Keysight Technologies Investment Value Deep-Dive

Hunting ten-year five-baggers among great growth stocks — pressing the upside question: "Can it get much bigger?"

• QHow large is its market ceiling? Is it expanding an existing pie, or creating an entirely new market?6/10

  A

  The ceiling is high enough, but the business is essentially “expanding and deepening an existing pie,” not creating a new market from scratch. That makes its growth “high quality but bounded,” rather than the Baillie-style ideal of “redefining a category.”

  Start with the pie itself. Keysight serves the electronics engineering value chain of “design-simulation-validation-manufacturing test-operations optimization.” At its 2023 Investor Day, the company put its serviceable available market (SAM) at about $20.9 billion, with long-term growth of about 4%–6%, and raised its own long-term core growth target to 5%–7% (the report explicitly labels this as a management estimate, not an externally audited figure). A $20 billion-level pool growing at single digits means the ceiling is still far from binding for a company with fiscal 2025 revenue of about $5.375 billion ($4.063 billion from products and $1.312 billion from services and other, per the report). Based on its self-reported framework, its combined share across the three major markets is about 26%, so there is still room for penetration, but the pool itself will not expand explosively.

  Then look at the real driver behind “expanding the pie”: it is not inventing new demand, but riding the tailwind of rising electronics complexity. 5G/6G, AI data centers, next-generation semiconductors, automotive electrification and intelligence, aerospace and defense: these end markets are becoming harder, and the unit value of test and validation rises with that complexity. That effect is clearly being amplified by the current cycle: Keysight’s fiscal 2026 second-quarter revenue (quarter ended 4/30) was $1.72 billion, up 31% year over year; orders were $2.05 billion, up 56% year over year; and AI-related business doubled in size. But the honest interpretation is this: this is an “existing pie being inflated at a specific point in time by the AI capex cycle,” not Keysight opening a market that did not previously exist. It is still selling test capabilities engineers already needed; this round of end-market demand is simply much stronger.

  There is some element of “creating a new market,” but it is limited. The company is indeed extending from hardware into software, subscriptions, SaaS, and integrated simulation (the report notes that services and other revenue rose from about 18.7% of revenue in 2020 to about 24.4% in 2025), and it has pushed the boundaries of its capabilities upstream and downstream through acquisitions such as ESI (simulation), Riscure (security), and Spirent (network and protocol testing). This is closer to “capturing a higher-value layer inside the existing pie, while also modestly expanding the serviceable scope,” rather than creating a new category.

  Conclusion: the ceiling is high enough that it is not a near-term constraint, but the growth story is “continuing to deepen and expand in a long-runway, slow-growing market where it is already a leader.” This is structural tailwind plus penetration within an existing pool, not creation of incremental new demand. On Baillie’s “market ceiling” dimension, it is solidly upper-middle rather than top tier. The upper bound is determined less by imagination space than by how deeply it can defend pricing and share in high-value segments.

  Jun 11, 2026
• QCan its revenue at least double over the next five years? Will growth be driven mainly by volume, price, or new businesses?4/10

  A

  Most likely not. On a through-cycle normalized growth basis, doubling revenue over the next five years (about a 15% CAGR) is far above the structural center of gravity of this business. It can briefly run at that level today because of an AI/data-center capex-cycle pulse, not because of a sustainable engine.

  First, set out the base and the normal growth center. Fiscal 2025 revenue was about $5.375 billion (per the report); the company’s self-reported long-term core growth target is 5%–7%, and its serviceable market’s long-term growth is 4%–6% (2023 Investor Day management estimate). Even at the top end of the target, a 7% CAGR would only reach about $7.5 billion in five years, up about 40%, still far from doubling (about $10.7 billion). Doubling in five years requires maintaining ~15% growth for five consecutive years, more than twice the industry’s structural center. Keysight has never sustained that historically (the report discloses fiscal 2021 +17.1%, 2022 +9.7%, 2023 +0.8%, 2024 −8.9%, 2025 +8.0%), which is a typical pattern of “long-term upward trend with periodic drawdowns,” not linear high growth.

