Rockwell Automation, Inc. (NYSE: ROK) — The Reshoring Crown Jewel, Priced for a Recovery It Has Only Begun to Deliver

Independent fundamental research. No recommendation and no price target appears in the analysis body; the single exception is the clearly-labeled “Author’s Take” block below, which is the author’s own subjective opinion. This is general information, not investment advice.

Report date: 2026-06-19 · Price (6/18/26): $473.79 · Market cap: ~$53.1B · Enterprise value: ~$56B · Fiscal year-end: September 30 · CIK: 0001024478

⚡ Author’s Take

This block is the author’s own subjective opinion. It is general information and not investment advice. The analysis that follows takes no position and carries no price target.

Verdict: HOLD / AVOID-here for new capital; not-a-short; accumulate-on-weakness toward ~$340–390 (≈27–30x forward adjusted EPS / ≈4.5x EV-sales). Medium conviction.

Rockwell Automation is the best business I have looked at in the multi-industrial cohort and one of the most richly priced. It is the purest large-cap way to own North American factory automation — the Allen-Bradley/Logix installed base is a genuine, decades-deep moat that shows up exactly where a real moat should: a Software & Control segment earning 30–35% operating margins and a Logix controller franchise that grew 20%+ even through a two-year industrial down-cycle. About 65% of sales move through a near-exclusive North American distributor channel that competitors cannot replicate, and ROK is the cleanest beneficiary of US reshoring, the data-center build-out (data-center sales more than doubled year-over-year in Q2-FY26), and “physical AI”/robotics. That is the bull case, and it is real.

The problem is the price and the timing. The stock has nearly tripled off its June-2022 low of ~$179 to an all-time high of $473.79 — and on its own multi-year history it has never been more expensive on sales (P/S ≈99.98th percentile, ~6.3x EV/sales) and sits in the 93rd percentile on earnings. You are paying ~37x forward FY26 adjusted EPS (~$12.80), ~25x EV/EBITDA, ~6x sales, and a ~2.7% free-cash-flow yield with a 1.1% dividend, for a business that grows mid-single-digit organically through the cycle (revenue actually fell ~8% from the FY23 peak to FY25, and FY25 organic growth was +1% — entirely price, with volume down 2%) and earns a ~15% ROIC. The market is capitalizing the early innings of a cyclical recovery plus a secular reshoring/AI acceleration as if both are permanent and compounding. The tape agrees with the bulls — beta ~1.28, a strong momentum and dividend-yield loading, a y1 return of ~+49% at a 1.6 Sharpe, no value loading — this is a crowded, fully-believed momentum winner, not a falling knife and not a coiled spring. It has already had its move.

So this is the EMR problem in a sharper form: a better franchise than Emerson in its North American niche, but priced even more aggressively on sales, on earnings that are recovering off a real trough. The honest framing is quality-compounder-at-the-wrong-price: I would happily own Rockwell, just not 60% above where the recovery was being given away a year ago. Conviction: medium. Flips bullish on a 20–25% drawdown that resets the multiple toward ~28x forward EPS while orders/book-to-bill and data-center/reshoring momentum hold; flips bearish if the capex unlock in its two biggest verticals (automotive, consumer packaged goods) keeps stalling on tariff/USMCA uncertainty and the short-cycle recovery rolls over into a second air-pocket, leaving 37x forward earnings stranded on flat volumes. Tag: you want to own the factory floor — you don’t want to pay the top tick for it.

📈 Stock Price Action — Five-Year Event Map

Over five years Rockwell round-tripped and then broke out: from the low-$300s in late 2021 it crashed to a 5-year low of ~$178.60 on 17-Jun-2022 in the rate-shock bear market, ground sideways through a two-year industrial destocking down-cycle (revisiting the low-$240s in 2022 and 2024), then ripped ~+165% to an all-time-high $473.79 on 18-Jun-2026 as the cyclical recovery, reshoring, and the data-center/AI vertical converged. It now trades at the very top of its range, ~$474, against a 52-week range of ~$317–$474, comfortably above its 21-/50-/200-day EMAs (~$453/$435/$392). The arc is not an earnings melt-up — FY25 revenue ($8.34B) is still below the FY23 peak ($9.06B); the multiple did most of the work.

Cycle narrative. (1) ROK entered the period as a beloved reshoring/automation play, closing its largest-ever acquisition (Plex) and re-rating into late-2021 euphoria. (2) It then fell almost in half in H1-2022 — a pure rate-driven de-rating of a high-multiple industrial, made worse by component shortages, bottoming near $179. (3) From that low it nearly doubled over a year as the CHIPS Act/reshoring narrative and a supply-constrained, record backlog drove a sharp recovery to ~$348. (4) The destocking down-cycle then bit: distributors and OEMs worked down inventory, orders softened, FY24 guidance was cut repeatedly, and revenue rolled ~8% off the FY23 peak — the stock gave back ~30% to the low-$240s. (5) Through FY25 the cycle troughed; management’s productivity/self-help program (incrementals lifted toward 50%) defended margins even on falling volume, and the stock recovered to ~$317. (6) The dominant, most recent leg: a ~50% run to June-2026’s all-time high as the recovery was confirmed (organic +9% in Q2-FY26, Logix +20%, data-center sales more than doubling), reshoring momentum built, and management raised FY26 adjusted-EPS guidance twice (to ~$12.80). DA Davidson initiated Neutral on 16-Jun-2026 — a fair marker of where the debate now sits: a great business that has already re-rated. Price moves are Fact; attributed causes are Interpretation.

1. Executive Summary

Rockwell Automation is the world’s largest company dedicated entirely to industrial automation and digital transformation — and the undisputed #1 in North American discrete (factory) automation. Founded in 1903, headquartered in Milwaukee, it sells the controllers, drives, motion, sensing, safety, visualization software, and lifecycle services that run factories and plants: the Allen-Bradley hardware brand, the Logix/ControlLogix/CompactLogix programmable controllers at the heart of its “Integrated Architecture,” the FactoryTalk software suite, and a growing software/SaaS layer (Plex MES, Fiix maintenance, Verve OT-cybersecurity) plus robotics (OTTO Motors autonomous mobile robots). FY2025 (ended 30-Sep-2025) revenue was $8.34B across three segments — Intelligent Devices (45%), Software & Control (29%), and Lifecycle Services (26%) — with ~63% of sales in North America and roughly 65% transacted through independent distributors (the two largest ~20% of total).

The business quality is genuinely high. Rockwell’s moat is a textbook case of customer captivity (switching costs) reinforced by a regional distribution/scale advantage and brand (see the Competitive Position section): once a plant is engineered around Allen-Bradley hardware and Logix control code, the cost and risk of switching — re-coding, re-validation (critical in pharma/food), operator retraining, downtime on a 15–30-year asset — make the installed base extraordinarily sticky. The financial fingerprint is unmistakable: a 48% gross margin, a Software & Control segment at 30–35% operating margins, ~15% ROIC, ~$1.4–1.5B free cash flow (≈17–18% of sales) on light ~2.5%-of-sales capex, and Logix growing 20%+ even during a down-cycle — pricing power and share that only entrenchment produces. Capital allocation is disciplined and shareholder-friendly: a ~16-year dividend-increase streak (Dividend Aristocrat, ~5% DPS CAGR, ~44% of FCF), steady buybacks that shrink the share count ~1%/year, low leverage (~$2.15–2.8B net debt, ~1x EBITDA), and a sensible bolt-on M&A program (Plex, Fiix, Clearpath/OTTO, Verve) — marred by one clear blemish, the impaired-and-now-dissolved Sensia oil-&-gas JV with SLB.

