IDEXX Laboratories

IThe business

IDEXX makes the machines, tests and software that a vet uses to find out what is wrong with a dog or cat. It is headquartered at One IDEXX Drive in Westbrook, Maine, and employs about 11,000 people who serve customers in more than 175 countries.²⁴

The business has one idea in it, and it is worth getting clear before any number. The company places an analyzer in a clinic. The clinic then buys a stream of single-use test consumables for it, year after year. The 10-K says the multi-year consumable stream is “significantly more valuable than the placement of the instrument.”⁵ That is the razor and the blade. About 79% of 2025 revenue was recurring diagnostics: consumables, outside laboratory work, rapid tests, service and accessories.⁵

Companion animal work (CAG) was 91.9% of revenue in 2025 C (calculated).⁵ The rest is water-testing kits sold to utilities, tests for livestock, poultry and dairy, and a small human diagnostics line. The US brought in 64% of revenue and everywhere else 36% C (calculated).⁵ The same wording appears in every filing: a veterinarian is the customer, and a pet owner is the person who pays the vet.

The four ways it earns

In-clinic analyzers and consumables. The Catalyst chemistry analyzers, the ProCyte hematology machines, the SediVue urine analyzer and the new inVue Dx cell analyzer sit in the practice and give a result while the owner waits. IDEXX VetLab consumables were $1,497M in 2025, the largest line, and grew 13.7% organically.⁵ The Catalyst base is over 80,000 machines worldwide, from about 77,000 a year earlier.⁶⁹

Reference laboratories. Samples go by courier to IDEXX labs in the US and abroad, which return a wider menu of tests and specialist reads, usually the next day. This was $1,424M in 2025, growing 5.9% organically, and 10.2% in the third quarter of 2026.⁵² It is a service business with trucks, scientists and fixed costs.

Rapid tests. The SNAP kits for heartworm, tick-borne disease and similar conditions are mature. Revenue fell 3.3% organically in 2025 as a pancreatic test moved onto the Catalyst platform, and rose 0.7% in the third quarter.⁵²

Software and imaging. Practice management systems such as Cornerstone, ezyVet and IDEXX Neo, plus digital X-ray and a telemedicine service, made $346M in 2025.⁵ In September 2026 IDEXX bought CoVet, an AI-based ambient listening and workflow tool for clinics, for about $250M including a $20M contingent payment.⁴ The release puts about $5M of CoVet revenue in the 2026 guide for the last three and a half months, which is roughly $17M a year and a price near fifteen times revenue C (calculated).² A third-party transcript gives a $20M run rate, but it is not clear whether that means CoVet or ezyVet, so I use the release.⁶

What the third quarter showed

Revenue was $1,214M, up 10%, with CAG Diagnostics recurring revenue up 11% organically.² The mix matters. Consumables grew 14.6% organically, on net new customers, a bigger base of instruments and what the company calls net price gains.² Reference lab revenue grew 10.2%. Instruments were flat at $51.5M, even though over 1,900 inVue Dx machines were placed in the quarter.² Placements under commitment contracts are recognised over time, so a big quarter of installs shows up in later consumables more than in instrument revenue. The company raised its 2026 revenue guide to $4,741M to $4,767M, and EPS to $14.96 to $15.16.²

The stock jumped on the day. It closed at $551.21 on 30 October and $671.64 on 2 November, a rise of 21.8% C (calculated).¹ It closed at $698.55 on 4 November, another 4% on.¹ In three sessions the shares gained about $11.6B of market value C (calculated).

The visit problem

The central fact about IDEXX in 2026 is that US vet visits are falling while its US revenue rises. On the call, management put US same-store clinical visits down about 1.8% in the third quarter, and said US CAG Diagnostics recurring revenue still grew 9.6%, outpacing visits by more than 1,100 basis points.⁶ The release rounds that US figure to 10%.² A year earlier the visit figure was −1.2%.⁹ Both numbers are from third-party transcripts, and the call is the only place the company puts them; the release points readers to a separate earnings snapshot for US practice metrics.²

How does revenue rise when visits fall? Management gave three reasons: more tests per visit, a wider menu, and price. The global net price gain was about 4% in the third quarter and is about 4% in the full-year outlook.⁶ Visits are a gauge of the number of times a pet is seen. Revenue per visit is up by a double-digit rate. That can last while owners keep paying, and it is exactly the point I can’t judge, so I come back to it in Valuation.

