Keysight’s Q4 beat says the recovery is real, but the stock should be valued on order quality and FY26 core growth, not acquisition optics

The print changes the debate because Keysight did not merely clear a low bar, it paired a street beat with order momentum and an FY26 framework that makes the next two quarters hard to dismiss as a one-quarter rebound. What was priced in was a recovery from the FY2024 trough, with revenue already having stepped from $1,217.0 million in Q3 FY2024 to $1,352.0 million in Q3 FY2025 and the Street looking for $1,384.3 million in Q4 FY2025. What actually surprised was that Q4 revenue came in at $1,419.0 million, a +2.5% surprise, while EPS of $1.91 beat the $1.83 estimate by +4.4%. The more important surprise sits behind the income statement: Neil Dougherty said “Orders of $1.533 billion were up 14% on a reported basis or 12% on a core basis.” That order number is the anchor for the thesis because it is larger than the reported quarterly revenue base of $1.419 billion and it came with only $22 million in orders from recently completed acquisitions and $4 million from currency, limiting the ability to dismiss the demand signal as deal-related noise.

That demand signal matters because the revenue trajectory had already been improving before the Q4 beat, but margins had not yet told the same story. Revenue moved from $1,216.0 million in Q2 FY2024 to $1,217.0 million in Q3 FY2024, then $1,287.0 million in Q4 FY2024, $1,298.0 million in Q1 FY2025, $1,306.0 million in Q2 FY2025, $1,352.0 million in Q3 FY2025, and $1,419.0 million in Q4 FY2025. That is a clean sequential recovery, including +5.0% QoQ and +10.3% YoY in Q4 FY2025. Yet gross margin in the quarterly history moved the other way into the print, from 63.2% in Q1 FY2025 to 62.3% in Q2 FY2025, 61.7% in Q3 FY2025, and 61.2% in Q4 FY2025. The stock reaction should therefore depend less on whether revenue beat, which it did, and more on whether investors believe mix, acquisitions, tariffs, and scale can stop gross margin from eroding as volumes recover. The Q1 FY2026 and Q2 FY2026 entries in the history give the bull case concrete numbers to point to, with revenue at $1,600.0 million and gross margin at 62.2% in Q1 FY2026, then revenue at $1,717.0 million and gross margin at 68.6% in Q2 FY2026. The defensible interpretation is that the Q4 FY2025 margin trough is not fatal if the next phase resembles those forward trajectory numbers, but it is also the part of the story that cannot be assumed away.

The capacity of the model to turn sales into earnings is where the company’s own reporting basis is stronger than the GAAP quarterly EPS line in the history, so the distinction matters. On the Street-comparison basis, Q4 FY2025 EPS was $1.91 versus $1.83, a +4.4% surprise. In the quarterly history, diluted EPS for Q4 FY2025 is $1.32, following $1.10 in Q3 FY2025 and $1.49 in Q2 FY2025, while Q4 FY2024 was -$0.42. Those are different reporting bases, and they should not be collapsed into one trend. The company’s own operational framing was more constructive: Dougherty said, “Looking at our operational results for Q4, we reported a gross margin of 64%, operating expenses of $539 million, and an operating margin of 26%.” The wording matters because management is trying to keep investors focused on operational profitability rather than the lower gross margin series in the quarterly history. That is fair only if Q1 FY2026 EPS guidance of $1.95 to $2.01 and the subsequent history entries of $1.63 in Q1 FY2026 and $2.02 in Q2 FY2026 resolve in favor of leverage rather than mix pressure.

That margin debate also reframes the FY26 guide, because acquisitions increase revenue visibility while complicating organic quality. Dougherty’s key commitment was that FY 2026 revenue growth, excluding acquisitions, is expected to be “at or above the high end of our 5% to 7% long-term target.” That phrase is the most investable line in the call: “excluding acquisitions” prevents the market from treating the FY26 story as purchased growth, and “at or above the high end” pins the organic algorithm to 7% or better without requiring us to manufacture a number beyond the pack. At the same time, he said the recently completed acquisitions of Spirent, the Optical Solutions Group, and PowerArtist are expected to contribute approximately $375 million of revenue in FY 2026. The market may be mispricing that combination by applying a conglomerate discount to acquired revenue while underweighting the core guide. The bear case is that $375 million of acquired revenue masks a slower base, but the company explicitly separated the acquired contribution from core growth, and Q4 included only $11 million of revenue from recently completed acquisitions and $7 million from currency against reported revenue of $1.419 billion. That makes Q4’s +10% reported growth and 9% core growth harder to dismiss.

The segment mix supports the same conclusion, with communications doing the heavier lifting and industrial recovering enough to reduce single-end-market risk. Communications Solutions Group generated $990 million in Q4 revenue, up 11% on a reported basis or 9% on a core basis. Inside that, commercial communications revenue was $660 million, up 12%, while aerospace, defense, and government was $330 million, up 9%. Electronic Industrial Solutions Group generated $429 million, up 9% on a reported basis or 8% on a core basis, with growth in semiconductor and general electronics. The market likely expected communications to improve, but the spread of growth matters: the biggest segment delivered 9% core growth, and the smaller segment delivered 8% core growth. That is not a narrow wireless-only rebound. It is also why the Q4 order number of $1.533 billion matters: broad segment revenue growth becomes more believable when orders are up 12% on a core basis, not just when shipments catch up to delayed demand.

