← back to ranking

TT · Trane Technologies plc

Construction · mkt cap $101.4B · calls: Q1 FY2026 vs Q4 FY2025
38.0 conviction · conf-adj 38

conf 7/10 partial

enthusiasm:27.0 · trend:8 · quantifies:0 · impact:0 · under_radar:0 · credibility:0 · business_impact:4 · disruption:0 · commitment:-4 · confirmation:3

Enthusiasm latest 9 / prev 8 (rising)

Management’s AI thesis is that AI-driven data center growth expands Trane’s thermal management TAM, while smarter chillers/buildings and “agentic AI software tools” create a broader efficiency opportunity. Credibility improved in Q1 FY2026 because management tied the story to acquired assets, backlog, revenue targets, service training, and specific system-level control use cases. Quantification is strongest around data center/Stellar revenue and backlog, while direct AI software revenue remains unquantified.

GROUNDED NEXT-FY IMPACT vs CONSENSUS

Grounded on actual base — revenue $21.3B · net income $2.9B · net margin 13.7% · diluted EPS 12.98

These are next-fiscal-year annual uplift estimates, not next-quarter numbers.

Aggregate next-FY est. rev uplift: % · next-FY EPS uplift: % · vs analysts: inline · priced in: high · confidence: 7/10

ClaimFigureArithmeticNext-FY Rev %Next-FY EPS %
Agentic AI software tackles 30% of building energy waste
engagement · soft		30%30% is a TAM/efficiency-opportunity statement about wasted building energy, not Trane revenue. No Trane $ figure, take-rate, attach revenue, or software-line base disclosed anywhere in the inputs to convert it. Unanchored -> cannot size.
Stellar Energy 2026 expected revenue $500M
revenue		$500M (2026)$500M / $21,321.9M rev = 2.345% topline. At company net margin 13.69% -> $68.4M NI / $2,918.6M = 2.345% EPS. Data-center cooling sold INTO the AI buildout = supplier-side, excluded from adopter headline.2.3452.345
Stellar incremental over $350M acquired base ($150M)
revenue		$350M base -> $500M (2026)$500M - $350M acquired base = $150M true incremental in next FY. $150M / $21,321.9M = 0.704% rev; @13.69% net margin -> $20.5M / $2,918.6M = 0.704% EPS. Supplier-side.0.7040.704
Stellar incremental captured in April guide ($50M)
revenue		$50M (2026, new vs Jan guide)$50M new-news above the January guide ($350M x 1.25 = $437.5M). $50M / $21,321.9M = 0.234% rev; @13.69% net margin -> $6.8M / $2,918.6M = 0.234% EPS. Supplier-side; only genuinely-new number vs prior estimates.0.2340.234
Stellar backlog $1B, ~half ships 2026
revenue		$1B backlog; ~half in 2026Backlog $1B / $21,321.9M = 4.69% (stock, not P&L). ~half shipping 2026 reconciles to the ~$500M revenue row above, so not additive. Supplier-side; shown for context, not double-counted into aggregate.4.69
Stellar target $1B+ revenue in 2-3 years
revenue		$1B+ in 2-3 yrs$1B / $21,321.9M = 4.69% of revenue, but lands in 2-3 years, not next FY. Next-FY-attributable portion already captured in the $500M 2026 row; out-year ramp not assigned to FY2026. Supplier-side.
Stellar mid-teens+ EBITDA target
other		mid-teens+ EBITDAOn $500M 2026 rev, ~15% EBITDA = ~$75M EBITDA, BELOW Trane's ~18.6% operating margin, i.e. margin-dilutive HVAC hardware, not high-margin software. EBITDA->NI needs D&A/tax not disclosed; characterizes the same $500M row, not additive. Supplier-side.
Data center HVAC content 'about the same' / reference designs 12-24mo, 2-4yr out
other · soft		~same content; 12-24mo / 2-4yrQualitative content/timing guidance about future data-center designs. No incremental $ or volume base to convert to revenue. Supplier-side, unsizable.
Applied systems service tail 8-10
engagement · soft		8 to 10Service-tail multiple/years for applied thermal systems; no installed-base $ or attach-rate base disclosed to size the recurring service revenue. Supplier-side, unsizable.

