SYNTHOS RESEARCH

Ecolab ECL

Basic Materials · Chemicals - Specialty · Synthos Deep Dive · 2026-08-04

$286.75
Watch

The Overview

Ecolab is the company that supplies the soap, the dishwasher chemicals, the water treatment and the pest control to a large share of the world's restaurants, hotels, hospitals, food plants and factories. It installs the dispensing equipment, sells the chemicals that go in it, and sends 25,000 people out to service the accounts. Customers rarely switch, because the cost is small and the consequence of getting it wrong — a food-safety incident, a shut-down boiler — is large. It has raised its dividend every year for 33 years.

In the last eight months it has done something quite different. It has spent $6.35 billion buying two companies that sell into artificial-intelligence data centres. The first, Ovivo Electronics, makes the ultra-pure water that semiconductor factories need. The second, CoolIT Systems, bought for $4.75 billion from the private-equity firm KKR, makes liquid cooling systems for the racks of computers that run AI. Ecolab borrowed $5 billion to pay for it.

That is a real change of character, and it may well be a good one. Data centres consume enormous quantities of water and power, and the cooling problem is genuinely the constraint on how much computing can be built. Management says this new division is already worth about $1.5 billion in annual sales and could be $4 billion by 2030.

Two things make us wait rather than buy.

The shares cost about 35 times what analysts expect the company to earn this year. That is a price for a company that grows well and reliably — and Ecolab's underlying sales grew about 2% a year in 2024 and 2025 before these purchases.

And the company has not told anyone what CoolIT actually earns. We know what Ecolab paid. We do not know the revenue, the profit or the margin. The first time we will see it is on 27 October.

Our estimate of fair value is $301, about 6% above the price. Analysts average $327.33.

One thing worth noting in Ecolab's favour: our expert knowledge base contains a claim recorded in April 2025 — fifteen months before any of this — arguing that data-centre cooling and semiconductor fabs would drive demand for exactly Ecolab's kind of water business. That was right. But all eight of the claims we hold on this company come from a single recording, so it is one voice being right, not eight.


Putting a number on it: our fair-value estimate is $301 against a current price of $286.75 — real upside if our numbers are right.

Our summary metrics

Downside Risk (lower = safer)6/10High

"Rated 6 — an unusually stable services business taking on unusually large balance-sheet and integration risk at exactly the moment its multiple is high. The supports are structural and real: gross margin of 44.1% in the June quarter; approximately 44% of revenue from a razor-and-blade model with installed equipment and multi-year contracts; a 33-year dividend-increase record; return on equity of 21.4%; four reportable segments all growing, with the smallest and fastest (Global Life Sciences) up 18% and operating income up 46%; and revenue diversified across seven geographies with no single non-US region above 20%. Against that: net debt was $7,590 million at 2025-12-31 on the FILED basis and total debt at 2026-06-30 is $13,176.2 million against $5,135.3 million of cash — with the cash earmarked for a $4.75 billion payment that has now been made. Pro-forma net debt is roughly $12.8 billion against a rebuilt fiscal-2025 EBITDA of $3,889 million, i.e. leverage moving from roughly 2.0x to roughly 3.0-3.3x depending on a CoolIT earnings contribution that has never been disclosed. Goodwill of $9,419.3 million and other intangibles of $3,449.1 million exceed total equity of $10,084.4 million, so TANGIBLE book value is NEGATIVE at approximately minus $2,813 million; the company's own 10-K names impairment of Nalco, Purolite and Ovivo Electronics goodwill as a specific risk. Free cash flow was $1,904.3 million in fiscal 2025 on filed figures — a 2.39% yield against the current market value — and capital expenditure rose 5.4% to $1,048.3 million with the first quarter of 2026 running 46% above the prior year. The valuation carries no cushion: 34.7x the 2026 consensus."

Growth Quality7/10High

"Rated 7 — genuine, accelerating, and now partly bought. Reported sales: $15,320.2M (2023), $15,741.4M (2024, +2.7%), $16,081.2M (2025, +2.2%), and the first half of 2026 at $8,481.5M against $7,720.2M, plus 9.9%. The June quarter alone was $4,415.4M, plus 10%, of which organic growth was plus 5% and acquisitions contributed $122.3M. Management guides second-half reported sales growth of 12-14% and ORGANIC growth accelerating to 6-7%, with adjusted operating margin near 19%. Every segment grew in the half at fixed currency: Global Water plus 9%, Global Institutional and Specialty plus 4%, Global Pest Elimination plus 9%, Global Life Sciences plus 13% with operating income plus 35%. Consensus has adjusted EPS at $8.159 (2026), $9.398 (2027) and $10.774 (2028) — growth of 8.4%, 15.2% and 14.6% on a fiscal-2025 actual of $7.53 — and management's own raised 2026 guidance of $8.05-$8.25 brackets the consensus almost exactly. The acceleration in 2027-2028 is the acquisition arriving: management states the Global High-Tech platform is approaching $1.5 billion of annualised sales, targets $4 billion by 2030 at 25% operating margins, and expects it to contribute more than two percentage points to annual sales growth one year after acquisition. That is company guidance and half-weighted. What holds this at 7 rather than 8: the organic base compounded at 2.2-2.7% for two years before this, the step-change is purchased with $6.35 billion of cash and $5.0 billion of new debt, and CoolIT's own revenue and margin have never been separately disclosed."

Exponential Potential4/10Moderate

"Rated 4 — the highest exponential score we can honestly give a specialty-chemicals company, and it rests on a business the company has owned for weeks. CoolIT Systems designs and manufactures coolant distribution units, cold plates and direct-to-chip cooling for data centres. Management states its pre-acquisition sales were up MORE THAN 100% in the first half of 2026, that the combined Global High-Tech platform approaches $1.5 billion annualised, and that it targets $4 billion by 2030 at 25% operating margins — which would be roughly a 28% compound rate and a margin well above Ecolab's 18.3% company average. If that happens it is a genuine step-function, and the knowledge base's own liquid-cooling lane (jordi_visser at skill 2.0, three separate claims, and jensen_huang's observation that 1.95 of every 2 tons of data-centre rack mass is cooling) says the end-market is real. Three things hold the score at 4 rather than higher. The $4 billion target is management's, dated 2026-07-28, and is half-weighted. High-Tech at $1.5 billion is 8.5% of consolidated revenue, so even a quadrupling leaves the company predominantly a dispenser-and-chemistry business selling to restaurants, hotels, hospitals and paper mills. And the remaining 91.5% is a superb but linear compounder whose growth function is price of 2-3% a year plus volume — the knowledge base's own characterisation, at conviction 65, is 'value-based pricing lets Ecolab take 2-3% price annually'. A 4: one real exponential option bolted to an excellent linear machine, bought rather than built, at a price of $4.75 billion."

Fair value$301 $215–$373
What does “fair value” mean?

Fair value is Synthos’s estimate of what one share is worth today, based on our model of the company’s future cash generation and the risks to it. It is not a price target or a prediction of where the stock trades next quarter — it’s the price where we think risk and reward are balanced. Above it, you’re paying for outcomes better than our base case; below it, the market is offering a margin of safety.

The model’s inputs and weightings are proprietary — they’re the product. What we publish is the output, We don’t publish a reverse-DCF cross-check for pre-profit companies — negative or missing earnings break that math — so take this number on our modeling alone.

What do the 5 tiers mean? (Core · Tactical · Watch · Hold · Avoid)
Buy — CoreOwn it as a foundation — start or add now, size it for years, let dips be gifts.
Buy — TacticalGood price + confirmed trend + a defined exit — buy the setup, not a marriage.
WatchWe want the business, just not at this price/setup — act only when the listed trigger hits.
HoldFine to keep if you own it — no reason to buy more; new money does better elsewhere.
AvoidDon't own it — the problem is the business or the expectations, so a cheaper price won't fix it.