The problem is entirely valuation and cycle-timing. ROK has just emerged from a brutal two-year destocking down-cycle — revenue fell ~8% from the FY23 peak ($9.06B) to FY25 ($8.34B), and FY25 organic growth was +1%, all price, with volume down 2%. A cyclical recovery is now clearly underway (Q2-FY26 organic +9%, two guide raises, FY26 adjusted EPS guided to ~$12.80). But the stock has nearly tripled off its 2022 low to an all-time high, and on its own multi-year history it has never been more expensive on sales (~99.98th percentile, ~6.3x EV/sales) and is near the top on earnings (P/E ~93rd percentile). At $474 you pay ~37x forward FY26 adjusted EPS, ~25x EV/EBITDA, ~6x sales, a ~2.7% FCF yield and a 1.1% dividend for a business whose through-cycle algorithm is mid-single-digit organic growth + ~1% M&A + margin self-help + buyback → low-double-digit EPS. The market is capitalizing the early innings of a recovery and the reshoring/data-center secular story as if they compound indefinitely.

The investment question is not whether Rockwell is a good business — it is one of the best in industrials — but whether ~37x forward earnings and ~6x sales correctly price a mid-single-digit organic grower with a ~15% ROIC at the start of a cyclical recovery, or whether the market has extrapolated a reshoring/AI re-rate that the through-cycle unit economics do not yet justify. This memo argues the moat is real and durable in its North American discrete core (and thin internationally and in process), the recovery is genuine but early, and the valuation embeds an optimistic, hard-to-falsify acceleration that leaves little margin of safety and poor risk/reward for new capital at the all-time high.

2. Business Overview

Rockwell Automation sells the hardware, software, and services that automate, control, and digitize industrial production. Its customers are manufacturers and plant operators — automotive, semiconductor, e-commerce/warehousing, food & beverage, life sciences, household & personal care, tire, oil & gas, mining, metals, chemicals, pulp & paper, water/wastewater, and increasingly data centers. The value proposition is throughput, uptime, safety, quality, labor productivity, and (lately) cybersecurity and AI-readiness on production assets that run for decades. Because the cost of a control failure (an unplanned line stoppage, an off-spec or non-compliant batch, a safety incident) dwarfs the cost of Rockwell’s content, the installed base is sticky and the franchise is aftermarket- and modernization-rich.

Reporting structure (FY2025). Rockwell reports three segments:

Source: FY2025 10-K (rok-20250930), Item 7 MD&A. Total segment operating margin shown; enterprise operating margin (which deducts ~$110M corporate) is lower — ROK began reporting “enterprise operating margin” in Q2-FY26 per SEC non-GAAP rules.

• Intelligent Devices (45%) — drives (PowerFlex), motion, advanced material handling (the OTTO/Clearpath autonomous mobile robots), safety, sensing, industrial components, and configured-to-order products. The hardware engine; lower-margin (~18%) and the most cyclically and competitively exposed part of the portfolio.
• Software & Control (29%) — the Logix controllers (ControlLogix/CompactLogix) at the core of the Integrated Architecture, control/visualization software and hardware, network and security infrastructure, digital-twin/simulation, and information software (FactoryTalk, Plex MES, Fiix CMMS). This is the moat engine — ~30% segment margin (34.9% in Q2-FY26) and the home of Logix’s 20%+ growth and the software/ARR story.
• Lifecycle Services (26%) — digital consulting (Kalypso), engineered-to-order solutions/professional services, and recurring/connected services (cybersecurity, safety, remote monitoring, asset management), plus the former Sensia JV (dissolved 1-Apr-2026). The longest-cycle, most project-dependent, lowest-margin (~14–16%) segment.

Revenue model. Roughly 65% of sales flow through independent distributors (the two largest ~20% of total) — a structurally important channel in North America that competitors do not have. Revenue is recognized as “Products & Solutions” vs “Services”; ROK does not break out software-vs-hardware dollars in its filings. Annual recurring revenue (ARR) is the strategic recurring layer — management does not disclose an absolute ARR dollar figure in the 10-K, only that it grew “over 6%” in Q2-FY26 (high-single-digit software ARR, mid-single-digit services) and is guided to high-single-digit growth in FY26 — below the long-stated “double-digit ARR” aspiration. Geographically, FY25 was ~63% North America, ~18% EMEA, ~12% Asia-Pacific, ~7% Latin America, with North America the only region that grew.

Verdict. A high-quality, installed-base-anchored discrete-automation franchise with a real recurring/software upgrade in progress, genuine pricing power (+3 pts price in FY25 and Q2-FY26), and a dominant North American channel — but one that is more hardware-led, more cyclical, and more regionally concentrated than the “software/Connected Enterprise” narrative implies. The economics are real; the disclosure flatters the software mix; the debate is entirely price (see the Valuation section).

3. Industry Dynamics

The industrial-automation industry is structurally attractive for entrenched leaders and brutally cyclical for everyone, and Rockwell sits at the favorable end of that spectrum in its North American core.

Market structure — a high-barrier oligopoly with regional fiefdoms. Global industrial automation is dominated by a handful of scaled players — Siemens (the global #1, Digital Industries), Rockwell, Schneider Electric, ABB, Mitsubishi Electric, Emerson, Honeywell, plus motion/robotics specialists (Omron, Yaskawa, Fanuc) and software entrants (Dassault Systèmes). Crucially, the market is regionally segmented: Rockwell is #1 in North American discrete automation; Siemens owns Europe; the Japanese and European majors are stronger in Asia. This regionalization is itself a barrier — local distributor density, installed base, and engineering talent are hard to replicate across borders, which is why Rockwell’s ~63% North America concentration is both its strength (dominant where it’s dense) and its ceiling (a distant challenger elsewhere).

Three features make the core of this industry one of the better structures in industrials:

1. Switching costs are extreme. A control platform runs a factory or plant for 15–30 years. Ripping out Allen-Bradley/Logix means re-engineering control code (Studio 5000 / ladder logic), re-validating safety and (in pharma/food) regulatory compliance, retraining operators, and risking downtime on an asset that earns continuously. Operators almost never switch mid-life; they expand and modernize on the incumbent. This produces decades-long annuity revenue per installed system.
2. Entry barriers are qualification- and trust-based, not merely capital-based. A new control platform must clear functional-safety certifications, earn reference installations, and win the confidence of plant managers who will not bet uptime on an unproven vendor. There has been no successful new global discrete-automation entrant in decades.
3. The channel and installed base compound. Rockwell’s near-exclusive North American distributor network (~65% of sales) provides local inventory, engineering, and support that deepen with density — a two-sided advantage competitors cannot easily attack.

The cyclical overlay — and where we are in the capital cycle. Demand is tied to manufacturing capital spending and short-cycle production levels, which track PMI, industrial production, and capacity utilization. Rockwell just lived through the bear phase: a 2023–2024 destocking down-cycle in which revenue fell ~8% from the FY23 peak as distributors and OEMs worked down inventory built during the supply-constrained 2021–2022 period. FY25 PMI sat below 50 for much of the year. The current upturn — short-cycle products recovering first (Q2-FY26 organic +9%, products especially strong), with long-cycle projects (automotive, CPG) still gated by tariff/USMCA uncertainty — is an early-cycle recovery, not yet a secular acceleration. In Marathon capital-cycle terms, the supply side is rational (a stable oligopoly, no capacity flood), but the valuation is at a multi-year peak precisely as the demand recovery is being extrapolated — the classic moment to demand a margin of safety.

