Johnson Controls

Phase A — Understand the business

Lens 1 · Company Overview

Johnson Controls designs, manufactures, installs, and services the systems that make commercial buildings run: HVAC equipment (the YORK chiller and applied-HVAC franchise), building controls / building-management software (OpenBlue), industrial refrigeration, fire detection & suppression, and electronic security — wrapped in a large direct-channel field-service business that produces recurring maintenance and retrofit revenue. The company describes itself, as of the latest 10-Q, as "a global leader in thermal management, mission-critical building systems, energy efficiency, and decarbonization... in rapidly expanding industries such as data centers, healthcare, pharmaceuticals, advanced manufacturing, and higher education". That phrasing is a deliberate repositioning — three years ago JCI marketed itself as a diversified industrial; today the masthead leads with "thermal management" and "data centers."

How it actually makes money: a mix of (1) products & systems — equipment sold and installed into new construction and retrofit projects, much of it recognized over time on a cost-to-cost (percentage-of-completion) basis on long-term contracts; and (2) services — maintenance, monitoring, repair/replacement, and long-term service agreements recognized over the contract term, the higher-margin, stickier half. Contract structure leans toward installed-base lock-in: equipment goes in, then a multi-year service annuity follows. Average service-contract life ~5 years for backlog purposes; some institutional building contracts run 25–35 years.

Customers: commercial, institutional, industrial, data-center, and governmental buyers — explicitly "data centers, hospitals, university campuses, advanced manufacturing, class A offices, and airports" as named high-growth verticals. No single-customer concentration is disclosed (customers.csv is empty; the 10-K notes governmental/institutional spending uncertainty but no >10% customer). Suppliers: steel, copper, aluminum, plus semiconductors, electronic components, and rare-earth-bearing parts — "some key parts may be available only from a single supplier or a limited group of suppliers". Competitors: Trane Technologies, Carrier, Daikin (applied HVAC); Vertiv and Schneider Electric (data-center thermal/power); Honeywell, Siemens (controls); plus regional players.

The defining corporate event: in July 2025 JCI closed the sale of its Residential & Light Commercial (R&LC) HVAC business to the Bosch Group (headline EV widely reported ~$8.1B for the broader JV; JCI received ~$5.6B after tax and expenses for its stake), reported as discontinued operations. With that, JCI became a pure-play commercial-buildings company and returned essentially all the proceeds to shareholders via a $5.0B accelerated share repurchase. Domicile: Ireland (Cork); operational HQ Milwaukee; ~23M sq ft across ~60 countries.

Lens 2 · Supply Chain

Upstream inputs → JCI → end customer, named at every stage:

• Raw materials: steel, copper, aluminum (commodity-priced; fixed-price backlog contracts create margin risk between order and conversion — see Lens 10). Hedged selectively via commodity contracts.
• Components: semiconductors, electronic components, rare-earth-mineral-bearing parts, refrigerants (notably R-454B, the low-GWP refrigerant in the new YORK YVAM chillers) — "not all of our business arrangements provide for guaranteed supply and some key parts may be available only from a single supplier". This is the genuine chokepoint: a single-source electronics or refrigerant constraint can throttle backlog conversion, and the 10-K explicitly says past disruptions "impact[ed] our ability to timely complete projects and convert our backlog."
• JCI manufacturing/assembly: ~6M sq ft owned + ~17M sq ft leased; applied-HVAC/chiller plants, fire/security manufacturing, monitoring centers.
• Recent vertical-integration move: acquired Alloy Enterprises (Q2 FY26) for "proprietary thermal management capabilities expected to enhance chillers, CDU [coolant distribution unit], and cold plate offerings". This is JCI buying its way up the liquid-cooling stack rather than only selling chillers at the facility level.
• Distribution & install: direct field channel (skilled-trade labor is itself a constraint — the 10-K flags skilled-labor shortages as a backlog-conversion risk).
• End customers / buyers: hyperscalers and colocation operators (data centers), hospitals, universities, governments, class-A office, advanced manufacturing.

Chokepoints: (1) single/limited-source semiconductors & refrigerants; (2) skilled field labor for installation; (3) commodity price moves on fixed-price backlog. None is a hard single point of failure, but all three gate the speed at which the record backlog turns into revenue.

Lens 3 · Competitive Advantages (Moats)

JCI's moat is not product superiority in any single category — it is the installed base + direct service channel + breadth, now sharpened by a credible new operating system.