  Now look at the quality of the current high-growth round. It is genuinely strong, but its driver mix means it should not be extrapolated. Fiscal 2026 second-quarter revenue was $1.72 billion, up +31%; orders rose +56% to above $2 billion; the company raised full-year growth guidance to the “high 20%” range and explicitly said its AI-related business doubled in size. By driver:

  • Volume is the main force, and it is highly cyclical. Second-quarter Communications Solutions Group (CSG) revenue was $1.231 billion, +35% (commercial communications +40%, aerospace-defense-government +24%), and Electronic Industrial Solutions Group (EISG) revenue was $486 million, +24%. This is volume release from AI data centers, 800G/optical, 6G pre-research, and defense orders all moving up at once. It is essentially an end-market capex cycle, not a permanent demand step-up.
  • Price contributes only modestly. This is not a consumer-staples business with free pricing power. The report describes its power as “technical pricing power,” not “monopoly pricing power.” Large customers, governments, and operators procure rationally, and price is tied to precision, functionality, and schedule. Price increases alone cannot support a doubling.
  • New businesses/acquisitions are real incremental contributors, but not enough to support a doubling: Spirent was acquired on October 15, 2025, for consideration of about £1.16 billion (about $1.46 billion). The report’s framework puts recent acquisitions’ fiscal 2026 revenue contribution at about $375 million. That is a meaningful extension, but relative to a revenue base above $5 billion, it is only a single-digit percentage point contribution.

  So how should “doubling” be viewed? Treating today’s “high 20%” full-year guidance as a five-year norm is the main trap in this question. The report itself repeatedly emphasizes that the second-quarter surge “mixes AI/data-center demand, an upturn in communications testing, and accounting effects from a one-time tariff refund,” so investors need to separate “fundamental improvement” from “cyclical benefits.” A more realistic judgment is this: at the cyclical peak, the next year or two may deliver double-digit growth and perhaps close to 20% year-on-year growth, but over a five-year horizon, after returning to the center of gravity, normalized growth is more likely in the mid-to-high single digits. Five-year cumulative growth is likely around 40%–60%, not 100%. Growth is mainly volume-led, with price as a secondary factor and new businesses as a supplement, but after looking through the cycle the combination still struggles to reach the doubling line.

  Jun 11, 2026
• QFive years from now, what will take over as the next growth engine? Does this “second curve” already exist today?5/10

  A

  The “second curve” does exist today and is already contributing revenue, but it looks more like a “natural extension and deepening” of the same platform than an independent new engine capable of recreating another Keysight. That means the likely successor five years from now is “software + services + acquisition synergies” lifting value capture from the existing business, not a disruptive new category.

  Start with the lines that already exist today and could take over:

  • The rising share of software, subscriptions, and services is the clearest “quality second curve.” The report discloses that services and other revenue grew from $789 million in 2020 to $1.312 billion in 2025, and its revenue share rose from about 18.7% to about 24.4%; deferred revenue at fiscal 2025 year-end was $884 million (above $767 million in 2024), and remaining performance obligations were about $629 million. This line will not double revenue, but it will continue improving cash-flow stability and gross-margin quality. It takes over the baton of “profit resilience,” not “scale explosion.”

  • Capability expansion after acquisition integration is the most realistic “scale second curve.” Spirent was acquired on October 15, 2025 (consideration about $1.46 billion), extending capabilities into network and protocol performance validation. Together with earlier acquisitions such as ESI (simulation), Riscure (security), and AnaPico, the company has added teeth at both ends of “design simulation-network testing-security validation.” The report estimates that recent acquisitions will contribute about $375 million of fiscal 2026 revenue, and Spirent synergies are expected to bring 150–200 basis points of operating-margin improvement by fiscal 2026. But note the key caveat: this curve’s quality as a successor depends on integration delivery. The report explicitly lists “failure to realize Spirent synergies, leading to goodwill impairment or ROIC falling near the cost of capital for a long period” as one of the signals that would overturn the investment thesis.

  • AI/data-center testing is currently the strongest, but also the most heavily discounted, “cycle-type second curve.” Second-quarter AI-related business doubled in size, and orders rose +56% year over year. Whether it can become a “sustainable successor” second curve rather than a single capex pulse is the most important question for the next three to five years. The report repeatedly warns that investors must distinguish “fundamental improvement” from “cyclical benefit.”

  An honest evaluation of the Baillie “second curve” dimension: Keysight’s strength is that its second curve is not a fantasy; it is visible in the financial statements today. Its weakness is that most of these curves are “extensions of the same platform into higher-value layers and adjacent testing subsegments.” They are strongly correlated with the core business, share the same customers and the same technology cycles, and lack a truly heterogeneous growth pole that is weakly correlated with the core business and can carry the flag independently when the core rolls over. In other words, it has successors, but the successors and the main runner are on the same track and will hit the same cyclical potholes together. That reduces the second curve’s value as “risk diversification and regeneration.”