The Road Ahead

What we expect to matter in each window, and the evidence that would prove us wrong.

Short term 0-6 months

Neutral
Driver
"A high-quality business that has gone nowhere for a year, at a full multiple, with a very large acquisition landing in the next print. Ecolab closed 2026-08-04 at $283.14, up 1.33% from $279.43, having traded $276.345 to $283.73 on 843 thousand shares — the thinnest turnover in this batch at roughly $239 million. That is 8.2% below a 52-week high of $308.35 and 15.2% above a low of $245.73, 5.0% above a 50-day moving average of $269.71 and 4.5% above a 200-day average of $270.84. RSI is 61.6 and MACD +2.59 — the least stretched configuration among the names in this batch that have risen. Relative performance is the important part: 3-month plus 10.8% against SPY plus 7.6%, but 6-month MINUS 0.2% against plus 11.1%, and 12-month plus 8.5% against SPY plus 24.3% and QQQ plus 30.8%. Ecolab has underperformed the index by roughly 16 points over a year while executing well and raising guidance twice. That is the setup a patient buyer wants and it is not yet cheap: 34.7x the 2026 consensus. The near-term swing factor is entirely the 2026-10-27 print, which carries the first CoolIT contribution, the first visible amortisation and financing drag, and management's own warning that both will weigh on reported earnings."
What we’re watching
"The 2026-10-27 print against consensus EPS of $2.18 and management's guidance of $2.13-$2.23. Four things inside it. First, any separate disclosure of CoolIT revenue, growth or margin — none exists today, and $4.75 billion has been paid. Second, the size of the acquisition amortisation and financing drag, which management has twice named as a headwind to adjusted earnings without quantifying. Third, whether organic sales growth reaches the guided 6-7% from the June quarter's 5%. Fourth, the adjusted operating margin against the guided approximately 19% for the half and the roughly 20% organic figure — which would validate a knowledge-base claim from April 2025 predicting exactly 20% by 2027. Also watch the pace of repurchase: treasury stock rose $669.3 million in the first half, ahead of the $783.8 million spent in all of 2025, and $4.75 billion has just left for CoolIT."
Confidence
Medium

Medium term 6-24 months

Tailwind
Driver
"The medium-term case is the clearest in this batch and the price already reflects a good deal of it. Consensus has adjusted EPS at $8.159 (2026), $9.398 (2027) and $10.774 (2028) — 8.4%, 15.2% then 14.6% growth — and the acceleration is legible rather than assumed. Its components: organic sales growth guided to 6-7% in the second half against 5% in the June quarter, driven by strengthening pricing; adjusted operating margin moving to approximately 19% in the second half from 18.3%, with an organic figure near 20%; and the Global High-Tech platform, approaching $1.5 billion annualised and targeted at $4 billion by 2030 at 25% operating margins, contributing more than two percentage points to annual sales growth from roughly mid-2027. Every segment is growing and the fastest-growing is the highest-margin. Against that stands the balance sheet: $5.0 billion of new notes at 4.600% to 5.350% adds roughly $250 million of annual interest expense against a fiscal-2025 base of $241.1 million — an effective doubling — and management has named the non-cash amortisation and financing costs of CoolIT as a drag on adjusted earnings in both the third-quarter and full-year outlooks. Total shareholder yield is modest at roughly 2.7%: a $2.84 trailing dividend, 1.00%, with a 33-year increase record, plus repurchases running near $1.3 billion annualised on the first-half pace."
What we’re watching
"Whether CoolIT's growth persists after the acquisition — management cites pre-acquisition sales up more than 100% in the first half, which is the kind of figure that decelerates. Whether the $4 billion-by-2030 High-Tech target survives its first two years of reporting. Whether leverage comes back down — pro-forma net debt of roughly $12.8 billion against a rebuilt fiscal-2025 EBITDA of $3,889 million is roughly 3.0-3.3x before any CoolIT contribution, against a company that has historically run near 2x. Whether the organic margin reaches the 20% that both management and the knowledge base's April-2025 claim point to. Whether Global Institutional and Specialty — 37% of sales, growing only 4% at fixed currency — reaccelerates. And whether the segment reporting is stable: the reportable segments were reorganised effective 2026-01-01 into Global Water, Global Institutional and Specialty, Global Pest Elimination and Global Life Sciences, which means the vendor's own segment history is discontinuous at that date."
Confidence
Medium

Long term 2+ years

Tailwind
Driver
"The long-run case is the one the knowledge base makes and it is a good one, provided the entry price is right. Ecolab sells water treatment, cleaning chemistry, pest elimination and infection prevention through dispensing equipment it installs and services, under multi-year contracts, to customers for whom the product is a small fraction of cost and a large fraction of risk. The knowledge-base characterisation — 'installed hardware, embedded chemistry, deep field-sales relationships and 3-5yr contracts create very high switching costs' at conviction 80 — matches what the filings show: 25,000 sales and service employees, 3,000 research and engineering staff, more than 170 countries, 40 industries, and a raw-material base of 10,000 fragmented inputs of which the largest is 4%. It has raised its dividend for 33 consecutive years. The underlying secular driver is water scarcity, which does not reverse. Layered on that is a new and much faster end-market: advanced computing needs ultrapure water for fabs and liquid cooling for racks, and Ecolab has now bought positions in both. The long-run risk is not competitive but financial and behavioural — this is a company that has just paid $4.75 billion for a business in the most crowded thematic market in the world, on top of $1.6 billion two quarters earlier, funded with debt, at a moment when its own equity trades above 30x forward earnings. Acquisitions made into a boom are the ones that get impaired, and Ecolab's own 10-K risk factors say so."
What we’re watching
"Whether CoolIT is still growing in 2028 and 2029, or whether $4.75 billion bought the top of a capital-expenditure cycle. Whether goodwill — $9,419.3 million against $10,084.4 million of total equity — is ever impaired; the 10-K names Nalco, Purolite and Ovivo Electronics specifically. Whether pricing power of 2-3% a year holds through a deflationary period, given the disclosed contract-renewal lag on passing through input costs. Whether the 33-year dividend record survives a leverage cycle. Whether Greater China — $722.4 million, 4.5% of revenue, and rendered in the payload under a corrupted label — becomes a material exposure or a material problem. And whether Christophe Beck's team continues to divest as readily as it acquires; the knowledge base credits 'willingness to divest underperformers' at conviction 68, and the ChampionX separation and the 2024 sale of the global surgical solutions business support it."
Confidence
Medium

Exponential Potential

Exponential Potential4/10Moderate

"Rated 4 — the highest exponential score we can honestly give a specialty-chemicals company, and it rests on a business the company has owned for weeks. CoolIT Systems designs and manufactures coolant distribution units, cold plates and direct-to-chip cooling for data centres. Management states its pre-acquisition sales were up MORE THAN 100% in the first half of 2026, that the combined Global High-Tech platform approaches $1.5 billion annualised, and that it targets $4 billion by 2030 at 25% operating margins — which would be roughly a 28% compound rate and a margin well above Ecolab's 18.3% company average. If that happens it is a genuine step-function, and the knowledge base's own liquid-cooling lane (jordi_visser at skill 2.0, three separate claims, and jensen_huang's observation that 1.95 of every 2 tons of data-centre rack mass is cooling) says the end-market is real. Three things hold the score at 4 rather than higher. The $4 billion target is management's, dated 2026-07-28, and is half-weighted. High-Tech at $1.5 billion is 8.5% of consolidated revenue, so even a quadrupling leaves the company predominantly a dispenser-and-chemistry business selling to restaurants, hotels, hospitals and paper mills. And the remaining 91.5% is a superb but linear compounder whose growth function is price of 2-3% a year plus volume — the knowledge base's own characterisation, at conviction 65, is 'value-based pricing lets Ecolab take 2-3% price annually'. A 4: one real exponential option bolted to an excellent linear machine, bought rather than built, at a price of $4.75 billion."