Assumptions: Incremental net margin = company net margin 13.69% (Stellar's mid-teens EBITDA is hardware-grade and below corporate ~18.6% operating margin, so 13.69% is if anything generous; no higher software margin warranted since the lone software claim is unanchored). Tax rate 21% default where after-tax conversion needed. Percentages reported as percentage points (2.345 = 2.345%). Phasing: next FY = 2026; Stellar 2026 revenue $500M, of which $150M is incremental over the $350M acquired base and only ~$50M is new vs the January guide. $1B revenue/backlog ramp lands 2-3 yrs out, not assigned to FY2026 beyond the $500M. Segment mapping: Stellar = data-center power/cooling sold into the AI buildout (SUPPLIER); agentic AI building software = Trane improving its own products (ADOPTER).

Top line: Essentially ALL of Trane's quantified AI-linked revenue is supplier-side (Stellar Energy data-center cooling sold INTO the AI buildout): $500M expected in 2026 = 2.35% of $21.32B revenue, of which only ~$150M (0.70%) is incremental over the $350M acquired base and just ~$50M (0.23%) is genuinely new vs the January guide. $1B backlog (4.7% of revenue), ~half shipping in 2026, reconciles to that same $500M. The only ADOPTER-side claim — agentic AI software addressing the 30% of building energy that's wasted — carries no Trane revenue figure, attach rate, or software-line base, so the adopter topline uplift is null/unsizable. As an adopter story the headline revenue uplift is null; the real AI dollars are a supplier/data-center cooling story.

Bottom line: Stellar targets mid-teens+ EBITDA; on $500M that's ~$75M EBITDA, BELOW Trane's ~18.6% operating margin, so the buildout is margin-dilutive hardware, not high-margin software. Flowing the genuinely-new $50M at company net margin (13.69%) gives ~$6.8M = 0.23% of $2.92B net income; the full $150M incremental is ~0.70% of NI — and all of it is supplier-side. The adopter agentic-AI software has no quantified earnings impact, so adopter EPS uplift is null.

[sized AI revenue is supplier-side (selling into the buildout), not adopter — uplift n/m] All sizable AI dollars are supplier-side Stellar data-center cooling, with only ~$50M genuinely new vs the January guide; consensus already reflects the Stellar ramp, so the incremental surprise is small. The adopter (Trane's own products) AI story — agentic building-energy software — is unanchored and unsizable. Hence adopter uplift null, supplier ~2.35% topline, priced-in high, inline vs expectations.

MODEL CONSENSUS (impact)

partial

Both agree: only adopter claim is the unsizable 30% energy-waste TAM; all real AI revenue is supplier-side Stellar (~2.35%). Adopter uplift null.

Conflicts reconciled
FieldOpus 4.8GPT-5.5
Rev uplift %–0
EPS uplift %–0
Priced inhighhigh
vs analystsinlineinline
Confidence77
Top lineEssentially ALL of Trane's quantified AI-linked revenue is supplier-side (Stellar Energy data-center cooling, capacity sold INTO the AI buildout): $500M expected in 2026 = 2.35% of $21.32B revenue, of which only ~$150M (0.70%) is incremental over the $350M acquired base and just ~$50M (0.23%) is genuinely new vs the January guide. $1B backlog (4.7% of revenue), ~half shipping in 2026, reconciles to that same $500M. The only ADOPTER-side claim — agentic AI software addressing the 30% of building energy that's wasted — carries no Trane revenue figure, attach rate, or software-line base, so the adopter topline uplift is null/unsizable. Net: as an adopter story, the headline revenue uplift is null; the real AI dollars are a supplier/data-center cooling story.Adopter-side quantified impact is 0 because the only adopter claim, 30% building energy waste, lacks a disclosed Trane monetization base. Supplier-side Stellar 2026 revenue is $500M, equal to 2.3450% of current revenue.
Bottom lineStellar targets mid-teens+ EBITDA; on $500M that's ~$75M EBITDA, BELOW Trane's ~18.6% operating margin, so the buildout is margin-dilutive hardware, not high-margin software. Flowing the genuinely-new $50M at company net margin (13.69%) gives ~$6.8M = 0.23% of $2.92B net income; the full $150M incremental is ~0.70% of NI — and all of it is supplier-side. The adopter agentic-AI software has no quantified earnings impact, so adopter EPS uplift is null.Adopter-side EPS impact is 0 from hard figures. Supplier-side $500M Stellar revenue at the current 13.6883% net margin implies $68.441M incremental net income, or 2.3450% of current net income.
ReasoningConsensus already bakes +$2.055B 2026 revenue (+9.64%) and +14.9% EPS ($12.98 -> $14.91). Stellar's $500M (2.35% of rev) sits well inside that, the true $150M incremental is 0.70%, and the only new-vs-prior-guide dollar is ~$50M (0.23% rev / 0.23% EPS) — a rounding-scale addition to a +9.64% revenue line that's clearly already in the numbers. On the ADOPTER axis this product ranks, the aggregate is null: every quantified figure is supplier-side data-center cooling, and the genuine adopter lever (agentic AI building software, 30% energy-waste TAM) is long-term and unquantified, so it cannot move estimates yet. There is no adopter-side gap above consensus to exploit.Consensus 2026 revenue of $23.377B is $2.055B above the $21.322B current base, or +9.6407%; consensus net income of $3.305B is +$386.656M, or +13.2480%, and consensus EPS of $14.91408 is +14.9005%. The hard adopter-side quantified uplift is 0%, while the supplier-side Stellar 2026 revenue claim is only 2.3450% of revenue and appears embedded in the much larger consensus revenue growth trajectory.