The newer products, sized

The company talks most about inVue Dx, Cancer Dx and FNA, so I tried to put sizes on them. inVue Dx is a bench-top analyzer for blood cells. The installed base reached over 9,000 at the end of June and nearly 11,000 at the end of September.¹⁰² The company has a goal of 20,000 placements over five years and expects to beat the 5,500 it wanted for 2026.⁶ On the call, management gave $3,500 to $5,500 of recurring revenue per analyzer a year, and the transcript elsewhere says $5,100 for the top of the range, so I use the wider one.⁶⁹ On that range, the 11,000 installed bring in $38M to $61M of recurring revenue a year, and 20,000 would bring in $70M to $110M C (calculated). That is 1.5% to 2.3% of this year’s revenue. It is a good product and a small number.

Cancer Dx is a blood test that screens dogs for lymphoma, with mast cell tumour detection added in late September to make it a multi-cancer panel. It launched abroad in the first quarter.¹¹ It costs about $15 when run with a standard blood panel, and over 14,000 customers have ordered it.⁶ At $15 a test it needs millions of tests to matter. I estimate a million tests a year is $15M E (estimate). The real prize is the pull-through: a vet who trusts the panel orders more bloodwork. I can’t measure that.

So the case does not rest on the launches. It rests on the older machines, the Catalyst and the lab network, keeping their pull while price rises and visits fall. The launches are the reason to think the company still has things to sell, and they sit on top.

IIThe moat

The protection has five parts, and they work together. A clinic that owns an IDEXX analyzer pays IDEXX for every test it runs, so leaving means replacing a machine, retraining staff and reworking the clinic’s routine. The company says customer retention is in the high 90s, which I take from the call and can’t check in a filing.⁶

Evidence it is widening

The installed base keeps growing at a double-digit rate. The premium base grew 12% over 2025 and 11% over the twelve months to September 2026.⁵⁴ In the third quarter the company placed about 5,780 premium instruments, including about 1,200 Catalyst machines to competitive or new customers, according to the call.⁶ Placements that win customers from rivals are a sign the protection is widening, and 1,200 in a quarter is a respectable number.

Gross margin tells the same story over a longer time. It went from 54.9% in 2016 to 61.8% in 2025 and was 62.9% in the third quarter of 2026 C (calculated).¹²² A company without pricing power doesn’t add seven points of gross margin in nine years while it also grows revenue 10% a year.

The cancer panel offers a small sign. Nearly a quarter of third-quarter Cancer Dx customers were clinics that did not use IDEXX as their primary reference lab, the company said on the call.⁶ A test that pulls in lab business from rivals is a way to widen the reference-lab position.

Evidence it is narrowing

The pressure is on the customer side. The 10-K says the number of owners of vet hospitals in the US is falling and more hospitals are owned by companies. It names Mars, the owner of Banfield, BluePearl and VCA, and National Veterinary Associates.⁵ A chain buys for many clinics at once, and it can negotiate. The same filing says consolidation “may intensify the competition” and that vertical integration among customers could hurt the company’s ability to compete.⁵ Mars also owns Antech and Heska, which sell the competing lab and in-clinic products.⁵ So one of IDEXX’s largest groups of customers is owned by a competitor.

Visits are the second pressure. They fell about 1.8% in the third quarter, and the pet population is not helping: puppy and kitten adoption is described as muted.⁶ The business has beaten visits by more than 11 points of growth for four quarters, which is a measure of how much pricing and testing intensity have carried it.⁶ If the protection is wide, that can go on. If clinics start to push back on cost, it is where it would show first.

The third is reward. A 50% return on invested capital, which is what IDEXX reports for 2025, tells every competitor where to look.⁷ The company defines that figure itself, excluding cash and investments, so I treat it as a guide.

IIIThe competitive advantage

I could not put IDEXX next to its main rivals with a table of margins, and I would rather say that than invent one. Mars is private. Zoetis reports diagnostics inside a much larger animal-health business, and I did not open its filings for this page. So I give the 10-K’s list of who competes where, and then IDEXX’s own numbers.