The secular overlay — real tailwinds, early monetization. Three genuine secular vectors support the bull case: reshoring of manufacturing to North America (Rockwell’s home turf, its single best structural advantage); the data-center/AI build-out (Cubic power distribution, Logix replacing commercial-grade controls in AI data centers, drives into chillers/HVAC — data-center sales “more than doubled” in Q2-FY26 but are still only low-single-digit % of sales); and “physical AI”/robotics and software-defined automation (OTTO AMRs, FactoryTalk + AI). These are real and Rockwell is well-positioned — but each is early and small relative to the $8.9B base, and the multiple already capitalizes them.

Software as the contested frontier. The battle is moving up the stack into MES, digital twins, asset performance, and OT-cybersecurity, where Rockwell (Plex, Fiix, Verve, FactoryTalk) competes with Siemens (Xcelerator), Schneider/AVEVA, Dassault, and PTC. The bull case is that Rockwell uniquely couples software with the largest North American discrete installed base; the skeptic notes ARR is growing high-single-digit (below the double-digit goal) and that true software remains a minority of revenue.

Verdict: structurally good industry for the regional leader — high barriers, installed-base annuity, rational supply — with a pronounced cyclical-capex overlay and a real but early secular tailwind. Rockwell is excellently positioned in the attractive (North American discrete) part. The industry quality is not in doubt; the open questions (see the Risk and Valuation sections) are cyclical timing and price.

4. Competitive Position

The moat is real, financially evidenced, and concentrated — wide and deep in North American discrete automation, thin internationally and in process. In Greenwald’s taxonomy, Rockwell’s advantage is customer captivity (switching costs) reinforced by economies of scale within a region (distribution density) and intangible brand — not a uniform, global moat across everything it sells.

Where the moat is strongest — Logix / Integrated Architecture (Software & Control). Rockwell’s defining differentiator, in its own words, is that it is “the only automation provider that can support many production disciplines — discrete, process, batch, safety, security, motion, robotics, power control — in a single hardware and software environment,” with Logix at its core. Once a plant standardizes on Logix, the control code, operator training, validated configurations, and the surrounding Allen-Bradley hardware all flow to Rockwell for the asset’s multi-decade life. The financial fingerprint of this captivity is the Software & Control segment’s ~30% (Q2-FY26: 34.9%) operating margin and Logix growing 20%+ during a down-cycle — outcomes that would deteriorate immediately if the captivity were not real. This is the part of Rockwell that deserves a premium.

Where the moat is reinforced — the North American distributor channel. ~65% of global sales move through independent distributors, and the two largest are ~20% of total sales. These distributors are effectively single-line (they carry Allen-Bradley, not Siemens), provide local engineering and inventory, and create a density of technical support that is self-reinforcing and very hard for a competitor to replicate from scratch in North America. This is a genuine local-scale/distribution moat — and simultaneously a concentration risk (see the Changes and Headwinds section) and the reason the moat is regional: that channel density does not exist for Rockwell in Europe or Asia, where Siemens owns the relationships.

Where the moat is brand/intangible. Allen-Bradley (the de-facto North American factory-floor standard since the early 20th century), ControlLogix, CompactLogix, PowerFlex, FactoryTalk, PlantPAx, and Connected Enterprise are entrenched marks. Notably, Rockwell states patents are not the moat (“we do not believe that loss or termination of any one of them would materially affect our business”) — confirming the advantage is installed base and channel, not raw IP. R&D is a moderate ~5.8% of sales (~$482M FY25; ~8% including engineering), consistent with a hardware-and-installed-base franchise rather than a technology-lead one.

Greenwald tests. (1) Market-share stability: the discrete-automation oligopoly has been stable for decades; Rockwell has held #1 in North America, with no successful new global entrant. Pass. (2) ROIC test: Rockwell earns ~15% ROIC (FY25 14.6% per the proxy; 15.0% per ROIC.ai) and 17–19% in better volume years (FY21–FY23) — comfortably above its cost of capital, evidence the captivity converts to economic returns. Pass — though note ROIC compressed from ~19% (FY21) to ~15% (FY25) as volumes fell and the Plex/Clearpath acquisitions added capital, so the returns are cyclically sensitive and the acquired growth diluted them somewhat.

Direct competition.

• Siemens AG — the global #1 in industrial automation, several times larger globally, and dominant in Europe; Rockwell’s principal rival and the reason ROK’s moat is regional. Rockwell wins North American discrete; Siemens wins almost everywhere else.
• Schneider Electric / ABB / Mitsubishi Electric — broad, globally larger automation/drives/PLC competitors, far more international than Rockwell.
• Emerson / Honeywell (and ABB, Siemens, Yokogawa) — the process/DCS incumbents, where Rockwell is a challenger, not a leader. Rockwell’s PlantPAx exists but lacks the installed-base captivity that protects Logix in discrete; the impaired-and-dissolved Sensia oil-&-gas JV is a tacit acknowledgment of process/energy weakness.
• Dassault Systèmes / PTC / AVEVA — the software/digital-twin frontier, where the battle is shifting and where Rockwell is one competitor among several.
• Omron / Yaskawa / Fanuc — motion, servo, and robotics.

Where the moat is thin (skeptic’s view). (1) Internationally it largely evaporates — ~63% North America, EMEA/APAC declining and a fraction of Siemens; ROK’s own strategy lists “market access in Europe and Asia” as a top acquisition priority, an admission of channel weakness abroad. (2) Process automation is not its turf — a small challenger to Emerson/Honeywell/ABB/Siemens/Yokogawa. (3) Commoditized hardware at the edges — drives, sensors, and industrial components (the Intelligent Devices long tail, 18% margin) are more contestable, and ID volume fell in FY25; the captivity premium is concentrated in Logix, not the whole portfolio. (4) Recurring revenue is under-delivering the aspiration — ARR growing high-single-digit vs the “double-digit” goal.

Verdict: a durable but regionally bounded moat — genuine, captivity- and scale-driven, and wide/deep in North American discrete automation (Logix + channel + brand); thin internationally, in process, and in commodity hardware. The moat is real enough to defend ~30% software margins, ~15% ROIC, and 20%+ Logix growth through a down-cycle — but it is a North American discrete moat, not a global one, and that distinction matters for both the growth runway and the multiple the market is paying (see the Valuation section).

5. Growth History and Forward Opportunities

History is a cycle, not a secular ramp. Reported revenue: $6.33B (FY20) → $7.00B (FY21) → $7.76B (FY22) → $9.06B (FY23 peak) → $8.26B (FY24) → $8.34B (FY25). The 2021–2023 surge was partly real demand and partly supply-constrained over-ordering (customers double-ordered into a shortage, building Rockwell’s record backlog); the 2023–2025 decline was the destocking payback. Stripping the noise, the through-cycle organic algorithm is mid-single-digit, and FY25 organic growth was +1% — entirely price (+3 pts), with volume down 2 pts. This is not a fast grower; it is a cyclical share-and-margin compounder.

The current recovery is genuine and broadening, but early. Q2-FY26 (reported 5-May-2026) was strong: organic sales +9%, reported +12% (3 pts FX), with double-digit growth in orders, sales, and EPS and a book-to-bill slightly above the normal 0.95–1.1 corridor. By segment: Intelligent Devices +9% organic (20.9% margin, +320 bps y/y), Software & Control +17% organic (34.9% margin, +480 bps), Lifecycle Services −1% (14.6% margin). Logix grew 20%+. Management raised FY26 guidance twice, now to 5–9% organic growth (7% midpoint), ~21.5% enterprise operating margin, adjusted EPS ~$12.80 (from ~$11.80), and 100% FCF conversion.