What could take this further than the base case — and where the market may be underpricing it. Full forward-growth and acceleration math in Deeper analysis, §4 below.

Deeper analysis

Technicals, fundamentals, valuation, and the full expert-claim evidence panel — the detail behind the numbers above.


Reference table

Street consensus$327.33 (+15.6%) · median $332.50 · high $360 (+27.1%) · low $275 (−2.9%) · 29 buy / 7 hold / 1 sell across 37 analysts
Valuation34.7x FY2026E adjusted · 30.1x FY2027E · 26.3x FY2028E (4 analysts) · 35.7x trailing adjusted EPS of $7.94 · 38.0x trailing GAAP · 7.9x book · tangible book is NEGATIVE
Guidance — RAISED 2026-07-28, absent from the payloadFull-year adjusted diluted EPS $8.05–$8.25 (+7-10%), up from $8.03–$8.23 · Q3 2026 $2.13–$2.23 · H2 reported sales +12-14%, organic +6-7%, adjusted operating margin ~19%
The two acquisitionsOvivo Electronics, $1.6bn cash, closed 2025-12-16. CoolIT Systems (Frigeo Holdings LLC, from KKR), $4.75bn, agreed 2026-03-20, closed by 2026-07-28. Funded by $5.0bn of notes (4.600% 2029 $1.2bn; 4.800% 2031 $0.9bn; 5.150% 2033 $1.5bn; 5.350% 2036 $1.4bn) plus an undrawn $4.75bn delayed-draw term facility
Balance sheet, filing-verified 2026-06-30Cash $5,135.3M · short-term debt $1,271.1M · long-term debt $11,905.1M · net debt $8,040.9M pre-CoolIT payment, roughly $12.8bn after · total equity $10,084.4M · goodwill $9,419.3M + intangibles $3,449.1Mtangible book approximately MINUS $2,813M
Q2 2026 resultsSales $4,415.4M, +10% (organic +5%) · GAAP operating margin 17.2%, adjusted 18.3% · GAAP diluted EPS $1.90, +3% · adjusted $2.09, +11% against a $2.08 consensus
ConvictionPositive-moderate, and concentrated in ONE source. 8 name-level claims, all from business_breakdowns, all dated 2025-04-30. Seven bullish. The lane called the data-centre direction fifteen months early
Technicals−8.2% from the 52-week high of $308.35, +15.2% above the low of $245.73; +5.0% above the 50-DMA, +4.5% above the 200-DMA; RSI 61.6; MACD +2.59; 12-month +8.5% vs SPY +24.3%a 16-point deficit

What the experts actually said 8 traceable claims on ECL · showing the highest-conviction voices

“Ecolab is a durable compounder with high switching costs, strong sales culture, and R&D economies of scope; long-term investors rewarded for patience.”
Business Breakdownsbullishconviction 802025-04-30business_breakdowns-P1dNs6LyOAo:8ea761029f
“10,000 fragmented raw materials (largest 4%) limits single-input risk, but contract-renewal lag delays passing through commodity inflation, pressuring near-term margins.”
Business Breakdownsneutralconviction 552025-04-30business_breakdowns-P1dNs6LyOAo:5831314a3b

Every claim reconciles to a real claim_id in the Synthos knowledge base — this is the evidence the verdict is built on, not vibes. Management (the company itself) is shown but half-weighted; one cautionary voice is included on purpose.

Price & moving averages 12 months · 50 & 200-day averages · 52-week range

241259277295313Aug '25Nov '25Jan '26Apr '26Jun '26Aug '2652w hi $308Price 28750-DMA 278200-DMA 27252w lo $246

Solid line = price · dashed line = 50-day average · dotted line = 200-day average · the two thin horizontal lines mark the 52-week high and low. Price above both averages is an uptrend.

Data summary: last close $286.75, 3% above the 50-day average ($278), 5% above the 200-day average ($272) — an uptrend. 7% below the 52-week high of $308, 17% above the 52-week low of $246.

Bollinger Bands 20-day average ± 2 standard deviations

235258281304327Aug '25Nov '25Jan '26Apr '26Jun '26Aug '26Price 28720-day avg 283

The shaded band widens when the stock gets more volatile. Riding the upper edge = strong momentum (sometimes stretched); the lower edge = weak / potentially oversold.

Data summary: price $286.75 is currently inside the band (band $275–$291).

RSI (14) momentum gauge · 0–100

705030Aug '25Nov '25Jan '26Apr '26Jun '26Aug '26RSI 57.4

Above 70 (overbought zone, shaded) = overbought, below 30 (oversold zone, shaded) = oversold. Currently 57.

MACD 12 / 26 / 9 · trend & momentum

0Aug '25Nov '25Jan '26Apr '26Jun '26Aug '26MACD 3.1signal 2.7

The MACD line crossing above the signal line (bars flip to the up color) = momentum turning up; crossing below (bars flip to the down color) = turning down. Bar height = the size of that gap.

Data summary: MACD is currently above its signal line by 0.38, positive momentum.

Relative performance vs S&P 500 & its sector (XLB (sector)), set to 100 a year ago

8695104113122Aug '25Nov '25Jan '26Apr '26Jun '26Aug '26S&P 500 119XLB (sector) 115ECL 103

Solid = ECL · dashed = S&P 500 · dotted = XLB (sector). A rising line means it is beating that benchmark — the sector line shows whether it is a leader or laggard within its own group.

Forward revenue & earnings actual → estimate · "FY" = fiscal year, "E" = estimate

06121723$13BFY21EPS $5$14BFY22EPS $4$15BFY23EPS $5$16BFY24EPS $7$16BFY25EPS $8$18BFY26EEPS $8$19BFY27EEPS $9$20BFY28EEPS $11

Darker bars = actual results, brighter = analyst estimates. Taller bars to the right = expected growth.

Key stats an RIA wants

Price$286.75
Market cap$81B
P/E trailing38×
P/E FY26E / FY27E35× / 31×
EV / Sales5.3×
EV / EBITDA26.8×
Gross margin44.1%
Net margin12.6%
Dividend yield0.99%
Beta0.89
52-wk range$246 – $308
RSI(14)53
50 / 200-DMA$278 / $272
12-mo return+2% (SPY +19%)
Street target$326 ($275–$360)
Analyst grades29 Buy · 7 Hold · 1 Sell
FMP ratingB
Next earnings2026-10-27 (Q3 2026 earnings, 84 days away; vendor consensus EPS $2.18 against management guidance of $2.13-$2.23, on revenue of $4,702,598,000. This will be the FIRST quarter with CoolIT Systems inside it and the first opportunity to see the acquisition's revenue, margin and amortisation drag, none of which has been separately disclosed.)

1. What the business is, and what it just bought

Ecolab reports four segments after a reorganisation effective 2026-01-01: Global Water (formerly Global Industrial, comprising Light & Heavy, Food & Beverage and Paper), Global Institutional & Specialty, Global Pest Elimination and Global Life Sciences, which was "elevated to a standalone reportable segment." The former healthcare operating segment moved into Institutional. Chief executive Christophe Beck; approximately 48,000 employees, of whom more than 25,000 are sales and service and approximately 3,000 are research, development, engineering and digital.