Where IDEXX wins

It wins on the economics of the installed base. For every dollar of instruments the company sold in 2025 ($200M), it sold about $7.50 of VetLab consumables ($1,497M) C (calculated).⁵ Company segment results show the effect: CAG made an operating margin of 33.1% in the third quarter and 33.5% over nine months, up from 32.9% a year earlier.² Water made 50.3% in the quarter, a small line with an unusually good margin, and livestock made 5.5%.²

It wins on return on capital. The company reports ROIC of 50.2% for 2025, and its bonus plan targets 47.8%.⁷ My own check is simpler. Operating income of $1.36B in 2025 sat on equity of $1.61B and net debt of about $0.7B at the year end C (calculated).⁵ The business needs little capital to run, whatever the share price.

It wins on reach. Innovations come from one R&D budget of $251M a year, 5.8% of revenue, spread over about 175 countries of sales.⁵² A competitor with a smaller base has fewer instruments over which to spread each new test.

Where it is exposed

It is exposed to its customers. When Mars owns clinics, and Mars sells competing diagnostics, the chain has a reason to move work in-house. The 10-K does not say how much of IDEXX’s revenue comes from chains, and I could not find that number.⁵ The company does say no single customer made up 10% of revenue in 2025.⁵ Erickson himself led the company’s strategic accounts team from 2018 to 2022, the period when corporate ownership rose “significantly,” per the proxy, so the company knows this problem well.⁷

It is exposed to one supplier. The Catalyst chemistry slides come from Ortho, now part of QuidelOrtho, under agreements to 2044 with an early-termination right in some cases.⁵ The 10-K names Ortho as a sole-source supplier whose products cannot easily be replaced.⁵ I treat this as a tail risk. It is about the biggest line in the business.

It is exposed to a slow rise in the cost of the checkup. The rapid-test line is already shrinking, and the company has a reason to keep a close eye on whether higher prices change how many tests vets order. In the third quarter, rapid assay volumes fell, while price rose.⁴

The newest platforms

The next point-of-care analyzer is called MultiCue Dx, and the company has scheduled an investor session about it for 18 January 2027, at the VMX conference in Orlando.² The third-party transcript spells the name Multi-Q, which I take to be a transcription error, and gives no detail.⁶ I don’t know what it does, so it is outside my model. If it is a platform on the scale of Catalyst, it matters a lot. If it is another menu extension, it matters little.

IVManagement

The top job changed hands in May. Jay Mazelsky was CEO from 2019 and stepped down when the 2026 annual meeting adjourned on 12 May. Mike Erickson, 52, succeeded him. Mazelsky stays as Executive Chair for a year, until his retirement in May 2027.⁷ Lawrence Kingsley, who had been the independent chair, became lead independent director when Mazelsky moved to Executive Chair.¹³ Erickson has been at IDEXX since October 2011, after eight years at McKinsey. He ran the strategic accounts team that deals with corporate clinic groups, and since 2023 he has run in-clinic diagnostics, which includes the inVue Dx programme.⁷ He is an insider with a strong record on the products the company is selling, and he has been CEO for about six months.

How they are paid, and what that pays them to do

Erickson’s terms are in an 8-K of 13 January. He gets a $1.0M salary, a bonus target of 120% of salary, and a 2026 equity grant of at least $7.8M, half options and half performance share units. If he is dismissed without cause he gets two years of salary and two more years of equity vesting.¹³ That is a lower package than his predecessor’s, at about $10.0M a year on target C (calculated). The 2026 bonus and share-unit metrics will appear in next year’s proxy. I have not seen them, so I read the 2025 design, which is the one in force for Mazelsky, as the likely guide.