Rows highlighted where the two models disagreed.

QUANTIFICATIONS
building energy waste addressable by agentic AI software tools: 30% (long term, both)
“And that's really going to be part of our future growth projections that we have going forward because we know that most buildings waste about 30% of the energy that they pay for.”
Stellar Energy backlog: $1 billion (Q1 FY2026, topline)
“If you look at it today, it's $1 billion in backlog.”
Stellar Energy shipment timing: about half (2026, topline)
“I think about half of that shipping here in 2026.”
Stellar Energy target revenue scale: $1 billion business (2 to 3 years, topline)
“Think of this as a business that in 2 to 3 years is a $1 billion business.”
Stellar Energy target EBITDA: mid-teens plus EBITDA (2 to 3 years, bottomline)
“Think of it with mid-teens plus EBITDA serving many verticals, not just data centers.”
Stellar Energy expected revenue: $500 million (2026, topline)
“Maybe just from a modeling perspective, as Dave said, we expect about $500 million of revenue this year from Stellar.”
Stellar Energy acquired base revenue: $350 million (base of acquired business, topline)
“We had about the base of that business that we acquired is around $350 million of revenue.”
Stellar Energy assumed growth: 25% (January FY2026 guide, topline)
“We also had anticipated about maybe 25% of growth off of that just based on where data center growth was going in our January guide.”
Stellar Energy incremental revenue captured: $50 million (April FY2026 guide update, topline)
“And think of that as probably around $50 million incremental revenue we got captured in April.”
Stellar Energy target revenue scale: $1 billion-plus revenue business (2 to 3 years, topline)
“And to Dave's point, we've got a lot of investments to make to take this from a $350 million business to a $1 billion-plus revenue business in 2 to 3 years.”
data center reference design horizon: 12 months or 24 months (future, topline)
“I think you got to take a reference design and think it's probably out there. We could argue whether it's 12 months or 24 months, but it's probably somewhere in between there.”
data center reference design horizon: two to four years out (future, topline)
“And think of these data centers as the data centers that will be built, you know, maybe two to four years out.”
data center HVAC content: probably about the same (two to four years out, topline)
“I'm gonna err on the side of saying it's probably about the same.”
applied systems service tail: eight to 10 (typical applied thermal management system, topline)
“And, you know, I think I could tell you right now, we think of it at the eight to 10.”
PAST (realized)
CURRENT (now)
FORWARD (guidance)
TRACK RECORD — PROMISE vs DELIVERY

—/100 (no quantified promises)   no-quantified-promises  6 calls reviewed

Across all six calls, Trane references AI only as an end-market tailwind (the 'AI revolution' driving data center HVAC demand) and as generic capability buzz — AI/digital building management, smart controls, connected services, and digital performance optimization — but never attaches a number plus a timeframe to any AI/ML/analytics capability. No quantified AI promise was ever made, so there is no judgeable AI promise-vs-delivery track record.