  Jun 11, 2026
• QWhat is its core competitive advantage? Will this moat widen or narrow over the next three to five years?6/10

  A

  The core competitive advantage is the switching-cost and trust barrier created by the combination of “high-end testing platform + workflow embedding + global service relationships,” rather than network effects. Over the next three to five years, I judge the moat as broadly stable and slightly wider at the margin, but not “meaningfully wider.” This is a real, verifiable moat, but its width has a ceiling.

  What exactly forms the moat (mapped item by item to the report evidence):

  • Switching costs (medium-high, the main fortress): once test platforms, automation scripts, instrument software, measurement methods, calibration systems, engineering processes, and staff proficiency are embedded in customer R&D and production lines, migration friction is real. The company further thickens stickiness through software subscriptions, KeysightCare, calibration contracts, and extended warranties. The report describes this as “significantly higher than ordinary hardware, but weaker than the extremely high switching costs of ERP/databases.” That is an honest calibration.

  • Scale and R&D barriers (real): the report discloses more than $1 billion of fiscal 2025 R&D spending and $303 million of R&D in fiscal 2026 first quarter alone. Replicating Keysight’s breadth and update cadence across multiple testing subdomains would require many years and billions of dollars of investment. The report infers that replicating a platform with comparable credibility would typically take 5–10 years or more.

  • Brand and trust (in high-end testing, this is the question of “whether customers dare to entrust critical validation to it”): the company’s electronic measurement business traces back to 1939; it serves 40,000+ customers, and no single customer contributed more than 10% of revenue for three consecutive years from 2023–2025 (per the report). That diversity itself reflects moat depth.

  • Network effects (weak): the report explicitly states that this is not a social platform or an exchange; more customers do not directly make the product more valuable for each individual customer. It warns against misreading a broad ecosystem as strong network effects. This point is critical because it separates the company from true platform companies with network effects.

  The moat has “actually converted into profit,” which is the key difference from narrow-moat growth stocks. The latest quarter provides hard evidence: in fiscal 2026 second quarter, both the communications and electronic industrial segments reached 33% operating margins, improving year over year, with gross margins rising on both sides. Together with the report’s long-term 60%+ gross margin and low capital-intensity disclosure (2020–2025 capex about 2%–4% of revenue), this shows that pricing power and stickiness flow through to the income statement, not merely to moat language.

  Will the moat widen or narrow over the next three to five years? My judgment is “stable with modest widening,” supported by evidence in both directions:

  • Forces that slightly widen it: software/subscription/SaaS/integrated simulation are strengthening; Spirent (completed on 2025/10/15, about $1.46 billion), ESI, and Riscure extend capabilities into network testing, simulation, and security, deepening workflow embedding.
  • Forces that cap widening: the report names three erosion risks: more workloads moving to software-defined simulation/cloud testing could change how value is captured; geopolitical disruptions such as export controls and tariffs; and volatility in customer R&D and capex cadence. At the same time, the company itself acknowledges that outside trademarks, no single patent constitutes a “material single-point barrier.” That means the moat depends on continued investment in combined capabilities, not on a permanent legal barrier.

  Conclusion: this is a “real, monetized, upper-middle-width” moat, far more solid than narrow-moat growth stocks relying purely on narrative. But its width is constrained by “no network effects, patents not forming a single-point barrier, and co-exposure to technology migration and geopolitics.” Over the next three to five years, slow widening is more likely than a step change. On Baillie’s “moat” dimension, it deserves an upper-middle score, not the highest tier.

  Jun 11, 2026
• QIf its core business is disrupted, does it have the DNA to reinvent itself? How does it deal with mistakes and bad news?5/10

  A

  It has the DNA for “evolutionary self-reinvention,” but lacks a track record of rebuilding from ruins after being disrupted overnight. In dealing with mistakes and bad news, the company shows institutionalized, honest disclosure, but it has not yet gone through a true existential crisis. For this moat-type platform company, regeneration is “strong at continuous incremental iteration, untested in paradigm-level self-rescue.”