1. Installed-base service annuity (the real moat). Decades of equipment installed in buildings worldwide generate a recurring stream of maintenance, monitoring, and retrofit revenue with high switching costs — once JCI controls the chiller plant and the building-management layer, the incumbent advantage on service renewals and upgrades is structural. Services grew across all three segments in FY25 and is the margin-mix driver.
2. Breadth / one-throat-to-choke. The ability to sell HVAC + controls + fire + security + digital as an integrated outcome (via OpenBlue) to a single building owner is a genuine differentiator for complex verticals (hospitals, data centers) where integration risk matters more than per-unit price.
3. Applied-HVAC engineering + the YORK franchise. In large applied chillers — the workhorse of data-center and mission-critical cooling — JCI has scale, an installed reference base, and the new low-GWP YORK YVAM centrifugal chiller (cited at ~40% energy reduction on R-454B).
4. Bargaining power: moderate. Over customers — strong on service renewals (lock-in), weaker on competitive new-build bids. Over suppliers — limited on single-source components (a real vulnerability), normal on commodities.

Durability verdict: the service-annuity moat is durable and underappreciated; the equipment moat is contestable and faces well-capitalized specialists (Vertiv, Schneider) moving fast in liquid cooling. The moat is "wide-ish and widening on services, narrow and contested on data-center hardware." Ground: positioning/bottlenecks commercial-layer files were missing for the datacenters topic, so Lens 3 is built from the 10-K business description + web competitive data, not a compiled positioning matrix.

Lens 4 · Segments

JCI resegmented on April 1, 2025 from four segments (Global Products + three Building Solutions geographies) into a pure three-geography structure: Americas / EMEA / APAC. All prior periods recast. There is no standalone "data center" or "products vs. services" reporting segment — the DC growth is embedded inside Americas (and the Applied-HVAC product line), which is the single most important disclosure-quality limitation in this name (see Lens 10/13).

FY2025 (year ended 2025-09-30) — continuing operations:

Consolidated FY25: gross profit $8,592M (GM 36.4%, up from 35.2%); SG&A $5,764M; continuing-ops net income attributable to JCI $1,721M ($2.63 diluted EPS); total net income $3,291M / $5.03 EPS including the $1,570M R&LC discontinued-ops gain. The Americas +1% reported masks +7% organic — the gap is divestiture noise (−$799M). EMEA margin expansion (+16% EBITA) is the cleanest operational story (productivity + services mix).

Q2 FY26 (quarter ended 2026-03-31) — the acceleration is visible by geography:

APAC sales rose 13% ex-FX, "led by over 20% growth in Applied HVAC primarily due to data center application growth" — the clearest segment-level fingerprint of the DC tailwind.

---

Phase B — Measure performance

Lens 5 · Earnings Result (latest print — Q2 FY26, reported 2026-05-06)

The beat-and-raise that re-rated the stock.

• Revenue $6,142M, +8% reported / +6% organic; modestly ahead of the ~$6.1B consensus.
• Adjusted EPS $1.19, +45% YoY, vs ~$1.11 consensus (≈7% beat); GAAP continuing-ops EPS $0.99–1.00.
• Gross margin 36.8% (gross profit $2,262M, +9%); Segment EBITA $1,114M (+18%); continuing-ops income before tax $738M (+47%) — operating leverage is real, helped by falling corporate expense ($152M vs $186M) and amortization roll-off.
• What drove it: large-project demand "including our solutions for large-scale data center projects," broad organic growth led by Americas, plus margin productivity.
• Guidance — RAISED. FY26 now: ~6% organic growth, ~50% operating leverage, adjusted EPS ~$4.85 (≈+30% YoY, +$0.30 vs the original guide), ~100% adjusted FCF conversion. Q3 guide ~$1.26 EPS / ~$6.43B revenue; Q4 ~$1.46. Tone shifted decisively more confident quarter-on-quarter.
• Balance-sheet flags (clean): net debt ~2.0x EBITDA, "within the long-term target range," ~$700M cash, strong liquidity. Receivables $6,269M and inventory $1,820M at FY25-end grew slower than sales — no working-capital red flag.
• Market reaction: stock subsequently made an all-time high of $148.21 on 2026-06-22, confirming the print + raise re-rated the name.
• Unusual vs. own history: orders +30% and backlog +26% are well outside JCI's historical low-single-digit organic cadence — this is a genuine demand-regime change, not noise. The one blemish: management flagged Middle East challenges upfront.

Lens 6 · Earnings Calls (sentiment trend)

No transcripts on the shelf (transcripts/ empty); this lens is ``, anchored on the Q2 FY26 call (2026-05-06).