PRICED-IN (REFINED)
HIGH (already in)

Est. revisions rising  ·  Fwd P/E 41.2  ·  EV/Sales 4.9x

AI claim maps to Product, Service

Estimate revisions are rising: buy/strong-buy counts increased from December to May, holds declined, and price targets moved up from last-year to last-quarter to last-month averages. Consensus also embeds solid forward revenue and EPS growth through 2026. With a 41.2x forward P/E and about 4.9x EV/Sales, valuation is already rich for a mature construction/HVAC business, so AI-driven upside flowing through Product and Service appears largely priced in.
COVERAGE — ENTHUSIASM TRAJECTORY + CATALYSTS
2Q4 FY20242Q1 FY20253Q2 FY20257Q3 FY20256Q4 FY20258Q1 FY2026

AI enthusiasm across 6 calls — trend ↗ rising

AI/data-center demand moved from absent to a concrete applied HVAC growth driver with backlog, bookings, acquisitions, and customer agreements.

RECENT AI CATALYSTS & NEWS
BUSINESS IMPACT - QUALITATIVE MATERIALITY

5/10 qualitative impact   moderate  long-term · soft evidence

Where AI matters: smart-building controls and energy optimization software

Trane has a credible adopter use case in AI-enabled building controls, autonomous building operations, and service productivity, with management pointing to wasted building energy as the problem to solve. But the hard numbers are almost entirely supplier-side data-center cooling sold into AI buildout, while direct AI software revenue, attach rates, and margin impact are not sized.

Caveats: Most quantified AI upside is excluded supplier-side data-center demand, not adopter-side AI use; Agentic building software remains unquantified and could stay a feature rather than a major revenue line; Hyperscaler/data-center cooling enthusiasm may obscure margin-dilutive hardware mix; Building automation AI could face integration, cybersecurity, and customer adoption friction

AI DISRUPTION / CANNIBALIZATION RISK  tailwind · 1/10

AI does not commoditize Trane's core HVAC equipment, thermal management, installed base service, or building-controls franchise; physical assets, regulation, uptime requirements, and channel/service relationships remain durable. AI could shift value toward smarter controls, but that is more likely to enhance Trane's offering than deflate the model.

OPTIONS / MARKET STRUCTURE

option liquidity: good

ATM IV
—
TYPICAL BID-ASK
—
OPEN INTEREST
—

proxy inputs — dollar-ADV $615M · beta 1.257 · px $458.92

source: proxy (no options chain on FMP)
FMP /stable/ exposes no options-chain endpoint on this key, so ATM IV, bid-ask spread and open interest are unavailable. Liquidity below is a PROXY from dollar-ADV, beta and price level (a stand-in for option depth), not measured option-market data.