Segment results, filed, from the 2026-07-28 release ($M):

Q2 2026Q2 2025growth (fixed FX)H1 2026H1 2025growth (fixed FX)
Global Water$2,215.5$2,014.9+10%$4,266.4$3,803.7+9%
Global Institutional & Specialty$1,617.4$1,562.5+4%$3,132.2$2,962.6+4%
Global Pest Elimination$350.5$321.2+9%$661.9$598.0+9%
Global Life Sciences$221.0$191.4+15%$421.0$355.9+13%
Total$4,415.4$4,025.2+8% fixed / +10% reported$8,481.5$7,720.2+7% fixed / +10% reported

Operating income, same basis: Global Water $348.8M (+8%), Institutional & Specialty $390.3M (+7%), Pest Elimination $70.6M (+14%), Life Sciences $58.5M (+46%), Corporate −$110.0M. Total $757.9M, +7%.

Every segment grew and the fastest-growing is the smallest. Global Life Sciences is 5.0% of sales and grew operating income 46% in the quarter and 35% in the half. Global Institutional & Specialty — 37% of sales — is the laggard at 4%, and it is the largest single profit pool at $390.3M.

Ovivo Electronics — $1.6 billion, closed 2025-12-16

From the 10-K: "On December 16, 2025, we acquired Ovivo's electronics business ('Ovivo Electronics') for total consideration of $1.6 billion in cash. Ovivo Electronics is a leading and fast-growing global provider of breakthrough ultrapure water technologies for semiconductor manufacturing." Reported within the Light & Heavy operating segment inside Global Water. This is the acquisitionsNet of −$1,621.3M in the payload's fiscal-2025 cash flow, and it explains the $1,319.7M increase in goodwill.

CoolIT Systems — $4.75 billion, and it is nowhere in the payload

From the 10-Q filed 2026-05-07: "On March 20, 2026, Ecolab entered into an agreement to acquire CoolIT Systems for $4.75 billion, subject to certain adjustments. CoolIT Systems is a pure-play data center liquid cooling company that designs and manufactures high-performance liquid cooling systems, including coolant distribution units (CDUs), cold plates and direct-to-chip cooling technologies."

The counterparty is disclosed in the financing 8-K: the merger agreement is "by and among Ecolab Inc., Ecolab U.S. 15 LLC, Frigeo Holdings LLC and KKR Frigeo Aggregator L.P."

The financing, from the 8-K filed 2026-05-29:

> "On May 19, 2026, Ecolab Inc. entered into an underwriting agreement... pursuant to which the Company agreed to issue and sell... $1,200,000,000 aggregate principal amount of its 4.600% Notes due 2029, $900,000,000 of its 4.800% Notes due 2031, $1,500,000,000 of its 5.150% Notes due 2033 and $1,400,000,000 of its 5.350% Notes due 2036... On May 29, 2026, the Company completed the offering. ... The Company intends to use the net proceeds from the sale of the Notes to fund the acquisition of Frigeo Holdings LLC ('CoolIT Systems')..."

$5.0 billion in a single day, with a special mandatory redemption at 101% of principal on three of the four tranches if the acquisition had not closed by 2026-09-16. Separately, the 10-Q discloses a $4.75 billion unsecured committed delayed-draw term loan facility entered 2026-04-10, undrawn as of that filing — a backstop that the notes made unnecessary.

And from the earnings release of 2026-07-28, management's own framing:

> "The recent close of our acquisition of CoolIT further strengthens our Global High-Tech growth engine, extending our leadership in water technologies, research, and services into the rapidly growing AI infrastructure market. CoolIT's growth momentum continues to strengthen, with pre-acquisition sales up more than 100% in the first half. Our Global High-Tech platform is now approaching $1.5 billion in annualized sales, and we expect it to grow to $4 billion in sales by 2030, with operating income margins of 25%. This business is now our largest growth engine and is expected to contribute more than two percentage points to Ecolab's annual sales growth one year after acquisition with increasing margins."

(All of the above from management, half-weighted per house rule. The purchase price, the financing terms and the closing are facts from the filings and are not.)

What is NOT disclosed anywhere in this archive: CoolIT's revenue, its operating income, its margin, or the multiple Ecolab paid. "Pre-acquisition sales up more than 100%" is a growth rate without a base. Global High-Tech "approaching $1.5 billion" is a combined figure including Ovivo Electronics and Ecolab's pre-existing high-technology business. $4.75 billion has been spent and this dive cannot compute what it bought. We say so plainly rather than estimate it.

2. The June quarter and the raised guidance

From the earnings release, all filed ($M except per share):

Q2 2026Q2 2025changeH1 2026H1 2025change
Reported net sales$4,415.4$4,025.2+10%$8,481.5$7,720.2+9.9%
Organic sales$4,282.1$4,090.0+5%$8,239.4$7,913.1+4%
Reported gross margin44.1%44.8%−70bp43.8%44.5%−70bp
Organic gross margin44.9%44.8%+10bp44.5%44.6%−10bp
Reported operating income$757.9$710.1+7%$1,379.9$1,265.4+9%
Special (gains) and charges$51.1$27.1$108.8$61.4
Adjusted operating income$809.0$737.2+9.7%$1,488.7$1,326.8+12.2%
Adjusted operating margin18.3%18.3%flat17.6%17.2%+40bp
Organic operating margin18.8%18.4%+40bp17.8%17.3%+50bp
Interest expense, net$73.1$63.2+15.7%$145.8$121.5+20.0%
Reported diluted EPS$1.90$1.84+3%$3.42$3.25+5%
Adjusted diluted EPS$2.09$1.89+11%$3.79$3.39+11.8%
Adjusted tax rate21.0%20.8%21.0%20.8%

Three observations.

First, reported gross margin fell 70 basis points while organic gross margin ROSE 10. The gap is the acquisitions: Ovivo Electronics and CoolIT carry lower gross margins than Ecolab's base business. That is a permanent mix effect, not a transitional one, and it will widen as High-Tech grows toward the $4 billion target.

Second, the gap between reported and adjusted earnings is widening. Special gains and charges were $51.1M pre-tax in the June quarter against $27.1M, and $108.8M in the half against $61.4M — primarily acquisition and integration costs for Ovivo Electronics and CoolIT. On an after-tax per-share basis the adjustment was $0.18 in the quarter against $0.07. Adjusted EPS grew 11%; reported EPS grew 3%. The 8-point difference is the cost of the acquisition programme, and management's guidance is on the adjusted number.

Third, interest expense rose 15.7% in the quarter and 20.0% in the half — before the $5.0 billion of notes had been outstanding for a full period. The notes were issued 2026-05-29 and carry coupons of 4.600% to 5.350%. On $5.0 billion the annual run-rate is roughly $250 million against a fiscal-2025 total interest expense of $241.1 million. Interest expense is on course to roughly double, and that is arithmetic rather than forecast.

The raised guidance, from the release: full-year 2026 adjusted diluted EPS of $8.05 to $8.25, up 7% to 10%, "higher than prior expectations of $8.03 to $8.23"; third-quarter adjusted diluted EPS of $2.13 to $2.23, up 3% to 8%; second-half reported sales growth of 12% to 14% with organic growth accelerating to 6% to 7%; and second-half adjusted operating margin of approximately 19% with an organic margin of approximately 20%. Both the quarterly and annual ranges "reflect strong underlying performance and... short-term impact from non-cash amortization and financing costs of the CoolIT acquisition"named as a drag, and not quantified.

The vendor consensus of $8.15854 for 2026 sits almost exactly at the midpoint of the guided range, so the estimate row is at least consistent with the company's own view. It does not, however, know that the range was raised.