The quality of the growth is improving but mixed. Where it’s working: data center (sales more than doubled y/y, driven by Cubic power distribution, Logix replacing commercial-grade controls in AI data centers, and drives into HVAC/chillers); e-commerce/warehouse automation (+30%, including OTTO AMRs displacing AGVs/forklifts); semiconductor (+high teens, AI/data-center-driven); energy (mid-single-digit, above plan, with the Sensia oil-&-gas business now back in-house). Where it’s still stalled: automotive and consumer packaged goods — Rockwell’s two largest verticals — where large CapEx remains delayed by tariff and USMCA uncertainty; growth there is coming from modernizations, mid-sized customers, and new offerings (AMRs), not greenfield capacity. Management is explicit that there is “no wholesale unlock” yet in auto/CPG.

Forward opportunities (real, but already in the price).

1. Reshoring — the single best structural tailwind, and Rockwell’s home-field advantage (~63% North America). New US capacity projects are building; an auto/CPG capex unlock (gated on tariff/USMCA clarity) is the key swing factor.
2. Data center / “industrial-grade” controls in AI infrastructure — highest-momentum new vertical, still small (low-single-digit % of sales), with a long runway if Logix continues to displace commercial controls.
3. Software-defined automation, AI, and robotics — OTTO AMRs, FactoryTalk + AI, cloud-native development enabling faster product cadence; the bet that Rockwell monetizes its installed base with higher-value software/recurring revenue.
4. Margin self-help — a productivity program lifting incrementals from the ~35% long-term algorithm to ~50% in FY26; management frames this as now “enshrined in the Rockwell operating model,” not one-off.
5. ARR / recurring revenue — the stated double-digit aspiration, currently running high-single-digit; the gap is the clearest place the narrative outruns the numbers.

Verdict: cyclical, mid-single-digit-organic growth with a genuine — but early and richly-extrapolated — reshoring/data-center/AI overlay. The recovery is real and broadening; the secular acceleration the multiple implies is a hypothesis, not yet a track record. High-quality growth in the right verticals (data center, e-comm, semi), gated by the two biggest verticals (auto, CPG) that have not yet unlocked.

6. Financial Quality

Rockwell’s economics are high-quality and improve with volume — the classic operating-leverage signature of a fixed-cost installed-base franchise — but the reported figures require normalization, and the returns are cyclically sensitive.

Margins and operating leverage. Gross margin is ~48% (FY25 48.1%, up from ~40–41% pre-2022 as price and productivity took hold; >50% in Q2-FY26). Operating margin is ~17% GAAP / ~20% segment (FY25), with incremental margins running ~50% in FY26 vs the ~35% long-term algorithm — i.e., as volume recovers, roughly half of each incremental sales dollar drops to operating profit. This is the core bull mechanic: a cyclical recovery should expand margins faster than revenue. The flip side is symmetric — the FY23→FY25 volume decline compressed operating margin from ~18.7% (FY23) to ~14.4% (FY24) before self-help rebuilt it.

Returns on capital. ROIC is ~15% (FY25 14.6–15.0%), down from ~18–19% in FY21–FY23 — compressed by both lower volumes and the capital added by the Plex (~$2.22B) and Clearpath/OTTO (~$609M) acquisitions. ROE is ~11% (FY25 11.3%), depressed by the low-volume year and by goodwill/intangibles on the balance sheet (tangible equity is negative — ~$4.7B disclosed intangibles + ~$3.8B goodwill against ~$3.7B total equity). The returns clear the cost of capital comfortably and would re-expand with volume, but they are cyclically sensitive and were diluted by acquisitions — a reason to be skeptical of paying a peak multiple.

Cash generation is the standout. Free cash flow was $1,358M in FY25 (~16% of sales; ~114% conversion of adjusted income) on light capex (~$186M, ~2.2% of sales); FY26 is guided to 100% FCF conversion of ~$1.44B adjusted net income. Over five years OCF has consistently met or exceeded net income (FY24’s $864M OCF was a working-capital trough, not a quality flag), and the business is genuinely capital-light. FCF/share has grown despite the revenue dip because of buybacks and margin self-help.

Quality-of-earnings flags (manage carefully).

• GAAP vs adjusted EPS gap is wide and widening. FY25 GAAP diluted EPS $7.67 vs adjusted $10.53 — a $2.86 (37%) gap, the widest in five years. The bridge is mostly legitimate non-cash/non-recurring items: the $224M Sensia JV impairment ($110M net of NCI/tax; ~$2.20/sh of the bridge), purchase-accounting D&A, and a $136M legacy-asbestos accounting-method change (~$0.91/sh). Adjusted $10.53 is a fairer run-rate than GAAP $7.67 — but the largest single adjustment (Sensia) is real value destruction being excluded from “adjusted,” so it should not be waved through uncritically. The forward FY26 adjusted EPS (~$12.80) is the right number to value, with eyes open to the ~$1B+ of cumulative impairments/charges that “adjusted” has absorbed over the period.
• FY23 GAAP was flattered by a one-time ~+$279M non-cash gain on the PTC investment (~$1.83/sh) — so the FY23 $11.95 GAAP EPS overstates the run-rate. Useful context for anyone anchoring on “ROK used to earn $12.”
• Sensia consolidation distorted the bottom line — GAAP net income ($749M) ≠ net income to Rockwell ($869M) in FY25 because $120M of the impairment loss was allocated to SLB’s minority interest; this reverses on the April-2026 dissolution.
• A 40M-share treasury retirement in FY25 is cosmetic (no change to shares outstanding) — don’t read it as a buyback.

Balance sheet. Conservative: ~$2.6B total debt, ~$468M cash, net debt ~$2.15–2.76B (~1x EBITDA), a laddered maturity profile, a renewed $1.5B five-year revolver (Nov-2025, undrawn), and a modest pension. Plenty of capacity for the dividend, buyback, and bolt-on M&A.

Verdict: yes, the economics improve with scale — high gross margin, ~50% incremental margins, ~15% ROIC re-expanding off a cyclical trough, capital-light ~16%-of-sales FCF, and a conservative balance sheet. The quality is real; the cautions are that returns are cyclically sensitive and were diluted by acquisitions, and that the headline EPS everyone quotes is adjusted, absorbing >$1B of impairments/charges across the period.

7. Capital Allocation

Management has allocated capital intelligently and conservatively, with one clear blemish and one governance gap.

Dividends — a ~16-year Aristocrat. Rockwell is an S&P 500 Dividend Aristocrat with ~16 consecutive years of increases. DPS: $4.08 (FY20) → $4.28 → $4.48 → $4.72 → $5.00 → $5.24 (FY25), a ~5% CAGR, paid out at a conservative ~44% of FCF (~$591M FY25). At $474 the yield is only ~1.1% — this is a grower, not an income stock.

Buybacks — steady, share-count-reducing. Repurchases ran ~$300M/yr (FY21–23), then $594M (FY24) and $419M (FY25), shrinking the share count ~1%/yr (115.2M → 112.4M over FY22–FY25) net of SBC. FY26 is guided to ~$850M of buybacks, and the board authorized a new $1.0B repurchase in June 2026 (on top of ~$927M remaining under the Sept-2024 authorization) — a meaningful step-up, though buying at an all-time-high multiple is value-questionable (the same critique applies to any buyer here, including the company).