What management is focused on (Q2 FY26):

• Data centers as the lead story: "Demand for our products, solutions and services remains strong, led by data centers where we're holding a leading position" (Weidemanis). CFO Vandiepenbeeck quantified the direct CDU contribution as modest-but-scaling — "CDU systems alone... expected to contribute around $100 million in revenue this year," with broader DC strength "across chillers, air handling units, and CDU systems." So the DC tailwind today is overwhelmingly chillers/AHU (large existing franchises) with CDU/cold-plate as the emerging optionality.
• The Weidemanis operating system (the Danaher import): framed as "Simplify [80/20], Accelerate [lean — weeks to days], Amplify [digital/AI — days to hours]." Execution metrics offered: ~1,400 colleagues engaged, ~1,000 leaders trained, >150 kaizens across ~20 priority areas. This is the DBS playbook being installed in plain sight.
• Guidance confidence: "raising our adjusted EPS guidance to approximately $4.85, representing roughly 30% growth and $0.30 higher than our original guide."

Tone shift over time: the trajectory from the FY25 10-K language (cautious on governmental/institutional spending uncertainty) to the Q2 FY26 call (momentum, differentiation, leading position) is a clear escalation in confidence as the order book inflected. Recurring new phrases: "momentum," "differentiation," "thermal management," "leading position in data centers." Things de-emphasized: the old diversified-industrial framing and the AFFF/litigation overhang (now largely settled). The single hedge they keep: Middle East softness and governmental-customer budget uncertainty.

Lens 7 · Comps

JCI vs. its building-tech and data-center-thermal peers. Multiples are ``, dated; where not sourced, marked n/a (no fabrication). Mixed source dates — treat as directional, not precise.

5-year average ROE column: n/a for the peer set (would be fabrication). JCI's own returns are inferred from continuing-ops net income ~$1.7B on ~$12.9B equity ≈ low-teens ROE, distorted post-buyback by the equity reduction.

Read: JCI at ~24.8x trailing / ~28.5x forward is the cheapest way to own large-cap AI-datacenter-cooling exposure — Vertiv trades at 2–3x JCI's multiple, Schneider at a premium. The market still prices JCI mostly as a buildings company; the DC mix is the embedded, under-priced call. That gap is the central thesis tension (Lens 12).

Lens 8 · Stock-Price Catalysts (>5% moves, last ~5 years)

Mostly ``; the pattern reveals what the market actually reacts to for JCI.

• Sep 2023 — ransomware attack (Dark Angels): ~27TB exfiltrated, ~$51M ransom demanded; JCI booked ~$27M response cost + ~$4M deferred revenue. A reputational/operational shock, modest financial dent.
• Apr 2024 — Tyco $750M AFFF/PFAS settlement with public water systems removed a major overhang.
• May 2024 — Elliott Management discloses >$1B stake (top-10 holder); Soroban also took a stake. This is the catalyst that started the value-creation clock.
• Jul 31, 2024 — CEO George Oliver announces retirement after "constructive" talks with Elliott; shares surged. The clearest "governance catalyst moves the stock" data point.
• Feb–Mar 2025 — Joakim Weidemanis (ex-Danaher) named/installed as CEO (effective Mar 12, 2025).
• Jul 2025 — Bosch R&LC HVAC sale closes (~$5.6B after-tax), funding the $5.0B ASR.
• May 2026 — Q2 FY26 beat-and-raise on record DC orders → re-rate to all-time high $148.21 (Jun 2026).

Pattern: JCI's stock reacts most to (a) governance/portfolio actions (Elliott, CEO change, the Bosch divestiture, buybacks) and (b) the order/backlog signal — far more than to any single quarter's revenue line. This is a self-help + secular-demand story, and the tape says the market rewards both. Macro (rates, construction cycle) is the background risk, not the primary driver right now.

---

Phase C — Judge people & books

Lens 9 · Management

This is an activist-driven management reset, and the operator quality is the whole bull case.

