026-09-13 (previous close $325.65) suggests buyers are defending the $320s. Entry-timing implication: this is not an oversold bounce opportunity and not a breakout - it's a stock hovering in the upper third of its range with moderate RSI. Adding at 16% above the 200-day requires conviction in the order book; a pullback toward the 50-day (~$325) or the $305-$315 zone would offer materially better risk/reward, while a decisive close above ~$375 would confirm the next leg. Beta of 1.208 means it will amplify broad tech drawdowns.

• AI infrastructure capex cycle: hyperscaler and networking spend on 800G/1.6T Ethernet, PCIe 6/7, HBM and advanced packaging is the dominant demand driver for Keysight's Communications Solutions Group - powerful multi-year tailwind, but lumpy and capable of a digestion pause.
• Semiconductor and electronics R&D budget cycles: Keysight sells into customers' capex/R&D budgets, making it a classic capital-equipment-adjacent cyclical whose orders can swing hard in both directions (revenue was flat from 2023 through 2025 before the current breakout).
• Interest rates and enterprise capex financing: tighter credit raises the hurdle for customer R&D spending, though Keysight's net cash position (~$2.61B cash vs $1.82B long-term debt, current ratio 2.02x) insulates the company itself from funding stress.
• Aerospace, defense and automotive electrification: resilient global defense budgets and software-defined vehicle investment support the EISG/automotive franchises and partially offset commercial electronics softness.
• Trade policy, export controls and tariffs: test and measurement equipment is dual-use and geopolitically sensitive; China restrictions, retaliatory measures or supply-chain tariffs could pressure a meaningful slice of revenue and complicate bookings timing.

• Q4 FY26 earnings, estimated for 2026-11-23 after market close - the key event, because it delivers final FY26 results and, more importantly, the first FY27 guidance and a read on whether the order surge converts into revenue.
• Order and book-to-bill momentum: a third consecutive quarter above $2B in orders would validate the AI-driven demand thesis and almost certainly force upward EPS revisions toward or above $12.00; a normalization would do the opposite.
• Analyst estimate and price-target revisions post-Q3 - current targets cluster at $397-$418 (high $440-$452), and further upward moves would mechanically support the multiple.
• Continued hyperscaler/network OEM capex announcements (AI data center buildouts, next-gen interconnect standards) that feed directly into Keysight's test roadmap.
• New product ramps in EDA/emulation and software (PathWave) plus any portfolio M&A that shifts mix further toward higher-margin recurring revenue.

• Cyclicality at a peak: revenue was flat for roughly nine quarters before the current +36.5% YoY print. Paying ~30x forward earnings for peak-cycle margins (25.0% operating margin in Jul-26 vs 15.3-17.9% in the prior two years) leaves little room for a normal order digestion phase.
• Valuation with no cushion: 8.79x P/S versus a 5.53x 5-year average, 34.3x EV/EBITDA, 8.9x book, and ROIC of only 8.0%. A reversion toward historical sales multiples would be a violent downside move, not a mild one.
• Customer concentration and order lumpiness in AI/semiconductor capex - the $2.91B 'total demand' figure implies large, chunky bookings that can reverse quickly if a handful of hyperscalers or chipmakers pause.
• Geopolitical/export-control exposure: China and trade restrictions on dual-use test equipment are a recurring, hard-to-quantify revenue risk.
• GAAP-to-non-GAAP gap and earnings quality optics: TTM GAAP EPS of $7.26 versus a ~$11+ non-GAAP run-rate means a large chunk of reported profit is adjusted away (amortization, stock-based comp, restructuring), and the Oct-24 quarter's -$73M GAAP net loss shows how volatile the below-the-line items can be.

Keysight is a genuinely high-quality franchise (64.7% gross margin, 20.3% TTM operating margin, 22% FCF margin, net cash balance sheet) that has just entered a real, order-confirmed upcycle rather than merely beating a soft quarter - three consecutive quarters of accelerating revenue growth (+23%, +32%, +36.5%), two straight record order quarters, and ~800bp of operating-margin expansion in six months. The problem is price, not the business: at $338.64 on 2026-09-13 you are paying 29.5x forward EPS and 8.8x sales for peak-cycle margins, so the stock is fairly valued-to-modestly-attractive rather than a bargain, and the Q3 beat failed to produce a durable breakout. My lean is to own it for the earnings-revision cycle (FY27 guidance on ~2026-11-23 is the swing factor) but not to chase it at 16% above the 200-day; a pullback toward $320 or below is where the risk/reward improves materially.

Primary method: a justified forward P/E of 30.5x - a deliberate discount to the 5-year average trailing P/E of 33.1x to reflect peak-cycle margins and a 34.3x EV/EBITDA - applied to consensus next-FY non-GAAP EPS of $11.485 gives $350. Cross-checks bracket that number: a simplified DCF with $1.9B of forward FCF, 12-15% near-term growth, a 4% terminal rate and a 9% WACC yields roughly $325-$355, and 25x forward EV/EBITDA on ~$2.4B of forward EBITDA plus ~$0.79B of net cash yields about $356. The result is only ~3% above the 2026-09-13 price of $338.64 and well below the $416.63 sell-side consensus, which I view as anchored to a more generous ~36x forward multiple and to EPS estimates that likely rise only if Q4 FY26 guidance confirms the order surge.