M&A — disciplined bolt-ons, one blemish. The strategy is to add ~1%/yr of growth via recurring-revenue software/digital/robotics tuck-ins: Plex Systems (Aug-2021, ~$2.22B — the largest deal in company history, cloud MES), Fiix (CMMS), Kalypso (digital consulting), CUBIC/Knowledge Lens, Clearpath/OTTO Motors (Oct-2023, ~$609M, AMRs), and Verve Industrial (Nov-2023, ~$183M, OT-cybersecurity). Most are sensible adjacency buys, and FY25 was a deliberate M&A pause (zero deals) to focus on integration and the balance sheet. The blemish is Sensia — the oil-&-gas JV formed with Schlumberger/SLB in 2019, impaired twice ($157.5M FY23 + $224M FY25, ~$315M gross) and dissolved 1-Apr-2026, ROK reabsorbing the oil-&-gas automation business. Sensia is the clearest capital-misallocation episode of the period and a reminder that Rockwell’s edge is in discrete, not process/energy.

Capex and R&D. Capex is light (~2.2–3% of sales), supporting the high FCF conversion; ROK has flagged a “>$2B over five years” investment program in facilities/digital/talent (a watch item — management itself lists the risk that this capex “may not earn its return”). R&D is ~5.8% of sales (~8% with engineering), moderate for the “software” framing.

Incentive alignment — solid, with one real gap. CEO Blake Moret (Chairman & CEO) earned ~$15.4M in FY25 (~92% at-risk). The annual incentive is weighted Adjusted EPS 40% / Organic Sales Growth 25% / Free Cash Flow 20% / Organic ARR Growth 15%, with a hard EPS gate; the long-term plan is 40% performance shares (100% relative TSR vs. selected GICS groups) / 30% options / 30% RSUs. The plan demonstrably bites — the FY23–25 PSU paid only 70% of target (35th-percentile relative TSR) and FY25 ARR scored 0%. The gap: ROIC/return-on-capital is not a payout metric — it is disclosed (FY25 ROIC 14.6%) but management is paid on EPS/sales/FCF/ARR and relative TSR, none of which penalize value-destructive capital deployment as directly as a return-on-capital hurdle would. For a serial acquirer that just impaired Sensia, that is a genuine (if common) governance weakness. Insider signal is neutral: CEO sales are 10b5-1-planned (diversification, not conviction), there are no open-market purchases, and individual insiders own <1% (directors/officers as a group 0.76%); the largest holders are Vanguard (~12%) and BlackRock (~8%).

Verdict: intelligent, shareholder-friendly capital allocation — a ~16-year dividend Aristocrat at a conservative payout, steady share-count reduction, low leverage, and a mostly-disciplined bolt-on M&A program — with two caveats: the Sensia write-offs (process/energy is not ROK’s edge) and the absence of an explicit ROIC incentive metric. The June-2026 buyback step-up at an all-time-high multiple is shareholder-friendly in intent but price-questionable in execution.

8. Changes and Headwinds — Last Two Years

Cyclical reset and recovery. The dominant change is the 2023–2024 destocking down-cycle and the 2025–2026 recovery (see the Growth section). Revenue fell ~8% from the FY23 peak; FY24 guidance was cut repeatedly as orders softened; the stock gave back ~30% to the low-$240s; then FY25 troughed and FY26 has seen two guide raises. Anyone evaluating Rockwell today is looking at a business in the early innings of a cyclical upturn — which is exactly why the all-time-high multiple is the central tension.

Leadership. Christian Rothe became CFO in August 2024, succeeding Nicholas Gangestad. Blake Moret remains Chairman & CEO. The board expanded to 11 directors in April 2026. Change-of-control agreements were refreshed for senior officers in late 2025 (routine).

Portfolio: Sensia dissolution (April 2026). The oil-&-gas JV with SLB was dissolved on 1-Apr-2026, with Rockwell reabsorbing the oil-&-gas automation business into Lifecycle Services. The transaction is EPS-neutral, lowers reported revenue (~$100M in H2-FY26, ~$50M/quarter), and raises Lifecycle and total margin percentages. It follows two impairments — a clean-up of the period’s weakest capital allocation.

Tariffs and trade — the live headwind. Tariff/trade volatility is the most-cited near-term risk in management’s commentary. Rockwell targets tariff EPS-neutrality via pricing (FY26: ~250 bps total price, of which ~100 bps tariff-based), but tariffs are simultaneously delaying large capex in automotive and CPG — the two biggest verticals — and customers are watching USMCA renegotiation. The net effect is a drag on the pace of the recovery, even if margins are protected.

Input-cost inflation. Management flagged a double-digit-million-dollar memory-cost headwind in H2-FY26 (plus transportation and commodity inflation), partly offset by price and safety-stock building. A concrete, current margin pressure that informs the (prudent) guide for flat-to-down sequential margins in H2.

Geopolitics. The Middle East conflict paused some Lifecycle Services projects (limited impact to date; management does not expect a material full-year hit). China is a modest exposure and a swing factor (semiconductor strength in Taiwan offset broader caution).

Secular acceleration in data center / reshoring. On the positive side, the data-center vertical inflected sharply (sales more than doubled), e-commerce/warehouse stayed strong (+30%), and reshoring sentiment improved — the structural tailwinds that the bull case (and the multiple) rest on.

Verdict: the changes are net thesis-neutral-to-modestly-positive on fundamentals but thesis-negative on risk/reward. The recovery, the Sensia clean-up, and the data-center inflection strengthen the business; tariffs, input inflation, and the auto/CPG capex stall cap the pace; and the valuation has run far ahead of all of it. The fundamentals are improving; the margin of safety has evaporated.

9. Risk Analysis (Risk Matrix)

The catastrophic-loss risk is low — Rockwell is a profitable, cash-generative, low-leverage franchise with a durable installed base; a total or near-total permanent loss of capital is hard to construct. The dominant, high-probability risk is valuation/cyclical-timing (#1, #2): paying an all-time-high multiple at the start of a cyclical recovery, where a stall or de-rating produces a large drawdown even if the business is fine. That is a risk to the entry price and return, not to the survival of the business — which is precisely the distinction the Author’s Take draws.

10. Valuation Discussion (Embedded Expectations)

No price target and no recommendation here. This section frames what the current price embeds.

Where the multiple sits — all-time high on its own history. At $473.79, Rockwell trades at:

The single most striking datum is the P/S in the ~99.98th percentile of Rockwell’s own ~10-year range — the stock has never been more expensive on sales. That is the same “richest-ever versus itself” signature as the EMR cross-read, but more extreme on the sales line and stacked on recovering (not peak) earnings.

Embedded-expectations / reverse-DCF intuition. To justify ~37x forward adjusted EPS and ~6x sales, the market must be underwriting something close to: (a) sustained 7–9%+ revenue growth (well above the mid-single-digit through-cycle organic algorithm) as reshoring + data center + share gains compound; (b) continued margin expansion beyond the already-raised 21.5% enterprise margin (toward the mid-20s segment level), on ~50% incrementals holding through the cycle; and © a durable re-rating that treats Rockwell as a secular software/automation compounder rather than a cyclical, regionally-concentrated, mid-single-digit-organic hardware leader. A simple reverse-DCF at a ~9% discount rate and a ~2.5% terminal growth requires roughly low-double-digit FCF growth for a decade to support today’s EV — achievable in a strong recovery, but it leaves no margin for a cyclical stumble, a stalled auto/CPG unlock, or ARR continuing to undershoot its double-digit goal.

What the market is pricing correctly. The moat is real (30% software margins, 20%+ Logix growth through a down-cycle); the recovery is genuine and broadening; reshoring and data center are real tailwinds where Rockwell is exceptionally well-positioned; margin self-help is delivering; capital allocation is sound. None of that is in dispute.

What the market may be pricing incorrectly. It is (a) capitalizing the early innings of a cyclical recovery as a secular ramp; (b) paying a peak multiple on earnings that are recovering off a trough, not at a sustainable peak; © under-weighting that the moat is North American discrete, not global, and that international and process remain weak; (d) extrapolating data-center/ARR contributions that are still small relative to the $8.9B base; and (e) leaving essentially no margin of safety against the high-probability risks (valuation de-rating, cyclical stall, tariff-gated capex).