  Start with the positive evidence for self-reinvention. It does actively follow technology migration rather than clinging to an old form:

  • The business has steadily migrated from “selling hardware boxes” toward “software + subscriptions + services + integrated simulation,” a real evolution in form. The report discloses that services and other revenue rose from about 18.7% of revenue in 2020 to about 24.4% in 2025, and acquisitions such as ESI (simulation), Riscure (security), AnaPico, and Spirent (completed on 2025/10/15, about $1.46 billion, extending into network and protocol performance validation) have expanded its capabilities toward both the design front end and the network back end. This shows willingness and execution ability to actively reshape how it captures value.
  • Its reinvestment intensity in R&D is enough to support iteration: the report discloses more than $1 billion of R&D in fiscal 2025, an average employee tenure of 12.4 years, and a three-year average attrition rate of about 6.2%. In an industry highly dependent on measurement science and accumulated application engineering experience, low attrition is the base fuel for “reinvention capability.”

  But the “limits of self-reinvention” need an honest label: all of its reinvention has occurred while “the logic of its own platform has not been overturned”:

  • Keysight’s evolution is “extending the same test-and-measurement core into adjacent fields.” It has never experienced a crisis where the core paradigm was replaced externally overnight and it had to tear itself down and rebuild. It does not have a real-world record of “resurrection after core-business disruption,” so a more precise description of its “self-reinvention DNA” is “strong continuous incremental evolution, untested cliff-edge self-rescue.” The real threat named in the report, workloads migrating toward software-defined simulation/cloud testing and potentially changing the way value is captured, is exactly the test of this DNA, but the showdown has not yet arrived.

  Now look at “how it deals with mistakes and bad news,” which is a window into management honesty. The evidence is mostly positive:

  • It does not hide a downturn; it answers with cash flow: the report discloses fiscal 2024 revenue fell to $4.979 billion (−8.9%). The company did not use aggressive accounting cosmetics, and still reported $833 million of operating profit, $1.052 billion of operating cash flow, and about $898 million of free cash flow, “slower growth without bleeding.”
  • It actively dissects one-off disruptions: in its 2025 10-K, the company proactively explained the deferred tax effect related to Singapore tax-rate incentives in 2024, warning that 2024 GAAP net margin should not be treated as permanent earning power. In fiscal 2026 second quarter, it also proactively disclosed receivables and liability items related to refunds from the IEEPA tariff court ruling. Separating one-time benefits from fundamental improvement is a sign of honest management.
  • Governance mechanisms for correction and restraint: the report discloses a clawback policy, prohibitions on hedging and pledging, restrictions on grant windows, and PwC as auditor continuously since 2013. These all reduce the probability that bad news is systematically hidden.

  Conclusion: on Baillie’s “self-reinvention DNA + how it deals with mistakes” dimension, Keysight earns an “upper-middle” result. Its evolutionary reinvention capability is real and visible, and its disclosure around bad news is transparent and measured. But it has no real-world record of rising from the ashes after disruption, so cliff-edge regeneration is “untested,” not “proven,” and does not merit the highest tier.

  Jun 11, 2026
• QDoes management, especially the founders, have a long-term view and deeply aligned interests with the company? Is it willing to sacrifice current profit for five to ten years out?4/10

  A

  Management has a standardized long-term governance framework and a clear long-term orientation, but this is “medium-strength alignment created by professional managers through compensation, ownership requirements, and governance mechanisms,” not the ultra-strong alignment Baillie values most, where a founder has a large amount of real personal wealth at stake. Is it willing to sacrifice current profit for five to ten years out? The institutional incentives point to the long term, but it lacks the founder-style internal drive of “personal wealth and company fate bound together.”

  Start with the structural fact of “no founder alignment.” Keysight was spun out of Agilent in 2014 and is run by a professional management team, with no founder/controlling shareholder. The report discloses that as of January 20, 2026, CEO Satish Dhanasekaran beneficially owned about 46,092 shares, and CFO Neil Dougherty about 93,508 shares. At the current share price, Keysight was around $324 (June 10, 2026 close), implying CEO holdings of about $15 million and CFO holdings of about $30 million. These are not small absolute sums, but relative to total shares of about 171 million, the percentages are extremely low (CEO about 0.027%). That is far from the Baillie ideal of “a founder owning 30% and being locked in with the company.” It is what the report calls “medium-strength alignment.”

  Now the positive institutional evidence for long-term orientation. The framework is indeed designed for the long term:

  • Mandatory share ownership requirements: the report discloses that the CEO must hold 6 times salary, the CFO/COO 3 times salary or 80,000 shares, and other executives 3 times salary or 40,000 shares. As of fiscal 2025 year-end, all NEOs met the ownership requirements. This institutionally ties executive wealth to the share price.
  • Compensation tied to long-term metrics: the report discloses that performance compensation is linked to revenue, EPS, operating margin, TSR (total shareholder return), and other measures, with a clawback policy, prohibitions on hedging and pledging, and restrictions on grant windows. These all suppress short-term arbitrage and point toward long-term value.