3. Balance sheet — the leverage step, and a debt figure counted twice

Filed, from the 2026-07-28 release ($M):

2026-06-302025-12-312025-06-30
Cash and cash equivalents$5,135.3$646.2$1,920.9
Accounts receivable, net$3,452.7$3,249.4$3,058.9
Inventories$1,643.9$1,490.4$1,569.4
Property, plant and equipment, net$4,445.8$4,276.6$3,938.8
Goodwill$9,419.3$9,227.0$8,047.2
Other intangible assets, net$3,449.1$3,688.5$3,197.1
Operating lease assets$747.2$765.9$762.8
Total assets$29,927.0$24,696.3$23,736.0
Short-term debt$1,271.1$870.4$688.5
Long-term debt$11,905.1$7,365.9$7,522.2
Operating lease liabilities (non-current)$576.7$596.5$600.6
Total Ecolab shareholders' equity$10,055.5$9,770.8$9,320.3
Noncontrolling interest$28.9$33.7$30.3
Total equity$10,084.4$9,804.5$9,350.6

Total debt at 2026-06-30 is $13,176.2 million against cash of $5,135.3 million — net debt of $8,040.9 million. But the cash is not spendable: it is the note proceeds sitting on the balance sheet waiting for the CoolIT closing. Once the $4.75 billion is paid, pro-forma net debt is approximately $12,791 million — against a rebuilt fiscal-2025 EBITDA of $3,889 million (operating income $2,912.6M plus filed depreciation and amortisation of $976.4M), i.e. roughly 3.3x before any CoolIT earnings contribution, against roughly 2.0x at 2025-12-31 on the filed basis. The true figure lands somewhere between 2.9x and 3.3x and cannot be computed more precisely because CoolIT's EBITDA has never been disclosed. No greater precision is available and none is claimed.

Two balance-sheet facts that matter and are easy to miss.

Tangible book value is negative. Total equity of $10,084.4M less goodwill of $9,419.3M less other intangibles of $3,449.1M is approximately MINUS $2,813 million, or about −$10.00 per share. The vendor's tangibleBookValuePerShareTTM of −$9.897 is consistent with this and is correct rather than defective — it is simply a quantity that cannot be used as a ratio denominator. The 10-K's own risk factors name it: "We incur significant expenses related to the amortization of intangible assets and may be required to report losses resulting from the impairment of goodwill or other assets recorded in connection with the Nalco, Purolite and Ovivo Electronics transactions."

And the payload's debt figure counts the same lease liability twice. Detailed in Section 6, but the arithmetic belongs here: the filed 2025-12-31 debt is $870.4M short-term plus $7,365.9M long-term = $8,236.3M. The vendor reports shortTermDebt $870.4M, longTermDebt $7,962.4M (which is $7,365.9M plus the $596.5M non-current operating lease liability) and separately capitalLeaseObligations of $596.5M, summing to totalDebt of $9,429.3M. The overstatement is $1,193.0M — exactly TWICE the operating lease liability. This is the HCA double-count signature, reproduced to the dollar.

Capital return. The dividend is $2.84 per share trailing, a 1.00% yield, with 33 consecutive years of increases; fiscal-2025 dividends paid were $753.6M. Repurchases were $783.8M in fiscal 2025, and treasury stock rose $669.3M in the first half of 2026 ($9,079.6M to $9,748.9M) — an annualised pace near $1.34 billion, 1.68% of market capitalisation. Total shareholder yield is approximately 2.68%, and it is reasonable to expect the repurchase to slow now that $4.75 billion has left for CoolIT.

4. Free cash flow — capex verified clean, trailing figures unverifiable

Annual capital expenditure reconciles to the filing exactly, in all three years:

($M)FY2023FY2024FY2025
Cash provided by operating activities (filed)2,411.82,813.92,952.6
Capital expenditures (filed)(774.8)(994.5)(1,048.3)
Vendor capitalExpenditure(774.8)(994.5)(1,048.3)
= Free cash flow1,637.01,819.41,904.3
Yield on $79.69bn market cap2.39%

All three years match to the dollar and the fiscal-2025 free-cash-flow figure of $1,904.3M is correct. Recorded as verified clean. The vendor's freeCashFlow field agrees.

The trailing figures, however, cannot be verified, and we do not use them. The vendor's freeCashFlowYieldTTM of 1.738%, operatingCashFlowPerShareTTM of $8.273 and freeCashFlowPerShareTTM of $4.924 imply a trailing operating cash flow of approximately $2,327M and trailing capital expenditure of approximately $942M. Neither can be reconciled, for two reasons. First, the archive contains no June-quarter 10-Q — the most recent quarterly report is the 10-Q filed 2026-05-07 for the March quarter, and the 2026-07-28 8-K carries an income statement and a balance sheet but no cash flow statement. Second, the implied trailing operating cash flow of $2,327M is inconsistent with what the March-quarter filing does show: first-quarter 2026 operating cash flow was $445.9M against $369.4M, up $76.5 million, which on a trailing basis from fiscal 2025's $2,952.6M gives roughly $3,029M at 2026-03-31 — $700 million above the vendor's June-quarter trailing figure. A $700 million single-quarter deterioration is possible but is not evidenced anywhere in the archive, and the depreciation series in the same file is demonstrably corrupt (Section 6). We therefore reject freeCashFlowYieldTTM, priceToFreeCashFlowRatioTTM (57.5x), evToFreeCashFlowTTM (63.8x) and capexToOperatingCashFlowTTM, and use the annual figures only.

The March-quarter filing does show capital expenditure accelerating sharply: $348.5M against $237.9M, up 46.5% year on year. Against fiscal 2025's $1,048.3M, that is a materially higher run-rate, and it compresses free cash flow before any acquisition effect.

On the knowledge base's claim of "90-100% FCF conversion" (conviction 72, 2025-04-30): the filings do not support it on the most recent year. Fiscal-2025 free cash flow of $1,904.3M against net income of $2,075.6M is 91.7% conversion — so the claim holds on that definition. Against ADJUSTED net income, which is the basis management guides to, fiscal-2025 adjusted earnings of $7.53 per share on roughly 284 million shares is approximately $2,139M, giving 89.0% conversion. The claim is broadly right and sits at the bottom of its own stated range. Graded as approximately correct.

5. Valuation — priced in or room?

At $283.14 (market capitalisation $79.69B; 282,423,768 shares issued and outstanding at the 2026-03-10 record date; 282.9 million weighted-average diluted in the June quarter):

TrailingFY2026EFY2027EFY2028E
Consensus adjusted EPS$7.94 (four filed quarters)$8.159 (12)$9.398 (14)$10.774 (4)
EPS growth+8.4% (on FY2025 actual $7.53)+15.2%+14.6%
P/E35.7x34.7x30.1x26.3x
Consensus EPS range$8.089–$8.418$8.958–$9.819$10.496–$11.150
Management guidance$8.05–$8.25
Consensus revenue ($M)16,843 (TTM)17,876 (15)19,268 (15)20,494 (12)
Revenue growth+11.2%+7.8%+6.4%
Price / book (filing $35.73)7.9x
Tangible bookNEGATIVE (~−$10.00)
Free cash flow yield (FY2025 filed)2.39%

Estimate coverage is adequate on the anchor year and thin beyond. 14 analysts on FY2027 EPS, range $8.958 to $9.819, a 9.6% spread — the anchor for all three fair values. FY2028 rests on FOUR analysts and is used only as a cross-check. Note the coverage asymmetry: 12 analysts carry a FY2028 revenue number but only 4 carry an EPS number, which is a direct consequence of nobody knowing what CoolIT's margin is.

est.ebitdaAvg and est.ebitAvg are rejected. In every year from 2023 to 2028 ebitdaAvg is exactly 26.6521% of revenueAvg and ebitAvg exactly 24.8741% — the fixed-ratio fabrication signature, with the additional tell that the implied EBIT margin of 24.88% is six and a half points above the company's own adjusted operating margin of 18.3%. The 2021 and 2022 rows break the pattern in the other direction (ebitAvg at 12.2% and 11.1% of revenue), confirming the forward rows are synthesised rather than collected. All forward valuation uses epsAvg only.

