IDEXX: The AI Sorting Machine Left a 46%-ROIC Compounder for Dead

IDEXX: The AI Sorting Machine Left a 46%-ROIC Compounder for Dead

The broad market rallied sharply in the second quarter of 2026, a surprisingly strong tape given the Middle East conflict, rising oil prices, and inflation worries that never fully cleared. In a market like that, money does not just rise — it rotates. And the rotation has been merciless to one corner that should be the GARP investor's natural habitat: the quality end of mid-cap growth. A marquee quality mid-cap growth strategy has lost roughly 4% over the past year even as the index it tracks gained, which tells you the quality tilt itself has been the drag rather than the sector.

The index itself betrays nothing. The iShares Russell Mid-Cap Growth ETF sits within half a percent of its 52-week high, up roughly 8% year to date and about 4% on a trailing twelve-month basis. So this is not a case of the whole group being thrown out. The damage is inside it, the product of a market-wide sorting of growth stocks into two buckets: names the AI trade blesses, and names it does not. Cloudflare, one of the largest holdings in a well-known quality mid-cap growth fund, is an AI-infrastructure darling up more than 55% this year and roughly 87% over the latest 120 trading days, near its high at a price-to-sales multiple around 44x. That is the blessed bucket. The cursed bucket, in the same portfolios, holds IDEXX Laboratories, a $43 billion veterinary diagnostics company that is down 19% year to date and sits about 29% below its 52-week high.

Here is the part worth slowing down on: IDEXX has done everything that is supposed to justify owning a compounder, and the market has rewarded it with a derating anyway.

The Numbers Never Broke

Second-quarter results, reported in early August, beat consensus by a comfortable margin. Revenue came in at $1.22 billion, up 10% as reported and 9% on an organic basis — "organic," meaning stripped of currency and acquisition effects, the cleanest read on underlying demand. Diluted EPS was $4.27 against the roughly $3.94 analysts had penciled in, up 18% as reported and 15% on a comparable basis that excludes currency and share-based-compensation tax benefits. Net income rose to $338 million from $294 million. And rather than blaming the environment, management raised full-year 2026 EPS guidance to $14.69-$14.94, $0.14 higher at the midpoint, while lifting its revenue outlook at the margin and guiding operating margin to 32.3%-32.5%.

This was not a one-quarter recovery act. Back in the first quarter, revenue grew 14% year over year with organic growth of 11%, EPS rose 17%, and management raised full-year guidance then as well. Two consecutive quarters of guidance going up while the stock goes down is not the signature of a broken business. It is the signature of a market that has stopped paying for incremental quality execution.

The Moat Is Compounding, Not Cracking

Before calling this contrarian, I need to check the moat, because a year of underperformance in a name with a premium multiple can just as easily be a real deterioration. The core of IDEXX is its IDEXX VetLab instrument-and-consumables model: place diagnostics instruments in veterinary clinics, then sell them the reagents, slides, and consumables for years. It is a razor-and-blade business, and the blade side is the metric that matters. Companion animal group diagnostics recurring revenue grew 11% as reported and 10% organic in the quarter, with the flagship VetLab consumables line up 15% as reported and 14% organic. The installed base of the company's newer inVue Dx handheld analyzer pushed past 9,000 instruments, with about 1,600 placements added in the quarter, and the overall premium instrument base expanded 11%. None of that reads like disruption or share loss; it reads like the compounding engine working exactly as designed.

The quality numbers support it. Return on invested capital sits near 46%, return on equity above 70%, gross margin around 62%, and free cash flow margin around 24% — meaning roughly a quarter of every revenue dollar converts to cash after capital spending. Trailing free cash flow is about $1.24 billion, up nearly 60% from a year ago, against only $134 million of capex; the balance sheet carries just $772 million of net debt for a company of this size. For an intelligent reader new to the name: ROIC measures how much profit a company earns on the capital it deploys, and 46% is exceptional — it means IDEXX can fund growth and repurchases out of its own earnings without diluting shareholders or levering up.

What ~38x Earnings Buys You

The honest part of this setup is that IDEXX is not dirt cheap, and I am not going to pretend otherwise. The stock trades at roughly 38x trailing earnings, about 37x the midpoint of the raised full-year guide, with a price-to-sales multiple near 9x. At the 52-week high, the multiple was far more demanding; because earnings kept growing into a 29% price decline, the multiple compression is steeper than the share price suggests. The stock has gone from egregiously expensive to expensively-priced-for-a-compounder.

That reframes the argument. This is not a value-trap bargain where the market multiple converges with the S&P 500; this is a quality premium that got cut by roughly a third while the earnings outlook was revised upward. The market is effectively pricing in a stall — a return to high-single-digit growth with nothing left over. The burden of proof sits with the bears to show that the recurring engine is actually breaking. The data points the other way: recurring consumables compounding at 14% organic, an expanding instrument base, and operating margins guided higher, not lower.

I should name the risk, because it is real. Growth has decelerated from 14% reported revenue growth in the first quarter to 10% in the second, rapid-assay products were roughly flat, and capital instrument revenue fell 19% largely as the inVue platform laps last year's big launch. If organic growth settles in the high single digits and stays there, a ~38x multiple becomes genuinely hard to justify, and the stock can keep grinding lower. That is the invalidation trigger, and I am watching it rather than hand-waving it away.

The Timing and the Trigger

The fundamental case is intact, which makes this a falling opportunity rather than a falling knife — but the price action has not yet confirmed it, and that matters for how I would hold the position. The stock sits at about $546, below both its 50-day and 200-day moving averages, with a 14-day RSI near 41, so the trend has not flipped. The 52-week low around $519 is only about 5% below the current price, meaning a test of the lows is a live possibility. There is no confirmed bear trap yet — no volume exhaustion or failed breakdown to mark the low-risk entry. As a cross-check on positioning, AInvest's aggregate signal still labels the stock a Buy, but consensus posture is confirmation at best; the valuation and business case carry the argument.

So I would not chase the stock from here, and I would not pretend the derating is finished just because the story reads better. The lower-risk path is to build on weakness and add as the price action confirms — a test of the lows on drying volume, or a reclaim of the 50-day line — while sizing for a medium-to-long-term horizon. The cruel irony of the current tape is that money is rotating out of exactly the kind of business the AI trade is supposed to make more valuable over time: recurring, durable, high-margin cash generation. When the market's sorting machine throws a 46%-ROIC compounder that keeps raising guidance into the trash bin, the mispricing is on the side of the name being discarded — and the patient investor is the one who picks it up on the way down, not the one who waits for the narrative to turn.