The second-order read-through for customers is that the test budget cycle is loosening across AI networking, RF, and semiconductor validation, but the magnitude from this pack argues for measured optimism rather than an across-the-board capex boom call. For NVIDIA, Keysight’s exposure is test and measurement and parametric testers, and the call explicitly surfaced NVIDIA’s $1 billion strategic investment in Nokia around AI-powered networks for future 6G RAN infrastructure. That does not tell us NVIDIA is buying more Keysight tools, but it does align with Keysight’s commercial communications revenue of $660 million, up 12%, and Communications Solutions Group revenue of $990 million, up 11% reported or 9% core. For Qualcomm, also a customer for test and measurement and parametric testers, the relevant signal is wireless growth inside that same $660 million commercial communications base, even though management did not quantify wireless separately. For Broadcom, the wireline component matters more: management attributed commercial communications growth to continued strength in wireline and growth in wireless, and the quantified base tied to that statement is $660 million, up 12%. There are no suppliers listed for Keysight in the pack, so the supply-chain implication is customer-side only: the read-through is incremental validation activity at named semiconductor and communications customers, not a documented procurement uplift for any Keysight supplier.

That customer-side interpretation is strengthened by the cash return and balance-sheet context, because Keysight is not financing the recovery by starving internal investment. Satish Dhanasekaran said, “In fiscal 2025, we achieved record free cash flow of $1.3 billion while investing in R&D, completing three acquisitions, and returning approximately $375 million through buybacks.” The quote earns its place because it ties capital allocation to the investment cycle rather than presenting buybacks in isolation. Cash and cash equivalents ended the quarter at $1.9 billion, cash flow from operations was $225 million, and free cash flow was $188 million. The company repurchased 595,000 shares at an average price of approximately $168 for a total consideration of $100 million, while full-year share repurchases totaled $375 million or approximately 30% of the $1.3 billion in free cash flow generated this year. The pushback is that annual interest expense is expected to be approximately $110 million at current debt levels, capital expenditures are expected to be approximately $160 million, and the non-GAAP effective tax rate is modeled at 14% for FY 2026. Those numbers do not break the thesis, but they narrow the margin for disappointment if acquired assets take longer to integrate or if gross margin does not move away from the Q4 FY2025 history level of 61.2%.

The peer lens also argues that Keysight’s stock should not be valued as the fastest cyclicality trade in test and assembly, but as the cleaner core-growth and cash-flow compounder within the group. In the latest reported quarter, ATEYY posted revenue YoY of +43.8% with gross margin of 67.4%, and 6871.T posted revenue YoY of +48.3% with gross margin of 47.3%. Keysight’s Q4 FY2025 revenue YoY was +10.3% with gross margin of 61.2% in the quarterly history, while management’s operational gross margin was 64%. That comparison cuts both ways. Keysight is not matching the highest peer revenue YoY rates in the table, but its margin profile is closer to the upper end than to lower-margin peers such as 6315.T at 36.2% gross margin or 6125.T at 25.4% gross margin. The investable point is not that Keysight deserves the most cyclical multiple in the group; it is that a 12% core orders increase and an FY26 core guide at or above the high end of 5% to 7% offer a cleaner growth bridge than a pure semiconductor equipment upcycle call.

The call delivery adds one caution flag to that constructive read, because headline sentiment improved but Q&A behavior was not cleanly better on the Q4 FY2025 call. The tone history shows Q4 FY2025 sentiment at 0.50 versus Q3 FY2025 at 0.45, guidance_tone at 0.52 versus 0.54, and uncertainty at 42.5 versus 56.4. Those numbers are directionally supportive: management sounded more positive overall and less uncertain than in the prior quarter. The conflict is qa_evasiveness, which was 43.0 in Q4 FY2025 versus -5.3 in Q3 FY2025, while prepared_sentiment fell to 0.03 from 0.74 and qa_sentiment rose to 0.50 from 0.22. That mix says the prepared script was restrained, Q&A sounded better, but evasiveness spiked. For a PM, the right conclusion is not to ignore the call tone, but to weight hard guideposts more heavily than verbal confidence. The later tone series shows Q2 FY2026 guidance_tone at 0.66, uncertainty at 37.1, and qa_evasiveness at -1.8, which would confirm that the Q4 evasiveness was more about transition complexity than deteriorating visibility.

The central risk to the thesis is that Q4 revenue quality is real but the margin and integration path disappoints, leaving investors with sales growth that does not translate into EPS upside. The company guided Q1 earnings per share to $1.95 to $2.01 based on a weighted diluted share count of approximately 173 million shares, while the quarterly history shows Q1 FY2026 revenue at $1,600.0 million, gross margin at 62.2%, and diluted EPS at $1.63. It then shows Q2 FY2026 revenue at $1,717.0 million, gross margin at 68.6%, and diluted EPS at $2.02. Those are the numbers that will confirm or break the argument. If revenue stays on a path from $1,419.0 million in Q4 FY2025 toward $1,600.0 million in Q1 FY2026 and $1,717.0 million in Q2 FY2026, with gross margin recovering from 61.2% toward 62.2% and then 68.6%, the market should stop treating Q4 as a one-off beat. If instead Q1 EPS fails to hold the $1.95 to $2.01 company guide on its own non-GAAP basis, or if organic commentary backs away from FY 2026 revenue growth excluding acquisitions at or above the high end of the 5% to 7% target, the bull case becomes a purchased-growth story built on approximately $375 million of acquisition revenue. The next dates to watch are the Q1 FY2026 and Q2 FY2026 reports tied to the fiscal periods ending 2026-01-31 and 2026-04-30, with orders relative to $1.533 billion, gross margin relative to 61.2%, 62.2%, and 68.6%, and Q&A tone relative to the Q4 FY2025 qa_evasiveness reading of 43.0 as the cleanest tells.