  “Is it willing to sacrifice current profit for five to ten years out?” The behavioral evidence is “yes, but more like rational capital allocation than a founder-style all-in bet”:

  • Sustained high R&D reinvestment: the company is willing to pressure short-term margin to preserve platform leadership. The report discloses more than $1 billion of R&D in fiscal 2025 and $303 million in fiscal 2026 first quarter alone, with capital and expenses continuously directed toward capability building.
  • Countercyclical acquisitions as bets on the future: Spirent was acquired on October 15, 2025, for about $1.46 billion, on top of ESI, Riscure, and others. These are payments for capability boundaries three to five years out. But the report also warns that this has increased goodwill and intangible assets, and that the “value proof over the next three years is not fully complete,” rating capital allocation as “rational and somewhat proactive, but not beyond doubt.”
  • Buybacks look rational rather than masterful market timing: the report discloses diluted shares fell from about 189 million to about 173 million in the first half of fiscal 2026, and recent buyback prices were broadly in the $140–170 range. In hindsight, they were well below today’s share price, so historical buybacks created value for shareholders. But this is disciplined capital return, not a sacrifice of the present for the long term.

  Conclusion: on Baillie’s “management long-term view and interest alignment” dimension, this is “middle to upper-middle.” The governance framework is disciplined, long-term incentives are in place, and there are real actions investing for the future. But there is no founder, executive ownership is extremely low as a percentage of the company, and alignment is institution-driven and medium-strength, not the top-tier Baillie pattern of “a founder putting personal wealth on the line and sharing the company’s fate ten years out.” The report’s corresponding score of 3.5/5 is consistent with this.

  Jun 11, 2026
• QIf it disappeared tomorrow, how much would customers miss it? Is its growth model sustainable, without depending on harm to society or regulatory arbitrage?6/10

  A

  If it disappeared tomorrow, core high-end customers would “miss it considerably, and would struggle to replace it seamlessly in the short term,” but it would not be so indispensable that “the entire industry stops.” Its growth model is highly sustainable and basically does not depend on harming society or crossing regulatory red lines. This dimension is clean and relatively strong. This question needs to be split into “indispensability” and “social/regulatory sustainability,” and answered honestly for each.

  Subquestion one: how indispensable is it? “Hard to replace in critical links, but not a sole source.”

  • The hard-to-replace part is real: once test platforms, automation scripts, measurement methods, and calibration systems are embedded in customers’ R&D and production-line processes, migration friction is high (the report characterizes switching costs as “medium-high”). In the highest-end areas such as RF/microwave, optical, high-speed digital, 6G pre-research, and defense validation, customers need “accurate measurement, fast measurement, and credible sign-off.” The report frames the brand as a question of “whether you dare to hand critical validation tasks to it.” Customers are deeply dependent on its precision and credibility, and short-term switching costs and risks are high. That is the source of “they would miss it a lot.”
  • But it does have substitutes: the report clearly states that Keysight has no perfectly overlapping one-to-one competitor, but there are partial competitors/comparables in different subsegments: VIAVI in communications network testing, Teradyne in semiconductor ATE, Fortive and AMETEK in general industrial instruments, with Asian competitors also creating pricing pressure (the report cites VIAVI risk disclosure). In other words, if customers were forced to leave Keysight, it would be painful and delay projects, but the industry would not be paralyzed. This is “high dependence,” not “irreplaceable.”
  • Hard evidence of dependence: the report discloses that the company serves 40,000+ customers annually and has had no single customer contributing more than 10% of revenue for three consecutive years. Being continuously chosen by such a broad customer base is itself a quantification of “how much customers would miss it.” It also shows that indispensability is spread across a massive customer base rather than tied to a few lifeline customers.

  Subquestion two: is the growth model sustainable, and does it harm society or regulation? This dimension is very clean.