1. CEO Joakim Weidemanis (since Mar 2025). Ex-Danaher EVP (13 years; most recently Diagnostics & China), a genuine Danaher Business System practitioner. He was hired specifically to install lean/80-20 operating discipline at a company that Elliott judged structurally under-margined. The "Simplify / Accelerate / Amplify" framework and the kaizen cadence (>150 completed) are early but on-brand. Track record: Danaher's compounding machine is the gold standard for industrial operating systems; the open question is execution at JCI's scale and field-services complexity. Tenure/skin: ~15 months in seat — too early to grade, insider-ownership data not on shelf (insider-transactions.csv absent).
2. CFO Marc Vandiepenbeeck (since Jan 2024). Internal, long JCI finance background — continuity through the transition.
3. Predecessor George Oliver ran the multi-year transformation to a pure-play buildings company and stepped down under Elliott pressure (advisor through Dec 31, 2025). The board responded to the activist quickly and cleanly — a governance positive.
4. Capital-allocation history (recent — strongly shareholder-friendly): sold R&LC to Bosch for ~$5.6B after-tax and returned essentially all of it via a $5.0B ASR completed Q2 FY26 at an average $114.43/share (43.69M shares) — well below the current ~$138, so the buyback is already accretive. ~$4.5B authorization remains; dividend raised to $0.40/quarter ($1.51/share FY25). Bolt-on M&A (Alloy Enterprises) is targeted at the liquid-cooling stack, not empire-building.
5. Red flags (governance): the new exec bench is heavily imported from outside (Brandt ex-Otis, Schlitz ex-ITW, Scalia ex-Hershey, Weidemanis ex-Danaher) — a deliberate culture transplant that can either accelerate the turnaround or create integration friction. Comp/related-party detail not on the shelf (proxy not ingested). Archetype: professional-manager / operating-system installer, not founder — exactly the right profile for a mature industrial being optimized, and exactly the profile activists install.

Lens 10 · Forensic Red Flags

Forensic-analyst lens. Every figure labeled.

• Segment opacity (the #1 disclosure issue). The April 2025 resegmentation to pure geography means there is no "data center" revenue line and no products-vs-services split by segment in the financial statements — the DC growth that's driving the multiple is narrated (orders +40% Americas, APAC Applied HVAC +20%) but not broken out. Investors are trusting management's qualitative attribution. Not improper, but it limits independent verification and is a classic spot where a growth story can be over-claimed.
• Percentage-of-completion (cost-to-cost) revenue recognition. A large share of revenue is recognized over time on estimated total contract costs, "reviewed monthly," with cumulative catch-up adjustments. PoC accounting is inherently judgment-heavy — the lever for smoothing or pulling forward margin. The auditor (PwC) flagged this as the Critical Audit Matter. Watch for estimate revisions and unbilled-receivable growth.
• Non-GAAP framing. Headline EPS is adjusted ($1.19 vs GAAP $0.99–1.00 in Q2 FY26) — a ~19% gap. Segment EBITA excludes restructuring, impairments, AFFF, divestiture gains, pension MTM, and amortization of intangibles. The amortization add-back ($87M in Q2) is legitimate for a serial acquirer but inflates "EBITA" vs cash earnings — keep an eye on adj-to-GAAP drift as the bull narrative builds.
• Impairments (FY25 Q4): non-cash charges of $156M (Americas Retail asset group) and $184M (EMEA Subscriber business) — both non-core units being de-emphasized; cleanup, not deterioration of the core.
• Leverage stepped up by design: net debt $9,501M, net debt/net cap 42.4% at FY25-end (up from 35.6%) — but that's mechanical: the $5.0B buyback shrank equity. Management states net debt ~2.0x EBITDA, in range. Interest of $326M due next-12-months, $3.3B thereafter.
• Cash vs. earnings: FY25 continuing-ops operating cash flow $2,554M (up from $1,568M) tracks net income reasonably — no obvious divergence; capex is light (PP&E net just $2,193M) confirming the asset-light model and ~100% FCF-conversion guide.
• Tax: Irish domicile; effective rate 12% FY25 (vs 7% FY24, which had non-recurring benefits). OECD Pillar Two 15% global minimum tax applies from FY25 — a structural risk that JCI's effective rate drifts up toward 15%+, a quiet headwind to the EPS-growth story.

Regulatory findings (required sub-section).