Scenario sketch (illustrative, not targets).

• Bear: the auto/CPG capex unlock keeps stalling on tariff/USMCA uncertainty, short-cycle recovery fades into a second air-pocket, FY27 organic growth reverts to low-single-digit, and the multiple de-rates toward its 5-year average (~4x sales / ~25x earnings). A large drawdown from $474 even with the business intact.
• Base: the recovery continues at a mid-single-digit organic pace, margins hold near the raised guide, adjusted EPS compounds low-double-digit, and the multiple slowly normalizes — total return roughly tracks earnings growth minus modest de-rating (low-single-digit to mid-single-digit annualized from here).
• Bull: reshoring and data center inflect into a multi-year capex super-cycle, auto/CPG unlock, ARR re-accelerates to double digits, incrementals stay ~50%, and the re-rating proves structural — EPS and the multiple both expand, justifying the price in hindsight.

Verdict: a great business priced for the bull scenario at an all-time-high multiple, with the risk/reward skewed unfavorably for new capital at $474. The embedded expectations require a secular acceleration that is plausible but unproven; the margin of safety is absent.

11. Variant Perception

Consensus. Rockwell is a high-quality reshoring/automation compounder in the early innings of a cyclical recovery turbocharged by data center, AI, and “physical AI”/robotics — worth a premium multiple because the secular tailwinds are durable and the margin self-help is structural. The sell-side is broadly constructive-but-full (DA Davidson initiated Neutral on 16-Jun-2026, a marker that even bulls see limited upside at the all-time high).

Strongest bull case. The moat is genuine and converting (30–35% software margins, 20%+ Logix growth through a down-cycle); Rockwell is the cleanest large-cap reshoring play (~63% North America); the data-center vertical (sales 2x+) and e-comm/warehouse (+30%) are inflecting; ~50% incrementals mean a cyclical recovery expands margins faster than revenue; capital allocation is sound (Aristocrat dividend, buybacks, low leverage); and if reshoring + AI infrastructure become a multi-year capex super-cycle, today’s multiple looks cheap in hindsight.

Strongest bear case. The stock has tripled to an all-time-high multiple (P/S ~99.98th pct, ~37x forward EPS) on earnings that are recovering off a trough, not at a peak; the through-cycle organic algorithm is only mid-single-digit (FY25 organic was +1%, all price); the moat is regional (thin internationally and in process — Siemens owns Europe, Emerson/Honeywell own process, Sensia was impaired and dissolved); the two biggest verticals (auto, CPG) remain capex-stalled on tariff/USMCA uncertainty; ARR is undershooting its double-digit goal; and the tape is a crowded, fully-believed momentum trade (beta 1.28, y1 +49%, strong momentum/dividend-yield loadings, no value loading) — i.e., consensus is long and offsides on price, not on quality.

The 3–5 assumptions that matter most.

1. Does the cyclical recovery sustain and broaden into auto/CPG, or stall in a second air-pocket? (The swing factor for FY27 organic growth.)
2. Is the reshoring/data-center/AI tailwind a multi-year super-cycle or a modest, slow overlay? (The justification for the secular re-rating.)
3. Do ~50% incrementals and the margin self-help hold through the cycle, or revert toward the 35% algorithm?
4. Does ARR re-accelerate to double digits, validating the software/recurring narrative the multiple pays for?
5. Does the multiple hold at an all-time high, or de-rate toward its own 5-year average? (The dominant driver of forward return from here.)

Falsification evidence. Bull falsified by: a second order/book-to-bill downturn, auto/CPG capex staying frozen through FY27, ARR stuck high-single-digit, and incrementals reverting to 35% — with the multiple de-rating. Bear falsified by: a sustained book-to-bill >1.1, an auto/CPG capex unlock, data-center scaling past mid-single-digit % of sales, ARR re-accelerating to double digits, and incrementals holding ~50% — validating the secular re-rate.

Factor-positioning read (the tape). FactorsToday shows Rockwell as a high-beta (~1.09–1.28), momentum-and-dividend-yield-loaded, market-sensitive name (Base-model R² ~0.44; market beta the dominant loading; a meaningful DividendYield loading ~0.6) with blistering recent risk-adjusted performance — y1 return ~+49% at a 1.63 Sharpe, m3 annualized ~+220% (i.e., a ~+33% quarter) at a 6+ Sharpe, m6 +50%. There is no value loading. The factor-similar peer set (DOV, PH, EMR, NDSN, MSM) is the multi-industrial complex. Interpretation: this is a crowded, fully-priced momentum winner that has already had its move — the empirical case for “great business, but the easy money is made and the risk/reward has inverted.” It is not a falling knife (no broken trend, alpha modestly negative but in a strong uptrend above all EMAs) and not a coiled-spring value name (no value loading, richest-ever multiple). The tape supports caution-not-shorting — consistent with the Author’s Take.

12. Fact vs. Interpretation Table

13. Open Questions

1. What is the absolute ARR dollar figure and the true software-vs-hardware revenue split? Rockwell does not disclose either in its filings — the recurring-revenue story is materially under-disclosed relative to peers.
2. How large can the data-center vertical realistically become? Management declined to size the TAM on the Q2-FY26 call beyond “low-single-digit % of sales, doubling”; the durability and ceiling are the key unknowns for the secular re-rate.
3. When does the automotive/CPG capex unlock? Gated on tariff/USMCA clarity; the single biggest swing factor for FY27 organic growth.
4. Will incrementals hold ~50% or revert to the 35% algorithm? Management signals “structural,” but the productivity program’s durability through a full cycle is unproven.
5. Will ARR re-accelerate to the double-digit goal, or stay high-single-digit? The clearest place the narrative outruns the numbers.
6. Does the “>$2B over five years” capex/investment program earn its return? Management itself flags the risk; no ROIC incentive to enforce discipline.
7. What is the normalized post-Sensia margin and revenue base? The dissolution is EPS-neutral but reshapes Lifecycle Services optics.

14. What Must Be True (Bull and Bear, with Falsification Tests)

For the BULL case to be right:

1. The cyclical recovery sustains and broadens — book-to-bill holds >1.0 (ideally >1.1), and auto/CPG capex unlocks in FY27.
2. Reshoring + data center + AI scale into a multi-year capex super-cycle, with data center growing past mid-single-digit % of sales and ARR re-accelerating to double digits.
3. Margin self-help proves structural — incrementals hold near 50%, enterprise margin pushes toward the mid-20s segment level.
4. The all-time-high multiple holds (or expands), treating Rockwell as a secular compounder.

Falsification test: two consecutive quarters of book-to-bill <0.95 with auto/CPG still frozen, ARR stuck high-single-digit, and incrementals reverting toward 35% — would falsify the bull and likely trigger a de-rating from the all-time high.

For the BEAR / valuation case to be right:

1. The recovery is early-cycle and the multiple is an all-time-high extrapolation of it.
2. Through-cycle organic growth stays mid-single-digit; the secular tailwinds are real but slow and small relative to the base.
3. The moat is regional (NA discrete) and does not justify a global-compounder multiple.
4. The multiple de-rates toward its own 5-year average, producing a large drawdown even with the business intact.

Falsification test: a sustained book-to-bill >1.1, an auto/CPG capex unlock, data center scaling past mid-single-digit % of sales, ARR re-accelerating to double digits, and incrementals holding ~50% — with the multiple holding — would falsify the bear and validate the secular re-rate.