Enterprise value is rebuilt rather than taken. The vendor prints enterpriseValueTTM of $88,304.0M, an $8,617.6M premium to market capitalisation that is neither the payload's own netDebt ($8,783.1M) nor the filed figure. Rebuilt on the 2026-06-30 balance sheet: $79,686.4M + $13,176.2M of debt − $5,135.3M of cash + $28.9M of non-controlling interests = $87,756.2M. Pro-forma for the CoolIT payment, roughly $92.5 billion. netDebtToEBITDATTM of 2.584x is rejected because both its numerator (double-counted leases) and its denominator (corrupt depreciation) are wrong; the correct pre-deal figure is approximately 2.0x and the pro-forma figure approximately 3.0-3.3x.

5a. What today's price assumes (the inversion)

At $283.14 — 34.7x the 2026 consensus and 30.1x 2027 — the price embeds roughly:

5b. The return bridge (why the multiple moves)

Expected return over the next twelve months decomposes as adjusted EPS growth (+15.2%, from FY2026E $8.159 to FY2027E $9.398) + multiple drift + shareholder yield (+2.7%).

Our base assumes the multiple COMPRESSES modestly, from today's 30.1x on the 2027 year to 32.0x — which is an expansion in the arithmetic and a compression in substance, and the distinction needs stating carefully. At $301 on FY2027E consensus of $9.398 the multiple is 32.0x, higher than today's 30.1x on the same year. That is not a re-rating call; it is the mechanical consequence of anchoring the base at a modest premium to the current multiple to reflect the earnings acceleration. The honest version is this: at an unchanged 30.1x on the 2027 consensus, fair value is $283 — spot exactly — and the entire twelve-month return would be the 2.7% shareholder yield plus whatever the estimates do. Most of the return in our base is earnings growth, and the modest multiple premium reflects only the higher growth rate consensus now carries.

Sensitivity, openly: on the FY2027E consensus mean of $9.398, 25x = $235, 28x = $263, 30.1x = $283 (spot), 32x = $301, 35x = $329, 38x = $357.

5c. Variant perception (where we differ, what would surprise)

Synthos fair values

All three anchors are multiples of the FY2027 consensus adjusted EPS distribution — mean $9.39839, low $8.958, high $9.81877, 14 analysts — cross-checked against the FY2026 and FY2028 rows and against management's guided 2026 range.

Base is 6.3% above spot; asymmetry roughly 1.32:1 to the upside (24.1% down, 31.7% up), before a shareholder yield of approximately 2.7%. That is a positive expected value and it is not a Buy, because the single largest variable in it — what $4.75 billion bought — becomes observable in eighty-four days and is unobservable today.

6. Data integrity — what we rejected and why

Seven findings. Ecolab's payload is stale in the way that matters most (it predates $6.35 billion of acquisitions and $5.0 billion of debt) and carries two precise, provable arithmetic defects, against a genuinely clean capital-expenditure series.

1. totalDebt counts the same operating lease liability TWICE — REJECTED and rebuilt. The filed 2025-12-31 debt is $870.4M short-term plus $7,365.9M long-term = $8,236.3M, with non-current operating lease liabilities of $596.5M disclosed separately. The vendor reports shortTermDebt $870.4M, longTermDebt $7,962.4M — which is the filed $7,365.9M plus the $596.5M lease — and, separately, capitalLeaseObligations of $596.5M, giving totalDebt of $9,429.3M. $9,429.3M − $8,236.3M = $1,193.0M, which is exactly 2 × $596.5M. This is the HCA double-count defect class, reproduced to the dollar on a different issuer. Consequential rejections: netDebt $8,783.1M (correct: $7,590.1M), netDebtToEBITDATTM 2.584x, debtToEquityRatioTTM 1.368x (correct on filed debt and equity: 0.842x), debtToAssetsRatioTTM 0.460 (correct: 0.334), and enterpriseValueTTM.

2. The depreciation series drops amortisation in fiscal 2025 only — a definition change in the most recent year — REJECTED. The 10-K reports Depreciation and Amortisation as two lines:

($M)FY2023FY2024FY2025
Depreciation (filed)616.7634.9672.6
Amortisation (filed)306.9300.5303.8
Filed total923.6935.4976.4
Vendor depreciationAndAmortization923.6935.4672.6

FY2023 and FY2024 carry the correct combined figure; FY2025 carries the depreciation line alone, omitting $303.8M — a 31.1% understatement. The same wrong figure appears in inc_a, cf_a and the derived ebitda. Consequence: inc_a.ebitda of $3,461.6M for FY2025 should be $3,765.4M — 8.1% understated — and every EBITDA-based multiple built on it is wrong in the flattering direction.

3. The quarterly depreciation series is corrupt in two of the last four quarters, and the diagnostic fails WITH it — REJECTED.

inc_q D&AVendorFiled / plausibleverdict
Q1 2025$236.1M$236.1M ($161.2 + $74.9)correct
Q2 2025$241.7Mnot separately filedplausible
Q3 2025$16.0M~$245Mgrossly WRONG
Q4 2025$178.8M~$253MWRONG
Q1 2026$263.1M$263.1M ($180.5 + $82.6)correct
Q2 2026$0~$265MWRONG — zero

The four fiscal-2025 quarters sum to $672.6M, exactly the corrupted annual figure, so the error is internally coherent and would survive a cross-file consistency check. The Q1 2026 figure of $263.1M reconciles exactly to the 10-Q's separate depreciation ($180.5M) and amortisation ($82.6M) lines, and Q1 2025's $236.1M to $161.2M and $74.9M. So the vendor CAN construct the field correctly and does so in two of six quarters.

Consequence, and it is the important one: capexToDepreciationTTM of 2.0577 is built on a trailing depreciation of $457.9M — which is precisely the sum of the four corrupt quarters ($16.0 + $178.8 + $263.1 + $0.0 = $457.9). The 10-Q filed 2026-05-07 discloses trailing depreciation of $691.9M and amortisation of $311.5M at 2026-03-31, i.e. $1,003.4M. On the filed basis the capex-to-depreciation ratio is approximately 0.94x, not 2.06x — a factor of 2.2. A reader taking 2.06x at face value would conclude Ecolab is investing at more than twice its depreciation rate; the filings say it is investing at slightly less. Note the contrast with two other names in this batch, where the diagnostic was clean and the datum corrupt: here the diagnostic and the datum fail TOGETHER, because the diagnostic inherits the corrupt series. The two failure modes are both real and neither can be assumed.

4. seg_geo has correct VALUES and corrupted LABELS — a new variant, and it hides China. The block ties exactly: FY2025 United States $8,480.2M + Europe $3,275.3M + Asia Pacific $1,364.5M + Latin America $1,101.2M + "C [N]" $722.4M + "Middle East And Africa" $638.2M + Canada $499.4M = $16,081.2M, the filed revenue, to the dollar. Every number is right. Cross-checking against the 10-K's segment-by-geography tables confirms each line: Greater China is $410.4M (Global Water) + $198.1M (Institutional & Specialty) + $61.3M (Pest) + $52.6M (Life Sciences) = $722.4M. So "C [N]" is Greater China with a mangled label. Separately, the filing's category is "India, Middle East and Africa" and the payload truncates it to "Middle East And Africa", dropping India from a $638.2M line. The practical consequence is specific and worth naming: an automated screen searching this payload for China exposure returns nothing, on a company with 4.5% of revenue there. This is a milder failure than the eight seg_geo defects found in the previous batch — the values are usable — but it is a failure, and the correct labels are used throughout this dive.