• SEC Litigation Releases / AAERs: None. "No LR found" and "No AAER found" for Johnson Controls in the 2021-06-30 → 2026-06-30 window, verified via SEC EDGAR EFTS.
• Non-SEC enforcement / litigation (web): The material item is the AFFF/PFAS liability. Tyco Fire Products (a JCI subsidiary) agreed to a $750M settlement in April 2024 with a nationwide class of public water systems. Earlier: $17.5M to ~1,200 Wisconsin residents (2021). Tail risk remains: the settlement does not resolve opt-outs, post-May-2024 first-detections, or property-damage/stormwater claims — PFAS is a multi-year overhang, not fully closed.
• 10-K Item 3 (Legal Proceedings): Gumm v. Molinaroli — a putative class action tied to the 2016 JCI/Tyco merger; dismissed in its entirety March 28, 2025, plaintiffs appealing. Plus the standard environmental, asbestos, and self-insured matters in Note 20 (restricted cash is held against asbestos liabilities).
• Summary: No SEC accounting-enforcement history. The genuine legal exposure is PFAS/AFFF (largely settled at $750M but with a residual opt-out/future-claim tail) and routine asbestos/environmental. Verified via SEC EDGAR EFTS (LR, AAER), web search, and 10-K Item 3 as of 2026-06-30.

---

Phase D — Project & stress-test

Lens 11 · Forward Projection (FY2026 → FY2028)

Built bottom-up from FY25 actuals + the raised FY26 guide; output ``, inputs labeled. No forecast.ts create (watchlist loop — not logging a Brier forecast).

Anchor: FY25 continuing-ops adj EPS base ≈ $3.70–3.75. FY26 guide adjusted EPS ~$4.85 is management's, not mine.

Base-case logic: the backlog ($20.0B, +26%) and orders (+30%) give unusual revenue visibility for ~6% organic; the margin lever (DBS) is the swing factor and the part the market hasn't paid for. ~15–16% EPS CAGR FY25→FY28 in the base, driven roughly half by organic+margin and half by buyback. Brier-tracked base call (not logged this run): "JCI FY26 (ending 2026-09-30) non-GAAP EPS ≥ $4.85," p≈0.70 — management just guided to it with two quarters left and a record backlog; the risk is a Middle East / macro air-pocket, not demand.

Lens 12 · Bull vs Bear

Bull case. JCI is a re-rating-in-progress: an activist (Elliott) forced a portfolio cleanup (Bosch sale) and installed a Danaher operator (Weidemanis) into a mature, under-margined buildings business just as the AI build-out turned data-center cooling into the fastest-growing demand vector in the company's history. The proof is in the order book, not the slide deck — orders +30%, Americas orders +40%, backlog +26% to a record $20B, APAC Applied HVAC +20% on DC. On top of that demand sits a self-help margin story (lean/80-20) that is barely started (>150 kaizens, 1,400 of 100k employees engaged) and a shareholder-friendly capital engine ($4.5B buyback left after a $5.0B ASR at $114). The contrarian point: the market still prices JCI at ~25x as a sleepy buildings company while Vertiv trades at 2–3x that multiple for the same secular tailwind. If DBS delivers even half a Danaher-style margin trajectory, the earnings and the multiple both move.

Bear case (2–3 permanent-impairment risks).

1. The DC tailwind is real but small and contested where it's hottest. Direct liquid-cooling/CDU revenue is only ~$100M this year; the big DC numbers are chillers/AHU that JCI has always sold. The high-growth liquid-cooling/immersion frontier is being taken by Vertiv and Schneider (Motivair), who are better-positioned at the rack/row level. JCI risks being the "facility cooling" incumbent that participates in DC growth at a lower multiple than the rack specialists deserve — i.e., the re-rating may never fully arrive.
2. It's still a construction-cycle, fixed-price-contract industrial. A real-estate/non-residential-construction downturn, plus commodity (steel/copper/aluminum) inflation on fixed-price backlog, can compress margins and slow backlog conversion fast. Governmental/institutional budget uncertainty is already flagged.
3. Execution + expectations risk. The DBS turnaround is ~15 months old and largely unproven at JCI's field-services scale; the bench is freshly imported. With the stock at all-time highs on a raised guide, expectations are no longer cheap — a single soft quarter (Middle East, China Systems, a backlog-conversion miss) could de-rate it. Pre-mortem (18 months out, thesis broke): DC order growth normalizes off the 2026 spike, DBS margin gains prove slower/lumpier than the Danaher comp implied, Pillar Two lifts the tax rate, and the stock that re-rated to ~28x forward on momentum gives the multiple back to ~20x — EPS still grows but the stock is flat-to-down.

Are multiples too high? ~25x trailing / ~28.5x forward is above JCI's own historical range for a low-single-digit-organic industrial, but well below the data-center-thermal peer set. The multiple is defensible only if the DBS margin story and the DC mix shift are real. It is priced for execution, not for failure.