The two falsification tests are near-mirror images, which is the honest statement of the situation: the business quality is not in dispute; the entire debate is whether an all-time-high multiple correctly prices an early-cycle recovery plus an unproven secular acceleration. Watch book-to-bill, the auto/CPG capex unlock, data-center scaling, ARR growth, and incrementals — these resolve the thesis.

15. Source Appendix

See Appendix B for the full source list. Primary sources: Rockwell Automation FY2021–FY2025 Forms 10-K (CIK 0001024478); FY2026 Q1–Q2 Forms 10-Q; FY2026 Q2 earnings call transcript (5-May-2026); DEF 14A 2026 (filed 22-Dec-2025); material 8-Ks (FY2024–FY2026). Quantitative cross-checks: ROIC.ai (statements, ratios, enterprise value, valuation multiples); AZI (price history, news, valuation-index own-history percentiles); FactorsToday (factor loadings, leaderboard, peers). Peer context: process-automation peer Emerson Electric (EMR).

APPENDIX A — Standard Diligence Questionnaire

Rockwell Automation, Inc. (NYSE: ROK) · Report date 2026-06-19 · Supplemental diligence questionnaire; Fact/Interpretation/Assumption labeled where it matters.

General

What thoughtful questions have other investors asked about this company? The recurring institutional debates: (1) Is the multiple defensible at an all-time high on earnings recovering off a destocking trough? (2) How big and durable is the data-center vertical (sales doubled but management won’t size the TAM)? (3) When does auto/CPG capex unlock — the two biggest verticals, frozen on tariff/USMCA uncertainty? (4) Are ~50% incrementals structural or cyclical? (5) Why is ARR undershooting the double-digit goal, and what is the absolute ARR figure (undisclosed)? (6) How much of the “software/Connected Enterprise” story is real vs a hardware company with a software wrapper? (7) Is Rockwell a secular reshoring compounder or a regionally-concentrated cyclical mispriced as the former?

Cyclicality & Earnings Nature

Are earnings at a cyclical high or low? Recovering off a low. FY25 revenue ($8.34B) is ~8% below the FY23 peak ($9.06B); FY25 organic growth was +1% (all price, volume −2%). FY26 is an early-cycle recovery (Q2-FY26 organic +9%, two guide raises). Earnings are neither at a sustainable peak nor at the trough — they are rebounding, which is the crux of the valuation tension. (Fact + Interpretation.)

Driven by external environment or internal actions? Both. Revenue swings are external (manufacturing capex cycle, PMI, destocking). The margin story is substantially internal — a productivity/self-help program lifting incrementals from ~35% to ~50%. (Fact.)

How stable are revenues? Cyclical, not stable — ~65% products/short-cycle, sensitive to PMI/IP; ~26% Lifecycle is longer-cycle/project-based; a recurring/ARR layer (high-single-digit growth) provides a stabilizing minority. (Fact/Interpretation.)

Outlook for products/services? FY26 guide: 5–9% organic growth (7% mid), ~21.5% enterprise operating margin, adjusted EPS ~$12.80, 100% FCF conversion. Discrete up low-double-digit; data center, e-comm/warehouse, semi strong; auto/CPG modernization-led, not greenfield. (Fact — management guidance, treat as hypothesis.)

How big is this market — growing, shrinking, domestic or international? Rockwell cites a ~$120B addressed market; secular tailwinds (reshoring, data center, AI/robotics, electrification) support low-to-mid-single-digit market growth through the cycle. Rockwell is ~63% North America — domestically concentrated, with international (EMEA/APAC/LatAm) currently a drag and a long-term opportunity/weakness. (Fact/Interpretation.)

Business Quality & Competitive Moat

Is the industry getting more or less competitive? Stable-to-intensifying — a high-barrier oligopoly in the core, but the battleground is shifting up into software/digital-twin (Dassault, Siemens Xcelerator, Schneider/AVEVA, PTC), where the contest is broader. (Interpretation.)

How profitable is the business (ROIC, ROE)? ROIC ~15% (FY25; ~18–19% in FY21–23), ROE ~11% (depressed by the low-volume year and intangible-heavy balance sheet). Gross margin ~48%; Software & Control segment ~30–35%. Above cost of capital, cyclically sensitive, diluted somewhat by acquisitions. (Fact.)

How profitable is the industry — competitors, barriers? Leaders earn high-teens-to-20s% margins on entrenched installed bases; barriers (switching costs, qualification, channel, brand) are high for incumbents — no successful new global discrete-automation entrant in decades. (Fact/Interpretation.)

Can the business be easily understood? Yes — sell controllers/drives/software/services that automate factories; razor-and-blade installed-base economics. (Fact.)

Can it be undermined by foreign low-cost labor? Not directly (it’s capital equipment/software, not labor-intensive output); indirectly, reshoring is a tailwind because labor cost/availability drives automation demand. (Interpretation.)

Do brands matter? Yes — Allen-Bradley is the de-facto North American factory-floor standard; brand + installed base + channel are the moat (patents explicitly are not). (Fact.)

Nature of competition? Siemens (global #1, dominant in Europe); Schneider/ABB/Mitsubishi (global); Emerson/Honeywell/ABB/Yokogawa (process, where ROK is a challenger); Dassault/PTC (software); Omron/Yaskawa/Fanuc (motion/robotics). Rockwell wins North American discrete; loses Europe and process. (Fact/Interpretation.)

Customers’ switching costs? Very high in the Logix/Integrated Architecture core — re-coding, re-validation (regulated industries), retraining, downtime risk on 15–30-year assets. Lower in commodity drives/sensors. (Interpretation, strongly supported.)

Financial Condition & Balance Sheet

Assets not fully recognized on the balance sheet? The installed base / switching-cost franchise and the Allen-Bradley brand are economically valuable but not capitalized; ~$4.7B disclosed intangibles + ~$3.8B goodwill are on the balance sheet (acquisition-driven). (Interpretation.)

Off-balance-sheet liabilities? Legacy asbestos and environmental liabilities (a $136M FY25 asbestos accounting-method change); pension (modest, ~$0.4B liability); operating commitments. No alarming off-balance-sheet exposure. (Fact.)

How conservative is the accounting? Mixed. Cash flow is high-quality (OCF ≥ net income, light capex, 100%+ FCF conversion). But the GAAP-to-adjusted EPS gap is wide (FY25 $7.67 vs $10.53, 37%), absorbing >$1B of impairments/charges over the period (Sensia, asbestos, restructuring) — mostly legitimate but worth scrutiny. (Fact/Interpretation.)

How CapEx-hungry? Light — capex ~2.2–3% of sales; a capital-light, FCF-rich model. A flagged “>$2B over five years” investment program is the watch item. (Fact.)

Capital Allocation & Management

How much FCF, and how is it used? ~$1.4–1.5B FCF (~16% of sales). Uses: dividends (~$591M, ~44% of FCF), buybacks (~$420–850M/yr), bolt-on M&A, modest debt management. Philosophy: dividend Aristocrat + steady buyback + ~1%/yr M&A. (Fact.)

Significant acquisitions recently? Plex (~$2.22B, 2021), Clearpath/OTTO (~$609M, 2023), Verve (~$183M, 2023), plus smaller digital/cyber tuck-ins; FY25 was a deliberate pause. The Sensia JV (with SLB) was impaired ~$315M and dissolved 1-Apr-2026 — the period’s clear blemish. (Fact.)

Buying back shares? Yes — ~1%/yr net reduction; FY26 guided ~$850M; new $1.0B authorization June 2026. Value-questionable at an all-time-high multiple. (Fact + Interpretation.)

Issuing large amounts of stock to insiders? No — SBC is modest (~$85–100M/yr), more than offset by buybacks. (Fact.)