5. seg_prod is discontinuous at the 2026-01-01 segment reorganisation — usable for FY2025, not comparable to FY2024. The FY2025 entry carries the new segments (Global Water $7,982.4M, Global Institutional and Specialty $6,104.6M, Global Pest Elimination $1,246.3M, Global Life Sciences $747.9M), summing to $16,081.2M — the filed revenue, exactly. The FY2024 entry carries the old segments (Global Industrial, Global Healthcare and Life Sciences), which the 10-K states were renamed and restructured "effective January 1, 2025" with "the Company's former healthcare operating segment moved into the Institutional operating segment" and Global Life Sciences "elevated to a standalone reportable segment." The FY2024 row also sums to $15,873.0M against a filed $15,741.4M — 0.84% over. Year-on-year comparison within this block is invalid, and all segment growth figures in this dive come from the filed release, which restates both years on the current basis.

6. researchAndDevelopmentExpenses disappears in fiscal 2025. inc_a reports R&D of $185.0M (2020), $186.0M (2021), $190.0M (2022), $192.0M (2023), $207.0M (2024) and $0 (2025). researchAndDevelopementToRevenueTTM is likewise 0. The 10-K states the company employs approximately 3,000 research, development, engineering and digital experts. R&D did not go to zero. Not used.

7. est.ebitdaAvg and est.ebitAvg carry the fixed-ratio fabrication signature and contradict the company's disclosed margin — REJECTED. In every year from 2023 to 2028 ebitdaAvg is exactly 26.6521% of revenueAvg and ebitAvg exactly 24.8741%. The implied forward operating margin of 24.88% is 6.6 percentage points above the company's own adjusted operating margin of 18.3% in the most recent quarter and above the ~20% organic figure management guides to for the second half. The 2021 and 2022 rows break the pattern (11-12% of revenue), confirming synthesis. All forward valuation uses epsAvg only.

Verified clean — recorded, because a clean field is a finding:

Peer set — REJECTED as unusable. The vendor peers are Agnico Eagle, Air Products, Barrick, BHP, CRH, Freeport-McMoRan, Hawkins, Newmont, Sherwin-Williams and Vale. Four are gold or diversified miners and one is an iron-ore producer. Only Air Products, Hawkins and Sherwin-Williams are chemicals companies of any description, and none is a water-treatment or hygiene-services business. Veralto, Xylem, Diversey's successor and Rentokil are all absent. No peer-multiple comparison is drawn. (Sherwin-Williams appears here and was one of the six confirmed operating-lease debt-inflation cases in the previous batch, which is a reminder that a peer set can be simultaneously irrelevant and populated by names with the same defect.)

Vendor composite rating — noted, not used. rating gives B / 3 overall, with 4 on discounted cash flow, 5 on return on equity, 5 on return on assets, 1 on debt-to-equity, 1 on price-to-earnings and 1 on price-to-book. The debt-to-equity sub-score is computed on the double-counted debt figure rejected in finding 1; on the filed figures the ratio is 0.842x, not 1.368x. The price-to-book sub-score of 1 reflects a genuine fact — 7.9x book on a balance sheet where goodwill exceeds equity — and points the same way this dive does.

Non-equity tripwire — checked and passed. ECL is common stock, NYSE-listed, with 282,423,768 shares issued and outstanding at the 2026-03-10 record date per the annual-meeting 8-K. Beta 0.895; the dividend has been raised for 33 consecutive years; volume 843 thousand shares (~$239M of turnover — the thinnest in this batch); the 52-week band of $245.73 to $308.35 is a 25% range, the narrowest here. This is common equity.

7. Knowledge base — eight claims, one source, one day, and it was right

Raw hits: 21. Exact case-sensitive entity matches: 8. Name-level claims on Ecolab: 8. Bullish: 7. Neutral: 1. Text-only, discarded from scoring: 13.

All eight name-level claims come from business_breakdowns and all eight are dated 2026's predecessor year — 2025-04-30. They are a single company breakdown distilled into eight structured claims. The formal concentration test therefore fails outright: removing one channel removes 100% of the name-level lane. We report kb_breadth as 1, not 8, and we do not treat seven bullish claims as seven independent votes.

The eight, verbatim:

> *conviction 80 · thesis · "Ecolab is a durable compounder with high switching costs, strong sales culture, and R&D economies of scope; long-term investors rewarded for patience."*

> conviction 80 · thesis · "Installed hardware, embedded chemistry, deep field-sales relationships and 3-5yr contracts create very high switching costs; Ecolab can effectively name its price."

> conviction 78 · thesis · "Data-center cooling and semiconductor fabs use enormous water volumes, driving strong demand for Ecolab's industrial water treatment; ~70% of sales touch water."

> conviction 72 · thesis · "Recession-resistant razor-and-blade model with 90-100% FCF conversion, 33 years of dividend raises, and ~20% long-term ROE."

> conviction 68 · thesis · "Disciplined capital allocation: culture-first bolt-on M&A, willingness to divest underperformers (Champion X), balanced buybacks and dividends."

> conviction 65 · fact · "Value-based pricing lets Ecolab take 2-3% price annually (up from 1-2%), reaching 20% operating margin by 2027 from ~18% today."

> conviction 60 · thesis · "Targets 5-7% annual revenue growth with large cross-sell (~$55B) and untapped (~$81B) opportunity, though weak fit with small mom-and-pop customers."

> conviction 55 · NEUTRAL · "10,000 fragmented raw materials (largest 4%) limits single-input risk, but contract-renewal lag delays passing through commodity inflation, pressuring near-term margins."

Four of the eight are gradeable against the filings and three of the four have aged well.

The data-centre claim (conviction 78) is the best-aged claim encountered in this batch. Recorded 2025-04-30. Eight months later Ecolab bought Ovivo Electronics — ultrapure water for semiconductor fabs — for $1.6 billion. Fifteen months later it closed a $4.75 billion acquisition of a data-centre liquid-cooling manufacturer and told shareholders the combined platform "is now our largest growth engine." The claim identified the strategic direction before the company took it. That is what this store exists to do and it is recorded as a hit.

The margin claim (conviction 65) is on track. It predicted "reaching 20% operating margin by 2027 from ~18% today." The June-quarter adjusted operating margin was 18.3%; the organic figure was 18.8%; management now guides to approximately 19% adjusted and approximately 20% organic for the second half of 2026 — a year early on the organic measure. Tracking.

The switching-cost and pricing claims (conviction 80 and 65) are consistent with the filings — 25,000 sales and service employees, installed dispensing equipment, multi-year contracts, and a raw-material base of 10,000 inputs of which the largest is 4%, which the 10-K corroborates.

The free-cash-flow claim (conviction 72) is at the bottom of its own range and is graded, not endorsed. "90-100% FCF conversion" against a fiscal-2025 free cash flow of $1,904.3M on net income of $2,075.6M is 91.7% — inside the range, at its floor, and before a first quarter of 2026 in which capital expenditure rose 46.5% year on year. The 33-year dividend record and the ~20% long-term return on equity in the same claim are both correct (21.4% trailing).

And one claim now reads against the company. "Disciplined capital allocation: culture-first bolt-on M&A" (conviction 68) describes a company that makes bolt-on acquisitions. A $4.75 billion purchase is 6.0% of the market capitalisation and is not a bolt-on. The claim's characterisation of Ecolab's M&A style has been superseded by Ecolab's own behaviour, and this dive treats that as a live question rather than as support.

The thirteen text-only hits carry zero weight and none names Ecolab. They are dominated by a genuinely relevant theme — liquid cooling as the physical constraint on artificial intelligence — and three of them come from jordi_visser at skill 2.0, the highest weight in the store:

> 2026-05-11 · bullish · conviction 80 · speaker: Jordi Visser · skill 2.0

> "The physical world was underinvested for this moment; agentic inference buildout needs power, liquid cooling, chemicals and optical fiber — all early-inning physical plays."