  • It does not grow by harming society: it sells test and validation capabilities that make communications, semiconductors, automobiles, aerospace and defense, energy, and other systems more reliable and safer. It is essentially a “picks-and-shovels” business that improves technology quality and safety. It does not have typical ESG controversies such as data misuse, addictive monetization, or externality dumping. Stronger growth often means downstream electronic systems are being more thoroughly validated. Social benefit and commercial interest point in the same direction.
  • Sustainability comes from structure, not overextension: the report discloses long-term 60%+ gross margin, low capex (about 2%–4% of revenue from 2020–2025), and cash-flow resilience from a rising service/subscription mix. In the latest quarter, both major segments reached 33% operating margin and gross margins rose on both sides. This is a sustainable model that earns through technology and stickiness, not growth propped up by subsidies, cash burn, or aggressive expansion.
  • Regulatory exposure is “passive compliance risk,” not “active line-crossing”: the one shadow that must be honestly noted is geopolitics. The report discloses that tariffs, export controls, and US-China trade frictions affected fiscal 2025 results, and in fiscal 2026 second quarter the company recognized refund-related receivables and liabilities due to the IEEPA tariff court ruling. But this is “business affected by the external regulatory environment,” not “growth through regulatory avoidance or harm to the public interest.” Its nature is compliance and geopolitical risk that needs monitoring, not a moral flaw in the business model.

  Conclusion: on Baillie’s “indispensability + social/regulatory sustainability” dimension, this is “relatively strong.” Customers would truly miss it, critical links are hard to replace in the short term (though it is not a sole source), and the growth model is clean, aligned with social benefits, and does not depend on harming others. The only long-term item to watch is passive geopolitical compliance exposure such as export controls/tariffs, not active misconduct.

  Jun 11, 2026
• QWhat are the unit economics of this business (gross margin, incremental returns)? Do they improve or deteriorate as scale increases? Where does the money it earns go?7/10

  A

  The unit economics are excellent: high gross margin, low capex, and strong incremental cash conversion. As scale increases, the business overall “gets better,” especially as the software/services share rises. The money it earns is mainly spent on R&D, acquisitions, and buybacks. This is Keysight’s strongest item and the core basis for calling it a “high-quality business.”

  Gross margin and earnings quality (structurally excellent):

  • The report discloses a long-term 60%+ gross margin and a fiscal 2025 gross margin of about 62.1%; in the latest quarter (fiscal 2026 second quarter), both Communications and Electronic Industrial segments reached 33% operating margins, with gross margins rising on both sides. This is a “technical premium” earned from accurate measurement, deep hardware/software integration, and strong service stickiness, not from volume-driven price cutting.
  • It is a company with “heavy expense-type reinvestment, light capex-type reinvestment”: the report discloses that capex was about 2%–4% of revenue from 2020–2025 (only $128 million in fiscal 2025, about 2.4%). Growth mainly depends on R&D and acquisitions rather than factory expansion. That means incremental revenue does not require proportional heavy-asset investment, so incremental returns are naturally high.

  Incremental returns and whether “bigger gets better or worse”:

  • The answer is “overall bigger gets better, especially the software/services layer.” The report discloses that services and other revenue grew from $789 million in 2020 to $1.312 billion in 2025, and its share rose to about 24.4%, with deferred revenue of $884 million. This part has low marginal cost and helps stabilize gross margin and cash flow. The larger its scale and share, the better the quality of earnings.
  • Cash conversion is “the more it grows, the more cash it produces”: the report discloses cumulative net income of about $5.166 billion from 2020–2025, cumulative operating cash flow of about $7.351 billion, and cumulative free cash flow of about $6.396 billion. Cash flow has been significantly above net income over the long term, a classic business where “accounting profit is good and cash profit is even better.” Latest-quarter free cash flow reached a record $472 million.
  • But there is an honest boundary to “bigger gets better”: scale returns are affected by the cycle and acquisition dilution. The report discloses that ROIC was very attractive in 2022–2023 but fell back to about 12%–13% in 2024–2025 (an analytical estimate based on GAAP operating profit, effective tax rate, and net operating invested capital, not an official company metric). ROE was about 15.5% in 2025, and ROA about 8.3%. These are still good levels, but not “extremely excellent today,” and goodwill and intangible assets from acquisitions have diluted book capital returns. So unit economics are extremely strong at the “operating layer,” and good rather than top-tier at the “invested-capital return including acquisitions” layer.

  Where the money goes (capital allocation):

  • R&D: more than $1 billion in fiscal 2025 and $303 million in fiscal 2026 first quarter alone (per the report), to preserve generational platform leadership.
  • Acquisitions: Spirent completed on 2025/10/15 for about $1.46 billion, together with ESI, Riscure, and others, extending into adjacent testing subsegments. Spirent synergies are expected to bring 150–200 basis points of operating-margin improvement by fiscal 2026, but integration delivery remains to be proven.
  • Buybacks: the report discloses diluted shares fell from about 189 million to about 173 million in the first half of fiscal 2026. Recent buyback prices were around $140–170, and in hindsight clearly below today’s share price. This was disciplined capital return that created value for shareholders.