Compensation policy of directors/management? CEO ~$15.4M (FY25, ~92% at-risk). Annual incentive: Adjusted EPS 40% / Organic Sales 25% / FCF 20% / ARR 15%. LTI: 40% relative-TSR PSUs / 30% options / 30% RSUs. The plan bites (FY23–25 PSU paid 70%). Gap: no ROIC payout metric for a serial acquirer. (Fact/Interpretation.)

Motivations of management? Equity-driven (low personal ownership, <1% individual / 0.76% group); CEO sales are 10b5-1-planned (neutral); no open-market insider buys. Alignment via comp, not large stakes. (Fact/Interpretation.)

Valuation & Market Data

ADR, MLP, or K-1 issuer? No — US C-corp, common stock, NYSE: ROK; standard 1099 dividends. (Fact.)

Dividend policy? ~16-year increase streak (Aristocrat), ~5% DPS CAGR, ~$5.24 FY25, ~44% FCF payout, ~1.1% yield at $474. A grower, not income. (Fact.)

How profitable? Highly — ~48% gross, ~17% GAAP / ~20% segment operating margin, ~15% ROIC, ~16%-of-sales FCF. (Fact.)

Net income diverging from cash from operations? No persistent divergence — OCF met/exceeded net income in FY25; FY24 OCF was a working-capital trough, not a quality flag. (Fact.)

Risks & Downside

What would cause the stock to decline? Multiple de-rating from the all-time high (the dominant risk); a cyclical stall / second order air-pocket; auto/CPG capex staying frozen; tariff/cost pressure outrunning price; reshoring/data-center under-delivery; ARR continuing to undershoot. (Interpretation.)

Risk of catastrophic loss? Low — profitable, cash-generative, low-leverage, durable installed base. The risk is to return from this entry price, not to survival. (Interpretation.)

Chance of a total loss? Very low — no plausible path to permanent total impairment of capital for a franchise of this quality and balance-sheet strength. (Interpretation.)

Recent News & Events

Has the business environment changed recently? Yes — a cyclical recovery is underway (Q2-FY26 organic +9%, two FY26 guide raises), with data center inflecting (sales 2x+) and reshoring sentiment improving; offset by tariff-gated auto/CPG capex and H2 input-cost (memory) inflation. (Fact.)

Significant acquisitions/divestitures? The Sensia JV dissolution (1-Apr-2026) — EPS-neutral, reabsorbs oil-&-gas automation, follows two impairments. No major M&A in FY25–FY26 (deliberate pause). (Fact.)

Change in accounting policies? FY25 legacy-asbestos accounting-method change ($136M); new “enterprise operating margin” presentation from Q2-FY26 (SEC non-GAAP requirement, presentation-only). (Fact.)

Recent changes — markets, facilities, management? New CFO Christian Rothe (Aug-2024); board expanded to 11 (Apr-2026); new $1.5B revolver (Nov-2025); new $1.0B buyback authorization (Jun-2026); data-center and e-comm/warehouse verticals scaling. (Fact.)

APPENDIX B — Source Appendix

Rockwell Automation, Inc. (NYSE: ROK) · Report date 2026-06-19 · CIK 0001024478

Sources are primary-first. Facts in the memo trace to these. Management commentary (transcript, guidance) is treated as a hypothesis validated against filings and external data. No price target or recommendation appears outside the Author’s Take.

Primary — SEC filings (EDGAR, CIK 0001024478; mirrored locally in output/ROK/sources/)

• Form 10-K, FY2025 (fiscal year ended 30-Sep-2025), filed 12-Nov-2025 — rok-20250930.htm. Segments, MD&A, risk factors (Item 1A), competition (Item 1), geographic/disaggregated revenue, capital structure, GAAP→adjusted reconciliation, Sensia impairment, asbestos accounting change.
• Forms 10-K, FY2021–FY2024 — rok-2021…/2022…/2023…/2024…0930.htm. Multi-year revenue, margin, ROIC, M&A (Plex, Clearpath/OTTO, Verve, CUBIC), dividend/buyback history.
• Forms 10-Q, FY2026 Q1–Q2 — quarterly financials through the period ended Mar-2026.
• DEF 14A 2026 proxy, filed 22-Dec-2025 — executive compensation (ICP metrics: Adj EPS 40% / Organic Sales 25% / FCF 20% / ARR 15%; LTI 40% relative-TSR PSU / 30% options / 30% RSU), beneficial ownership, ROIC disclosure (not a pay metric), say-on-pay.
• Material 8-Ks (FY2024–FY2026) — earnings prints; CFO transition (Rothe succeeds Gangestad, Aug-2024); 10b5-1 plan disclosures; $500M term loan (May-2025); $1.5B revolver (Nov-2025); board expansion to 11 (Apr-2026); buyback authorizations.
• Forms 3/4 — insider-ownership cadence (442 Form 4s over 5 yrs; routine grant/vest/10b5-1 sales; no open-market purchases).

Primary — earnings call transcript (via ROIC.ai)

• Q2 FY2026 earnings call, 5-May-2026 — Blake Moret (Chairman & CEO), Christian Rothe (CFO). Organic +9%, Logix +20%, data-center sales 2x+, e-comm/warehouse +30%, semi +high-teens; FY26 guide raised (5–9% organic, 21.5% enterprise op margin, adj. EPS ~$12.80, 100% FCF conversion); Sensia dissolution; tariff/pricing; H2 memory-cost inflation; ~$850M buyback; book-to-bill commentary.
• Prior calls enumerated (Q1 FY2026 through FY2024) via ROIC list_earnings_calls.

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

• ROIC.ai MCP — income statement, balance sheet, cash flow (FY2020–FY2025); profitability ratios (ROIC 15.0% FY25, ROE 11.3%, gross margin 48.1%); enterprise value (FY25 EV ~$42.5B at then-price; recomputed ~$56B at $473.79); valuation multiples (EV/sales, EV/EBITDA, P/E history); company profile.
• AZI — price history CSV (5-yr OHLCV, EMAs, beta 1.28; 5-yr low $178.60 on 17-Jun-2022; ATH $473.79 on 18-Jun-2026; 52-wk $317–$474); valuation_index own-history percentiles (P/S 99.98th, P/E 93.42nd, P/B 78.99th, composite 90.80th); news feed (DA Davidson Neutral initiation 16-Jun-2026; $1B buyback authorization + dividend 10-Jun-2026).
• FactorsToday — factor loadings (market beta ~1.09–1.28 dominant; DividendYield ~0.6; no value loading; Base-model R² ~0.44); leaderboard (y1 +49% at 1.63 Sharpe; m3 annualized +220%; lifetime max drawdown −75%); related stocks (DOV, PH, EMR, NDSN, MSM); stock-info (rs_12m +49.96).

Secondary / peer cross-read

• Process-automation peer Emerson Electric (EMR); used for industry framing, competitive cross-read (ROK = discrete, EMR = process), and valuation-percentile comparison.
• General financial media / sell-side initiations referenced via the AZI news feed (DA Davidson, 16-Jun-2026) — used only for consensus framing, not as fact.

Notes on reliability

• ROIC.ai/AZI/FactorsToday are third-party aggregated data, not primary; for US filers EDGAR and the 10-K/10-Q are authoritative and any material discrepancy resolves to the filing. Computed multiples at $473.79 use ~112.1M diluted shares and FY25/FY26 figures as labeled.
• Rockwell does not disclose an absolute ARR dollar figure, a software-vs-hardware revenue split, or Discrete/Hybrid/Process % splits in its filings — those are sourced from the earnings deck/call and flagged accordingly (an under-disclosure noted earlier).