CONFIRMATION — INSIDERS · 13F · LANGUAGE
Mixed — insiders selling, institutions adding, management language 2/10 hedged.
INSIDERS selling 14 open-market sell(s) vs 0 buy(s) — net distribution
INSTITUTIONS (13F) adding as of 2026-03-31: 199 new / 157 closed positions; 784 increased / 646 reduced; institutional ownership -0.16pp; +45 net 13F holders
MGMT LANGUAGE 2/10 hedged AI is barely discussed; automation/digitalization references are broad, product-adjacent, and mostly positioning language without concrete AI commitments.
commit “Our high-efficiency systems and smart controls help customers save energy, lower operating cost and increased resiliency”
hedge “megatrends around sustainability, digitalization, and rising energy demand are intensifying the need for our systems and services”
hedge “we are well positioned to deliver differentiated shareholder value in 2026 and beyond”
VERBATIM AI QUOTES
“And long term, we believe that all buildings will be smarter. All buildings will be more resilient. And we believe that we're going to be part of the solution there with our agentic AI software tools to make buildings a lot smarter.”
— David Regnery, Q1 FY2026
“And that's really going to be part of our future growth projections that we have going forward because we know that most buildings waste about 30% of the energy that they pay for.”
— David Regnery, Q1 FY2026
“As far as our position in data centers, we like our position, right? We're thought of it as the thermal management experts, okay? We're working with hyperscalers. We're working with other influencers, chip manufacturers and designing what some refer to as reference designs, others refer to as data centers of the future.”
— David Regnery, Q1 FY2026
“But look, we get called on for a reason because of our expertise. And as far as the TAM goes, it keeps expanding, okay? And the data center vertical keeps moving with innovation, and we keep pushing some of that innovation and developing that innovation.”
— David Regnery, Q1 FY2026
“The other addition that we made was in LiquidStack, which really expanded our offering in CDUs. So another nice addition that's off to a great start, and we'll continue to leverage that.”
— David Regnery, Q1 FY2026
“I mean, again, with the end in mind, I mean, the service opportunity with the recent last few years' growth in data centers is still well in front of us.”
— Christopher Kuehn, Q1 FY2026
“I think you could think of buildings being smarter, I think you're going to see chillers being smarter. And we're doing a lot of work around that.”
— David Regnery, Q1 FY2026
“So having a chiller that knows when to run in a free cooling mode only, or having a chiller that knows when to run in a vapor compression cycle and for how long, understanding weather patterns and the impact that they have on these micro grids that are being created here with these chiller farms and knowing when to cycle which units, that's all part of the efficiencies.”
— David Regnery, Q1 FY2026
“We're working very closely with many influencers in the data center vertical. So think of hyperscalers, think of chip manufacturers, like Nvidia and others. We're helping them design data centers of the future. Or you may have heard them referred to as reference design data centers.”
— David S. Regnery, Q4 FY2025
“And I have not seen a reference design or data center of the future that does not include chillers, just to be very clear.”
— David S. Regnery, Q4 FY2025
“Now I think that when we talk about some of these future designs, you're gonna see a lot of innovation around the thermal management system, specifically around the chiller that is really exciting.”
— David S. Regnery, Q4 FY2025
“But I want everyone to realize that, you know, Trane Technologies is at the forefront of this innovation. And we're helping our customers think through what's possible.”
— David S. Regnery, Q4 FY2025
“I think the amount of power that is being consumed by the thermal management system may be less. Okay? So think of it as a trade-off between you will still need the thermal management system. How often it runs or the power that it consumes may be less.”
— David S. Regnery, Q4 FY2025
ANALYST QUESTIONS ON AI
Q (Q1 FY2026, Amit Mehrotra): Dave, I just would like to see if you can talk about what you think your TAM is within data centers and how Stellar may change that.
A: As far as our position in data centers, we like our position, right? We're thought of it as the thermal management experts, okay? We're working with hyperscalers. We're working with other influencers, chip manufacturers and designing what some refer to as reference designs, others refer to as data centers of the future. But look, we get called on for a reason because of our expertise. And as far as the TAM goes, it keeps expanding, okay?
Q (Q1 FY2026, Amit Mehrotra): I don't know if this question is for Chris, maybe if you can talk about data center service revenue and when you expect that to kick in?
A: I mean, again, with the end in mind, I mean, the service opportunity with the recent last few years' growth in data centers is still well in front of us.
Q (Q1 FY2026, Andrew Obin): Can you maybe talk about absorption chiller technology at Trane? What do you guys have? Do you need to add capacity? And does technology need to evolve to support behind-the-meter needs?
A: And long term, we believe that all buildings will be smarter. All buildings will be more resilient. And we believe that we're going to be part of the solution there with our agentic AI software tools to make buildings a lot smarter.
Q (Q1 FY2026, Noah Kaye): Maybe just want to ask about some of the improvements that the company made to the reference design for large-scale data center deployments.
A: So having a chiller that knows when to run in a free cooling mode only, or having a chiller that knows when to run in a vapor compression cycle and for how long, understanding weather patterns and the impact that they have on these micro grids that are being created here with these chiller farms and knowing when to cycle which units, that's all part of the efficiencies.
Q (Q1 FY2026, Nigel Coe): And then on the AI reference design, to what extent is that helping drive higher content for Trane.
A: As far as the reference design, look, every reference design I've seen as chillers in it. I would also tell you that in data centers, as in other verticals, we love to think of it at a system level, and we have the opportunity to think at a system level based on the breadth of our portfolio.
Q (Q4 FY2025, Andrew Alec Kaplowitz): But how is Trane adapting in thermal management as it adapts? To maybe a little more liquid cooling.
A: We're working very closely with many influencers in the data center vertical. So think of hyperscalers, think of chip manufacturers, like Nvidia and others. We're helping them design data centers of the future.
Q (Q4 FY2025, Amit Singh Mehrotra): I think the question really is about, you know, how much you need to run it and do runtimes get affected, and do you need as many of them?
A: The frequency at which the compressor side runs will vary, but which could impact the services side as to how often you service these pieces of equipment.
Q (Q4 FY2025, Thomas Moll): When we're two to four years out, do you think that fraction is higher, lower, about the same as today?
A: I'm gonna err on the side of saying it's probably about the same.