  Conclusion: on Baillie’s “unit economics and incremental returns” dimension, this scores “strong.” Gross margin is structurally high, capex is extremely light, incremental cash conversion is excellent, and the rising software/services share makes scale effects positive. The only discounts are that acquisition-diluted invested-capital returns are good rather than top-tier, and the cycle will make the return center fluctuate. The money goes to R&D, acquisitions, and buybacks. The direction is rational, historical buyback timing has been good, but current acquisition synergy delivery remains an unresolved verification point.

  Jun 11, 2026
• QWhat conditions must hold simultaneously for it to rise fivefold in ten years? Are those conditions realistic? What expectations are implied in today’s share price?3/10

  A

  For the stock to rise fivefold in ten years (about a 17.5% annualized share-price return), three things must hold for a long time at the same time: high growth, margin expansion, and no major compression in the valuation multiple. This combination is not impossible, but each item is optimistic, and the combined probability is uncomfortable for conservative investors. Today’s share price already embeds the most important expectation: that the market will be willing to pay a high-quality premium multiple for it over the long term.

  Break down the conditions needed for “fivefold” and compare each with reality:

  • Condition one: earnings must rise about 5 times over ten years if the multiple is unchanged, or more if the multiple compresses. 5 times earnings growth is roughly a 17% EPS CAGR. The report discloses that the company’s self-reported long-term core growth target is only 5%–7%, and its serviceable market’s long-term growth is 4%–6% (2023 Investor Day management estimate). To generate a 17% EPS CAGR, the revenue center alone is far from enough. It must be supplemented by sustained margin expansion, continued buybacks, and incremental acquisitions. Realism assessment: medium-low. The report’s own three-scenario assumptions for first-10-year growth are conservative 4%, base 7%, and optimistic 10%; even the optimistic case is below the 17% needed for fivefold.
  • Condition two: margins must structurally step up and stay there. The current cycle has pushed margins very high (both major segments reached 33% operating margin in the latest quarter), but the report discloses that fiscal 2024 operating margin once fell back to 16.7%, showing that part of the high margin is cyclical benefit. To achieve fivefold, the company must turn the current high margin into a sustainable center of gravity and actually deliver Spirent’s 150–200 basis points of synergy. Realism assessment: uncertain; integration delivery is a core verification point named by the report.
  • Condition three: the valuation multiple must not fall back toward “ordinary high-quality industrial” levels. This is the most fragile part. Keysight’s current price is about $324 (June 10, 2026 close), market cap about $55.3 billion, trailing PE about 53 times, and forward PE about 30 times. The report repeatedly stresses that “the most fragile assumption is not one or two years of growth, but whether the market will forever be willing to pay a 40–50 times cash-flow multiple for it.” If the multiple returns to the normal range for high-quality industrials in ten years (20–25 times), even severalfold earnings growth would be heavily offset by multiple compression. Realism assessment: low. Sustaining a high multiple for ten years is itself a strong assumption.

  What expectations are implied in today’s share price? This is the core of the question:

  • The report’s owner-earnings discount method can reverse-engineer market expectations. The report’s three-scenario intrinsic-value estimates are: conservative about $107 (4% growth in the first ten years), base about $156 (7%), and optimistic about $259 (10%). The report gives a reasonable intrinsic-value range of $150–210. The current price of about $324 is already about 25% above the report’s optimistic-scenario intrinsic value (about $259). In other words, market pricing is even more optimistic than the optimistic assumption of “10% growth in the first ten years + 8% discount rate.”
  • In plain English: buying today already assumes the full optimistic script of “excellent double-digit-type growth for many years, delivered margin expansion, and a market that keeps assigning a high valuation multiple.” On that basis, the report measures the current price premium over reasonable intrinsic value at about 69%–137% (relative to the base range). TTM free-cash-flow yield is only about 2.2%, and conservative owner-earnings yield is even below 2%, while the contemporaneous US 10-year Treasury yield was about 4.50% (the report cites data as of May 26, 2026). Buying means actively giving up a higher risk-free yield to bet on that script.
  • One detail needs an honest update: the report’s snapshot price was $355.74, with a trailing PE of about 63.9 times; by 2026/6/10, the price had fallen about 9% to $324, and trailing PE had fallen to about 53 times. Some valuation has been digested, but the price is still meaningfully above the upper end of the report’s reasonable intrinsic-value range ($210). The conclusion is unchanged: expensive, and expensive because of the multiple.