Mazelsky’s 2025 pay was $14.58M: $1.15M of salary, $5.40M of stock awards, $5.40M of options and a $2.59M bonus. That is 218 times the median employee’s $66,956.⁷ At target, 80% of the CEO’s pay is long-term equity, split evenly between options and performance share units.⁷

The one-year bonus is weighted 40% to organic revenue growth, 20% to operating profit, 20% to diluted EPS and 20% to ROIC. In 2025 organic growth was 9.6% against a target of 8.1%, operating profit $1,320.6M against $1,307.7M, EPS $12.49 against $12.25 and ROIC 48.3% against 47.8%.⁷ The financial score was 132%, the committee scored the non-financial goals at 153%, and the committee added a 24% personal factor for Mazelsky, for a payout at 173% of target.⁷ The performance share units pay on a three-year average of organic revenue growth and comparable operating profit growth, half each, with payouts from 20% to 200%.⁷

So what does that pay them to do? Grow revenue and profit, and hold capital efficiency. That is a sensible mix, and ROIC is the right piece to include. Two things stand out. First, EPS is a fifth of the bonus, and EPS rises when the company buys back shares at any price, including 46 times earnings C (calculated). Second, nothing in the plan looks at cash after the costs of standing still, or at the price paid for the company’s own stock. A team can meet every target and still spend $800M on buybacks at a rich price. I do not say they have. I say the plan would not notice.

How much of their own money is in

The ownership guideline asks the CEO to hold ten times salary and each executive vice president four times.⁷ At 8 March 2026, Mazelsky owned 83,803 shares, about $58.5M at today’s close, plus rights over another 238,917 through options and share units that could vest or be exercised within 60 days C (calculated).⁷ All 20 directors and executive officers together owned 0.79% of the company.⁷ Erickson is not in the table of named officers. His Forms 4 show about 15,100 shares after his February transactions, near $10.5M at the close, and he sold 5,371 shares that month after exercising options struck between $141.60 and $178.26 C (calculated).⁸

The record of what they did with it is one-way. In the year from 3 November 2025, the Forms 4 show six insiders selling 57,585 shares worth $37.7M and no open-market purchase by anyone C (calculated).⁸ Mazelsky’s part was 23,326 shares at about $658, all from options exercised at $141.60 on the same days.⁸ That is $37.7M against a $55B company, so size is not the point. The pattern is. On 4 November three officers filed notices of proposed sales for 38,025 shares, about $26.6M at the close, led by Mazelsky’s 25,140 shares.¹⁴¹⁵¹⁶ A notice is not a sale. It is a sign that people who know the business best are willing to sell right after the shares rose 27% in three sessions.

What they did with capital over a cycle

From 2016 to 2025 IDEXX earned $6.09B of net income and produced $4.96B of owner cash, 82% of it C (calculated).¹² It spent $5.13B on its own shares, 103% of owner cash, and $0.35B on acquisitions C (calculated).¹² It pays no dividend. Diluted shares fell 11% in nine years, from 90.9M to 81.0M, and shareholders’ equity went from −$0.11B to $1.61B C (calculated).¹²

Was it good judgement? It depends on price. Over 2021 to 2025 the company retired 9% of its shares at an average of $488, and the same shares are worth $698.55 today, 43% more.⁷ That has worked. The nine months of 2026 cost $588.62 a share, and $563.01 in the third quarter, with the stock now 19% and 24% above those levels C (calculated).² On my base value of $306, each dollar spent at $588.62 bought about 52 cents of value C (calculated). I give that figure with the usual warning: it uses my discount rate, and a lower one would raise it.

The timing record is uneven. The stock fell 54% from August 2021 to October 2022, and the company spent $747M in 2021 and $820M in 2022 into the fall C (calculated).¹¹² It then spent only $72M in 2023, as the shares recovered from the low, followed by $837M in 2024 and $1.22B in 2025 at higher prices.¹² I don’t read that as an attempt to time the market. It reads as a steady program funded by cash flow and some borrowing, with the amount set by what the business throws off.

Acquisitions are small. In ten years it spent about $352M on businesses, then $227M in the first nine months of 2026, most of it for CoVet C (calculated).¹²² Paying about fifteen times revenue for a tool that size is the kind of price I watch. It is also a small bet next to the $816M spent on shares in the same period.

VFinancials

I use the ten 10-Ks from 2016 to 2025, which the SEC provides in structured form, and then the first nine months of 2026.¹²⁴ Owner cash here is operating cash flow less capital spending less stock pay. It is a rougher measure than I use in Valuation, where I adjust for tax timing and one-off payments.