> 2025-11-19 · neutral · conviction 72 · channel: jensen_huang

> "Nearly all datacenter rack mass (1.95 of 2 tons) is cooling; GB300 compute itself is tiny, underscoring power/cooling as the constraint."

These describe the end-market Ecolab has just paid $4.75 billion to enter and they are supportive of it. They name no company in this file, they enter no score, and they are quoted for context only. The jensen_huang material is promoter-class under the four-lane policy in any case.

Homograph check. ECL is a three-letter ticker and is also the standard abbreviation for "expected credit loss" in bank and insurer disclosures, which makes it a documented collision hazard of exactly the CEG/PWR/EMR type. The case-sensitive entity run held and produced no false positives; all eight exact matches carry both Ecolab and ECL in the entity list. No case-insensitive-only matches were found at all, which is the cleanest possible result for a ticker of this shape.

Attribution note. All eight name-level claims carry speaker: null and are attributed only to a channel — the weakest sourcing the four-lane policy admits. No claim carries a management speaker_role on Ecolab.

Conclusion. Breadth 1, claim count 8, net conviction positive-moderate. The Synthos knowledge base is positive on Ecolab, on a single well-constructed analysis from April 2025 that has aged unusually well and that anticipated the company's strategic direction by more than a year. It is one source on one day and it is scored as one source on one day. It supports the business quality argument in this dive and it does nothing to answer the question the verdict turns on, which is what $4.75 billion bought.

8. Technicals

9. Insiders — the chief financial officer exercised and sold, everything else is a director grant

DatePersonRoleTypeSharesPriceHeld after
2026-07-30Scott D. KirklandChief Financial OfficerM-Exempt (acquire)3,972$117.7320,881.412
2026-07-30Scott D. KirklandChief Financial OfficerS-Sale3,357$276.9617,524.412
2026-07-30Scott D. KirklandChief Financial OfficerM-Exempt (dispose)3,972$117.730
2026-06-30John J. ZillmerdirectorA-Award130.18$049,365.43
2026-06-30Julie WhalendirectorA-Award130.18$0471.76
2026-06-30Suzanne M. VautrinotdirectorA-Award130.18$011,805.02
2026-06-30Lionel L. Nowell IIIdirectorA-Award130.18$06,773.07
2026-06-30Tracy B. McKibbendirectorA-Award130.18$011,667.65

Five of the eight transactions are identical 130.18-share director dividend-equivalent awards dated 2026-06-30 and carry no information whatever.

The one decision in the file is the chief financial officer's. Scott Kirkland exercised 3,972 options at $117.73 and sold 3,357 shares at $276.96 on the same day, 2026-07-30 — two days after the second-quarter release and the raised guidance — retaining 17,524.412 shares. That is an 84.5% same-day disposal of the exercised position, at a price 2.2% below today's close.

The honest reading. A same-day exercise-and-sell at a $117.73 strike is the standard mechanism for realising an in-the-money option and carries limited signal; the officer retained more than 17,500 shares, roughly five times what he sold. But it is the chief financial officer, it is two days after a guidance raise, and it is the only non-mechanical transaction in the file. There is not one open-market purchase by anyone. Reported, weighted lightly, and not built into any conclusion.

10. Verdict, kill-criteria and flip conditions

Watch.

What is right about this company, stated first. Ecolab is one of the highest-quality industrial franchises in the market. Gross margin 44.1%; adjusted operating margin 18.3% and rising; return on equity 21.4%; four segments all growing; thirty-three consecutive years of dividend increases; 25,000 sales and service people embedded in customer sites under multi-year contracts; a raw-material base so fragmented that the largest single input is 4% of the total. The June quarter was good — sales +10%, organic +5%, adjusted EPS $2.09, +11% — and management raised full-year guidance to $8.05-$8.25. Consensus has EPS accelerating 8.4%, 15.2%, 14.6%. And the strategic direction is, on the evidence available, correct: data centres and semiconductor fabs consume enormous quantities of water and are constrained by cooling, which the knowledge base said in April 2025 and which Ecolab has now spent $6.35 billion acting on.

Why this is not a Buy — three reasons, in order of weight.

First, and decisively: $4.75 billion has been spent and this dive cannot compute what it bought. CoolIT Systems' revenue, operating income and margin are not disclosed anywhere in this archive. What is disclosed is a growth rate without a base ("pre-acquisition sales up more than 100%"), a combined platform figure that includes two other businesses ("approaching $1.5 billion in annualized sales"), and a 2030 target ("$4 billion... with operating income margins of 25%"). The programme's own precedent is explicit: a name that cleared 2.19:1 on arithmetic was declined in the last batch because a $6.25 billion cash acquisition six days old was unquantified. This is $4.75 billion, weeks old, and equally unquantified. The difference is that here the first quarter containing it is dated: 2026-10-27.

Second, the price carries no cushion. 34.7x the 2026 consensus and 30.1x 2027, against a base of $301 that is 6.3% above spot with 1.32:1 asymmetry and a 2.7% shareholder yield. Free cash flow was $1,904.3M in fiscal 2025 — a 2.39% yield — and first-quarter capital expenditure ran 46.5% above the prior year.

Third, leverage is roughly doubling into it. $5.0 billion of new notes at 4.600% to 5.350% takes pro-forma net debt to approximately $12.8 billion against a rebuilt fiscal-2025 EBITDA of $3,889 million — roughly 3.0-3.3x against roughly 2.0x — on a balance sheet where goodwill of $9,419.3M already exceeds total equity of $10,084.4M and tangible book value is negative by approximately $2,813 million. Interest expense is on course to roughly double.

And one fact that argues the other way and belongs in the verdict rather than buried: Ecolab has underperformed the S&P 500 by roughly 16 points over twelve months while raising guidance twice and executing the most strategically interesting capital allocation of its recent history. That is the shape of a name that becomes buyable, and it is why this is a Watch with a dated trigger rather than a Hold.

Pre-registered UPGRADE conditions — what would take this to Buy — Tactical:

Pre-registered KILL criteria — what would take this to Hold or Avoid:

Where ECL fits in the Synthos Framework Portfolio. Not entered today. Placed on the watch list with a trigger price of ~$265 and an event trigger of the 2026-10-27 CoolIT disclosure, either of which would support a 1.5-2% entry in the industrials/materials sleeve. Sizing note: this is a high-quality compounder that has just made an out-of-character acquisition six percent of its own size into the most crowded theme in the market, and the correct response is to wait one quarter for the disclosure rather than to pay 34.7x forward for an earnings stream nobody has seen. Logged as a tracked Synthos call (Watch) as of 2026-08-04 at $283.14, with the fair-value anchors, kill criteria and upgrade conditions all gradeable.

Single biggest risk: that $4.75 billion has been paid for the top of a capital-expenditure cycle. CoolIT sells coolant distribution units, cold plates and direct-to-chip cooling into data centres, in a market whose growth rate — "more than 100% in the first half" — is precisely the kind of number that attracts a full price and then decelerates. Ecolab funded it with $5.0 billion of fixed-rate debt maturing between 2029 and 2036, which is permanent, while the revenue it bought is not contractually recurring in the way the rest of Ecolab's business is. The company's own 10-K risk factors describe the mechanism: intangibles are amortised, excess purchase price becomes goodwill, and if future cash flows fall short the goodwill is impaired. Ecolab already carries $9,419.3 million of goodwill against $10,084.4 million of total equity. A CoolIT impairment would not threaten the company — the base business generates $1.9 billion of free cash flow and has raised its dividend for 33 years — but it would end the re-rating this price is anticipating, and at 30.1x the 2027 consensus the equity has no protection against it.


Provenance & disclosures