  Conclusion: on Baillie’s dimension of “what conditions must hold for a fivefold return in ten years + what expectations are implied in today’s price,” the honest judgment is that the probability of the required combination of “high growth × margin expansion × no high-multiple compression” holding for a long time is low. Even the report’s optimistic scenario does not reach the required growth rate, while the current price has already priced in expectations beyond that optimistic scenario. This is not a “bad company” issue. It is that “the price has prepaid for nearly perfect execution over the next ten years,” leaving a negative margin of safety.

  Jun 11, 2026
• QWhy has the market not realized all this yet? Is it hard to understand, overlooked, or simply too long-term? What will become the “narrative inflection point”?3/10

  A

  The market has in fact “understood it and respects it.” This is not a buried gem. Quite the opposite: the market has already fully, and perhaps excessively, recognized its quality and assigned a high premium accordingly. So the honest answer for Keysight on this question is that “there is no unrecognized perception gap; instead, there is a risk of overpricing.” This is the mirror image of Baillie’s ideal of a “dust-covered growth stock that the market does not understand, underestimates, or cannot look far enough to see.” That point must be stated honestly; one should not force a “the market has not seen it” story onto a growth narrative.

  Evidence one: the market has already recognized it; it is not “hard to understand.”

  • Sell-side coverage is dense, sentiment is warm, and pricing is not low: 13 analysts give a consensus “Buy” rating, the average 12-month target price is about $383, the current price is about $324 (June 10, 2026 close), and trailing PE is about 53 times. A stock covered by 13 institutions, with bullish consensus and trading above 50 times trailing earnings, is clearly not an “obscure stock nobody understands.” The market has priced in its high-end testing platform position, cash-flow quality, and AI tailwind.
  • The report’s own judgment is consistent with this: current valuation “already prices it as a high-quality, sustainably growing asset that will maintain a high valuation for many years.” The issue is “not whether it is a good company, but whether it is a good price today.” This is exactly the picture of “the market has understood it and paid a high price.”

  Evidence two: the so-called “not looking far enough” dynamic is actually reversed. The market may be “looking too far and too optimistically.”

  • The current price is already about 25% above the report’s optimistic-scenario intrinsic value (about $259), and meaningfully above the upper end of reasonable intrinsic value ($210). The report identifies the most fragile assumption as “whether the market will always be willing to pay a 40–50 times cash-flow multiple for it.” That means the risk is not that “the market has not realized it will improve,” but that “the market has already priced it as staying excellent forever and always deserving a high multiple.”

  So what would the real “narrative inflection point” be? For Keysight, the inflection is more likely to be a “downward correction” than an “upward discovery.” Both directions should be listed honestly:

  • Downside inflection points (more realistic and more likely): ① the AI/data-center capex cycle peaks; once AI-related business doubling and orders +56%, which currently drive high growth, normalize, the market will re-examine whether “high 20% growth” can be extrapolated (the report notes that second-quarter high growth mixes an upturn in the cycle with accounting effects from a one-time tariff refund); ② acquisition synergies from Spirent (about $1.46 billion) and others fail to materialize, leading to goodwill impairment; ③ margins fall from current highs; ④ valuation multiples converge from 40–50 times cash flow toward high-quality industrial norms (20–25 times). Any one of these would trigger multiple revision. The report’s bear case suggests a medium-to-long-term share-price drawdown of 50%–70%.
  • Upside inflection points (possible, but partly priced in): if the software/services share continues rising and turns cash-flow stability into something “SaaS-like,” Spirent synergies beat expectations, and the long-term non-GAAP operating-margin target is achieved, the market may be willing to re-rate it from a “cyclical test-equipment vendor” into a “higher-quality platform company.” But the report judges that even this would struggle to push reasonable intrinsic value above $355. Upward room has already been heavily consumed by the current high price.

  Conclusion: on Baillie’s “why has the market not realized it + narrative inflection point” dimension, Keysight scores low, not because it is bad, but because it does not fit the premise of the question. The market is not failing to understand, disrespecting, or looking too short-term; instead, it has already fully recognized the quality and assigned a high premium. The perception gap is neutral to negative, as the market may be too optimistic. The true narrative inflection point is more likely to be a “downward re-rating” triggered by an AI-cycle fade or failed acquisition integration than an “upside surprise” waiting to be discovered. This is also the fundamental reason the report gives it “Watch” rather than “Buy”: the good company’s good story is already embedded in the price.

  Jun 11, 2026