Revenue grew from $1.78B in 2016 to $4.30B in 2025, or 10.3% a year over nine years C (calculated).¹² Operating income grew 16.3% a year and EPS 20.5% a year C (calculated). The gap between the three is margin and share count, in that order. Operating margin went from 19.7% to 31.6%, and diluted shares fell 11%. Revenue rose in every one of the nine years since 2016, including 2022 and 2023, when operating margin fell and then recovered.¹²

Two years need a note. In 2022 operating margin dipped to 26.7%, alongside research spending that rose to $0.255B from $0.161B and then dropped back to $0.191B in 2023. I have not traced the cause and do not want to guess.¹² In 2024 margin dipped to 29.0% on a litigation accrual the company puts at $0.56 of EPS, and the 2025 margin of 31.6% includes the lapping of it. Comparable operating margin rose 90 basis points in 2025, against 270 reported.³

Earnings against cash

Net income over ten years was $6.09B and owner cash $4.96B, 82% of it C (calculated).¹² The gap has three sources. Stock pay is 7% of the earnings total. The company funds instrument placements with up-front incentives and contract assets, which grew $36M in the first nine months of 2026.⁴ And receivables grow with sales.

The cash numbers move around with taxes. In 2025 operating cash flow was $1.182B, but it included $131M of deferred taxes, a timing benefit from new US rules on research deductions, and it was reduced by about $80M paid to settle a litigation matter accrued in 2024.⁵ Remove the first and add back the second and the 2025 owner cash is $0.95B, not $1.00B, or 89% of net income instead of 94% C (calculated). In the first nine months of 2026 the deferred-tax line was only $19M, against $100M a year earlier.⁴ So 2026’s cash conversion is cleaner, and the company guides free cash flow to 95% to 100% of net income.²

Balance sheet

At 30 September 2026 IDEXX had $197.5M of cash, $849M of gross debt, mostly a $474M drawing on its $1.25B credit facility plus senior notes, and so net debt of $651M C (calculated).²⁴ That is about 0.4 times EBITDA, and covenants allow up to 3.5 times.⁴ Net interest is about $32M a year.² Equity was $1.87B. The company has used the credit line to help fund buybacks, but the debt is small beside $1.5B of operating income.

The share price is the other record. From $106.21 on 4 November 2016 to $698.55 is a rise of 6.6 times, or 21% a year, against EPS growth of 20% a year from 2016 to the 2026 guide C (calculated).¹¹²² So the price has tracked earnings, and the multiple has barely moved. At the end of 2016 the shares were 48 times that year’s EPS, and today they are 46 times the 2026 guide C (calculated). The stock has had four falls of more than 30% from a high, and the last, 34% from November 2025 to September 2026, has just been reversed.¹

VIValuation

First, how this goes badly wrong

Before the numbers, I asked what would most likely hurt this investment, and worked back. Three things could, and none needs a surprise.

• Price stops outrunning visits. US clinic visits are down about 1.8% and global net price is up about 4%.⁶ Revenue per visit has to rise for the model to hold, and vets pass that on to owners, who are already paying more for each visit. If a pet owner starts declining a blood panel because the bill is high, growth drops to the mid-single digits in the US first. Every number on this page rests on that.
• The link between machine and consumable weakens. A rival or a chain gives clinics a cheaper reader, or Mars moves more of its own clinics’ testing into its own labs, or the single supplier of the Catalyst slides has a bad year.⁵ The company’s margin and its ROIC both depend on the consumable following the machine.
• The price itself. At 46 times earnings the shares assume the first two don’t happen. A move to the 27 times that the stock sold for at its March 2020 low gives $411, and 33 times, the September 2026 low, gives $505 C (calculated).¹ That is a fall of 41% or 28% on a business that does nothing wrong. This one is not about the business. It is about what I pay for it.

The bear case below is built from the first two. The third is why I show a margin of safety. I keep one more outside the model because I can’t size it: AI tools that change how a vet decides which test to order.