Contrarian view (what the market refuses to see): Bears anchor on "it's just an HVAC company at a 25x peak multiple." The under-appreciated truth is the margin story, not the revenue story — the market is paying for the data-center order book (visible, possibly peaking) while under-paying for the Danaher operating-system margin expansion (invisible, just starting). The asymmetry is in the latter.

Lens 13 · Devil's Advocate (short-seller)

Dismantling the bull case.

• The "data-center pure-play" framing is a marketing overlay on a slow industrial. ~$24B of revenue, and the genuinely new DC liquid-cooling piece is ~$100M. Strip the narrative and you have a ~6%-organic buildings company at a near-record multiple. The orders +30% is real but lapping easy comps and front-loaded by a one-time AI capex surge — what's the 2027 order growth when hyperscaler capex digests?
• Revenue concentration risk is hidden by the resegmentation. Because there's no DC segment, you cannot see customer concentration in the DC business. If a handful of hyperscalers drive the order spike, a single capex pause (or in-housing of cooling, as some hyperscalers design their own) hits hard and invisibly.
• The moat is weaker than bulls think at the high-growth edge. JCI's durable moat is the service annuity on the installed base — a low-growth, defensive asset. At the frontier (liquid cooling, CDUs, cold plate), it's a follower buying capability (Alloy) against entrenched specialists (Vertiv, Schneider). Bulls conflate the defensive moat with frontier leadership.
• Most dangerous competitor bulls underestimate: Vertiv. Purpose-built for data-center thermal/power, faster, and already the category reference — it can out-innovate JCI at the rack level even if JCI owns the chiller plant.
• Capital-allocation skeptic note: the $5.0B ASR returned the Bosch proceeds to shareholders — shareholder-friendly, but it also levered the balance sheet (net cap 35.6%→42.4%) and arguably reflects a lack of higher-return internal reinvestment options. Buying back stock is what you do when you can't grow the core organically fast enough.
• Assumptions that must hold for today's price: (1) DC/AI building-cooling demand stays elevated through FY27; (2) DBS delivers Danaher-like margin gains at field-services scale; (3) no construction-cycle downturn; (4) tax stays near 12% despite Pillar Two. If growth disappoints 20–30%: organic falls to ~3–4%, FY27 EPS lands closer to the bear ~$4.85, and a momentum name at ~28x forward compresses toward ~20x — meaningful downside from $138.
• The one scenario that permanently impairs: hyperscalers structurally in-house data-center cooling design (as they've done with servers/networking), turning JCI's DC opportunity into commoditized facility equipment — plausible over a 3–5 year horizon and the real bear tail.

Lens 14 · Management Questions (ordered by information value)

1. Quantify the data-center business — what was DC-related revenue and orders in FY25 and the latest quarter, and what % of the $20B backlog is data-center? (The single biggest disclosure gap.)
2. Of the DC opportunity, how much is large applied chillers/AHU (your incumbent franchise) vs. liquid-cooling/CDU/cold-plate (the contested frontier), and where do you actually win head-to-head with Vertiv and Schneider?
3. What is the realistic margin ceiling the operating system ("Simplify/Accelerate/Amplify") can drive over 3 years, and what's the cleanest metric for us to track it (incremental margins? services mix?)?
4. How concentrated is the DC order book by customer, and what is your exposure if one or two hyperscalers pause capex or in-source cooling?
5. With ~$4.5B buyback authorization left and net debt ~2x, what's the priority between buybacks, bolt-on M&A up the cooling stack, and deleveraging through FY27?
6. What does the Alloy Enterprises acquisition give you that you couldn't build, and should we expect more frontier-cooling M&A?
7. How exposed is FY27 EPS to OECD Pillar Two — where does the effective tax rate settle?
8. What is the commodity/fixed-price-contract margin risk in the current backlog if steel/copper inflate, and how much is hedged or repriceable?
9. What's the single-source-component (semiconductor/refrigerant) risk to converting the record backlog on schedule?
10. The Middle East was flagged as a challenge — what's the scope, and is China Systems still a drag?
11. What is the residual PFAS/AFFF tail exposure beyond the $750M settlement (opt-outs, post-May-2024 first-detections, property/stormwater)?
12. Services vs. products & systems — what's the mix today and the target, given services is the higher-margin annuity?
13. How much of the order surge is genuine new demand vs. customers ordering early to secure capacity/lead times?
14. What's the OpenBlue (digital/software) attach and monetization trajectory, and is it a margin driver or a cost center?
15. With a bench heavily recruited from outside (Danaher, Otis, ITW, Hershey), how are you managing culture/integration risk during the transformation?

---