What the price implies

At $698.55 the market values IDEXX at $55.2B on 79.01M diluted shares, the 78.79M outstanding on 28 October plus 0.22M of dilution at the third-quarter rate C (calculated).¹⁴ That is 46.4 times the $15.06 midpoint of 2026 guidance, 41.7 times the Nasdaq consensus of $16.76 for 2027, and 36.7 times $19.03 for 2028.²¹⁷ The owner yield is 2.0% E (estimate), against the 4.90% ten-year Treasury.¹⁸

Owner cash is where this page differs from the usual sums. Reported 2026 earnings are about $1.19B at the guide. Take free cash flow at 97.5% of that, the middle of the company’s range, and subtract about $65M of stock pay that the free-cash-flow figure adds back, and the owner is left with about $1.10B E (estimate).²⁴ That is 92% of net income. The capital needed to stand still is smaller than I expected: capital spending of about $160M a year, close to depreciation and amortization, which ran at $119M in nine months, plus the growth in customer contract assets, which was $36M in nine months.²⁴ The gap to reported earnings is stock pay, mostly, and it widened in 2025 when tax timing helped.

What does the price pay for? Holding margin at today’s 32.2%, it needs revenue to grow 20.0% a year for ten years, to $29.4B in 2036, or 6.2 times 2026 C (calculated). IDEXX grew 10.3% a year over the last nine years. Even with a margin of 36%, the price needs 18.2% growth. And on my base growth, no margin gets there: at 60%, double today’s, the base value is $468.

What I can’t judge

Three inputs decide this valuation and I can’t know any of them. The first is how far price can keep running ahead of visits. IDEXX has done it for years, and management says it is not trying to maximise price.⁶ Nobody I can find has measured where clinics start to ration tests. The second is the ceiling on margin. Management expects 50 to 100 basis points of expansion a year, but a 33% operating margin is already high for a company that makes and services hardware.⁶ The third is what the next platform adds. MultiCue Dx will be shown in January.²

So here is the range, not one number. Read my $306 with the grid beside it.

The discount rate

I use 10.0%. The ten-year Treasury yields 4.90%, and Damodaran’s implied equity risk premium was 4.20% on 1 October.¹⁸¹⁹ My ten-year beta on month-end returns is 1.24, which gives 10.1%. The five-year beta of 1.51 gives 11.3%, and the three-year beta of 1.01 gives 9.2% C (calculated).¹²⁰ The range is wide because the stock moved with the market in some years and not in others. I take the ten-year figure, rounded. It matches the 10.0% on my Intuitive Surgical page and sits above the 9.0% on my Mastercard page, which I treat as a steadier business.

The rate that makes my base cash flows equal the price is 6.2%. That is barely above the Treasury yield, with the risks of a single company on top. To be fair to the market, the answer is sensitive to that rate. At 8.0% my bull is $610 and at 7.0% my base is $560 and my bull $782, above the price E (estimate). A buyer who accepts a return near 7% on a steady consumable stream is not being irrational. The bet is on a low required return as much as on growth.

Three cases

The model is simple on purpose. I set revenue growth and operating margin for each year from 2027 to 2036. I tax operating income at 20.5%, the rate the company guides for 2026.² Then I take off capital spending above depreciation at 0.1% of revenue, which is what 2026 looks like, and 15% of added revenue for working capital and the customer contract assets that fund new placements. Stock pay stays deducted because it is already an expense in operating income. What is left is owner cash before interest. The first cash flow, for 2027, arrives about 14 months from the price date.

My base 2027 operating income of $1.69B implies about $17 of EPS on my rough sum for shares and interest, close to the Street’s $16.76, and my 2028 of $1.86B implies about $19, in line with $19.03 E (estimate).¹⁷ So the near years are not where I differ from the market. The disagreement is about everything after that.

Where does $6.4B of 2030 revenue come from? About 8% a year, with price at 3% to 4% and the rest from testing intensity and a base of installed instruments that adds 9% to 10% a year E (estimate). That is a bit below what the company has done and a bit above what visits alone suggest. Inside it, inVue Dx at 20,000 machines is worth $70M to $110M, so it supplies about 1% of the sum.

The price sits 128% above my base, and that is the plain result. It is also 68% above my bull case. Why might the market be right? Because this stock has always sold above what my method gives. At its three earlier lows it traded at 41 times (2018), 27 times (2020) and 40 times (2022) that year’s EPS C (calculated).¹¹² The market has paid a premium for this company throughout the ten years in my data, for the same reasons: recurring revenue, high returns on capital, a habit of beating forecasts. The 22% jump on 2 November says the third quarter put the visits worry to rest for now. If that is right, my growth fade is too steep, and the grid shows what that is worth. Even so, the best cell is $425.

Other uses of the same dollar

The ten-year Treasury pays 4.90%. Against that, my base here returns 6.2% a year and my bear 5.1%, with the bull at 7.4% E (estimate).¹⁸ So the base earns about a point and a third over a government bond for taking the risks of one company. The closest comparison on this site is Intuitive Surgical, another installed base with consumables on top. There I have a base return of 7.0% a year, against a 10.0% hurdle, at a price 96% above my base value. Mastercard shows 8.4% against 9.0% at 13% above base. IDEXX is priced like Intuitive, and offers less per year on my numbers. Those pages use their own prices, rates and methods, so I treat gaps of a point as noise. The gap to Mastercard, over two points, is not.

What past lows looked like

The stock has fallen more than 30% from a high four times in ten years: 31% in 2018, 37% from February to March 2020, 54% from August 2021 to October 2022, and 34% from November 2025 to September 2026 C (calculated).¹ The 2022 low was $324.64 on 14 October, 40.4 times that year’s $8.03. The September 2026 low was $504.70, 33.5 times the 2026 guide. The shares closed under half of today’s price on 24 October 2022, four years ago.¹ That shows how fast the rerating has been, not that it will reverse.

The margin of safety

• Price$698.55Close, Q4 2026 [1]
• Base value$306My estimate E (estimate)
• Margin of safety below$214set from base and bear C (calculated)

I set the margin of safety at the lower of two prices: the one at which my $207 bear value is three quarters of what I pay ($276), and 70% of my base value ($214). The second is lower, so it is $214. At $214, my bear value is 97% of the price and the base is 43% above it. At $214 the bear returns 9.8% a year and the base 12.7% E (estimate). The stock last closed below $214 on 24 March 2020.¹ At $698.55 the price is 226% above that level, so there is no margin of safety on my numbers. It does not cover the stress case either: growth of 3% and margin falling to 25%, which gives $173.

Two different questions

How good is the business, and how good is the price? They are separate questions. As a business IDEXX scores well on most tests I have: recurring revenue, an installed base that compounds, a 50% reported ROIC, a gross margin that rose seven points in nine years and a share count that fell 11%. Its flaws are the falling visits, a dependence on one supplier for its biggest line and customers that a rival increasingly owns. As for the price, it is far above my numbers: it is 128% above my base, 68% above my bull, and earns 6.2% on my base. A strong business can be a weak purchase at the wrong price. This looks like that. I could be wrong on the growth fade, and the market is telling me so.

VIIThe case for the price

What most people miss

Three things cut in IDEXX’s favour and get little attention. First, the exclusive slides. Ortho, which makes the chemistry slides for Catalyst, is barred from selling them in the veterinary market outside the EU to anyone but IDEXX, under agreements that run to December 2044.⁵ Second, the base is growing faster than visits are shrinking. VetLab consumables grew 14.6% organically in the third quarter on a premium base that grew 11%, while US visits fell about 1.8%.²⁴⁶ Third, the share count. IDEXX retired about 9% of its shares from 2021 to 2025 and pays no dividend, and its 2026 outlook has average shares down about 2% again.⁷²

The bull case

In plain words: the blade is growing at double digits, and the new products add on top. I found no named investor who has published a thesis on it, so I show who owns it and what management says. At 30 June, Fundsmith held 1,019,887 shares, Polen Capital 532,384 and Baillie Gifford 353,797, per their 13Fs.²¹²²²³ Those positions are four months old and don’t say why. Berkshire Hathaway, Pershing Square, Coatue, Tiger Global and ARK held none at their latest 13Fs, on my check. Vanguard (12.2%, and a new 13G on 30 October for 5.04% by one of its units) and BlackRock (9.6%) tell me about index membership, not conviction.⁷²⁴

The clearest bull voice is the company itself. On 2 November the new CEO said innovations such as inVue Dx with FNA and a multi-cancer Cancer Dx panel help vets detect disease sooner and work more efficiently, and in August he called the model “durable, compounding growth.”²¹⁰ Management also told analysts it expects margin to expand 50 to 100 basis points a year.⁶ I found no named sell-side view I could verify, so I leave that out.

Against reality

If IDEXX does what its bulls say, my base is a fair picture of it. Consensus has EPS of $16.76 in 2027 and $19.03 in 2028, which is 11% and 14% growth on the 2026 guide.¹⁷ My base has operating income growing 10% a year to 2030, and with 2% fewer shares each year that reaches about 12% a share E (estimate). If IDEXX grows revenue 8% or more through 2028, gets margin to 34% and keeps buying back stock, $306 is possible. So is my bull of $416.

The price asks for much more. For $698.55 to earn 10% a year, the shares must reach about $1,023 by 2030. At 25 times earnings that is EPS near $41, against about $24 in my base for that year, or 72% more C (calculated). Equivalently, revenue must grow 20% a year for ten years at today’s margin.

Multiply it out. The price needs four things. Growth of 10% or more for ten years: 25%. A margin of 40% or more: 50%. No disruption from a chain, a supplier or a visit slump: 75%. A multiple near 37 times at the end: 40%. Those are my guesses, and they give 4% E (estimate). They overlap, so perhaps 5%. My base needs growth fading from 8.5% to 4%: 55%; margin to 36.7%: 60%; no disruption: 75%, or 25% E (estimate). The price asks for far more than my base, and my base is not likely either.

What a 10x would take. Ten times $698.55 is $6,986, about $552B of market value on today’s shares. At 30 times earnings, which is generous, that is about $18.4B of net income, 15 times this year’s $1.2B. At a 25% net margin, which is where IDEXX is now, that is about $74B of revenue, 15 times this year’s $4.75B, or 32% a year for ten years. Buying back 2% of the shares a year would cut those sums by about a fifth, to roughly $60B of revenue. IDEXX itself rose 6.6 times in ten years from $106.21, on EPS growth of 20% a year, and it started from about $9.7B on that year’s diluted share count, not $55B C (calculated).¹¹² I know of very few companies that have made a tenfold gain from a $55B base in ten years, and I can’t name one in diagnostics. For IDEXX, a tenfold rise from here is close to impossible, and I would not build a case on it.

Where I come down

Price moved me: a 4% annual gain with falling visits is stronger than I expected, and it is why my base starts at 8.5% growth. The 22% jump in the shares did not, because it reflects one quarter. Visits stabilising near flat, with price still at 3% to 4% and the installed base growing 10% or more, would raise my values most. A big acquisition at a rich price, or recurring growth under 8%, would lower them.

VIIIRisks, and what would change my view

The three ways this goes badly wrong are at the start of Valuation. These are the rest.

• Leadership change. A new CEO since May, with the former CEO as chair until May 2027 and three officers filing sale notices within days of a 27% rise.^7141516 A new CEO’s first year is when choices such as the size of acquisitions get made.
• Pet population and spending. Puppy and kitten adoption is muted and owners are paying more per visit.⁶ A downturn in discretionary pet spending would hit visits first and testing second.
• Currency. 36% of 2025 revenue came from outside the US, and the company says a 1% stronger dollar would cut the rest of 2026 revenue by about $4M and EPS by about $0.02.⁵⁶ The company hedges part of it.
• Tariffs and supply. The company lists changes in tariff and trade policy among the risks to its outlook.² It buys sole-source components, with Ortho the one it names.⁵
• Competition from chains and rivals. Mars owns hospitals, labs and point-of-care brands, Zoetis owns Abaxis, and Mindray and Fujifilm compete abroad.⁵
• Acquisitions. CoVet cost about $250M for roughly $17M of annual revenue on the release’s guide C (calculated). A string of such deals, or one large one, would change the capital story.⁴²
• Tax. The 2025 cash flow was helped by $131M of deferred tax from a change in US rules on research deductions. The effective rate was 20.3% for nine months and is guided to about 20.5%.⁵⁴² A change in international taxes or in the foreign-derived income deduction would move EPS.
• Regulation and AI. The 10-K flags the EU AI Act for products using AI, and the company is putting AI into imaging, software and CoVet.⁵ I can’t tell how fast AI changes how a vet decides which test to order. It could expand testing or shrink it.
• Share price. The shares have fallen 34% and then risen 38% from the low in under eight weeks.¹ A holder at $698.55 is exposed to a return of the multiple to 27 to 41 times earnings, whatever the business does.