The Instrument of Separation
In the spring of 2023, Agilent Technologies reported something that would have been unremarkable for most companies but was, for Agilent, a kind of seismic event: revenue declined. After three consecutive years of double-digit growth — a run that had taken the company from $5.3 billion to $6.8 billion in annual sales and vaulted its market capitalization past $50 billion — the instruments stopped shipping at quite the same velocity. Pharmaceutical companies, which had gorged on COVID-era diagnostics spending, began digesting their capital budgets. China, which had accounted for roughly a fifth of Agilent's revenue and an outsized share of its growth, turned abruptly hostile to procurement from American life-sciences suppliers. The stock, which had quintupled from its 2015 levels, gave back nearly 30% in a matter of months. Wall Street, with its usual subtlety, began asking whether the Agilent story was "over."
It was not over. It was, in fact, a useful moment to understand what the Agilent story actually is — because the company's durability has never been a function of any single end-market cycle. Agilent makes the machines that separate, identify, and quantify the molecular composition of virtually everything: the purity of a drug compound, the pesticide residue on strawberries, the chemical fingerprint of a lithium-ion battery cathode, the contaminants in a municipal water supply. Its instruments — liquid chromatographs, mass spectrometers, gas chromatographs, spectrophotometers — are the sensory organs of the global quality-assurance infrastructure. They sit in laboratories at Pfizer and Novartis, at the FDA and the EPA, at Samsung SDI and CATL, at every major university chemistry department, at crime labs and food-safety authorities and environmental-testing firms in 110 countries. The installed base numbers in the hundreds of thousands. And every one of those instruments consumes a recurring stream of columns, reagents, sample-preparation kits, and software subscriptions that Agilent sells at gross margins north of 60%.
This is the quiet empire that Ned Barnholt, Bill Sullivan, and Mike McMullen constructed from the wreckage of Hewlett-Packard's instrument division — a business that HP's leadership, in the late 1990s, considered a distraction from the real action in PCs, printers, and enterprise servers. What HP discarded became, over two decades, one of the most consistently profitable analytical-instrument companies on Earth, a textbook study in the compounding power of installed-base economics, and a surprisingly instructive case for any operator who has inherited a legacy business and wondered whether it could become something more.
By the Numbers
Agilent Technologies at a Glance
$6.51BFY2024 revenue
~$42BMarket capitalization (mid-2025)
~27%Operating margin (non-GAAP, FY2024)
~18,000Employees worldwide
$1.4B+Annual free cash flow (FY2024)
110+Countries with Agilent customers
~35%Revenue from recurring consumables & services
3Business segments
What Hewlett-Packard Threw Away
The origin story requires a brief detour through Palo Alto in the 1930s, which is to say through the garage where
Bill Hewlett and Dave Packard built their first audio oscillator in 1939. The founding mythology of Silicon Valley begins with two Stanford electrical engineers and an order from
Walt Disney — eight oscillators for the sound system of
Fantasia — but what gets lost in the retelling is that HP's instrument business was not a precursor to the computing revolution; it
was the company, for decades. Hewlett-Packard made oscilloscopes, signal generators, frequency counters, and, eventually, gas chromatographs and mass spectrometers. The Test &
Measurement division and the Chemical Analysis Group were the original HP — the business that funded everything else.
By the mid-1990s, HP had become a sprawling $40 billion conglomerate, and the instrument divisions, though profitable, were growing in the low single digits while the computing businesses were doubling. Carly Fiorina, who became CEO in 1999, saw the instruments as ballast. The board agreed. On November 1, 1999, Hewlett-Packard spun off its test-and-measurement and chemical-analysis businesses into a new public company. They called it Agilent Technologies — a name conjured by a branding consultancy that was meant to evoke agility and intelligence and instead evoked nothing at all, which may have been the point. The new entity debuted with $8 billion in revenue, 43,000 employees, and the awkward energy of a middle-aged professional who has just been told their services are no longer required.
Edward "Ned" Barnholt, who had spent 30 years at HP and led the Test & Measurement organization, became Agilent's first CEO. He was a measured, operational leader — not a visionary, not a dealmaker, but a man who understood instrument businesses at the circuit-board level. His immediate problem was that Agilent was too many things: semiconductor test equipment, fiber-optic components for telecom networks, electronic test instruments, and life-sciences analytical instruments, all bundled together under a corporate umbrella that lacked strategic coherence. The telecom bubble was about to burst. Semiconductor test was violently cyclical. And the analytical-instruments business — the gas chromatographs, the HPLCs, the mass specs — was the quiet, boring, recurring-revenue gem buried inside a portfolio of volatile hardware businesses.
Barnholt began the triage. When the dot-com crash obliterated demand for telecom components, Agilent's revenue cratered from $8.3 billion in FY2001 to $6 billion in FY2002. The company lost money. Barnholt cut 8,000 jobs. He sold the semiconductor-test business. He divested the healthcare-solutions group. Piece by piece, the conglomerate was being carved down to its core, though it would take another CEO and another decade to finish the job.
1939Bill Hewlett and Dave Packard build their first instrument in a Palo Alto garage.
1966HP enters the analytical-instruments market with gas chromatography products.
1999HP spins off Agilent Technologies; IPO at $30/share raises $2.1 billion — the largest tech IPO in Silicon Valley history at the time.
2001Dot-com crash; Agilent revenue falls from $8.3B to $6B in one year.
2005Bill Sullivan succeeds Ned Barnholt as CEO.
2006Agilent sells its semiconductor-test business to Verigy for $895 million.
2014Agilent spins off Keysight Technologies (electronic test & measurement), completing the focus on life sciences and diagnostics.
2015Mike McMullen becomes CEO and launches the "Build and Buy" strategy.
The Sullivan Pivot
Bill Sullivan, who succeeded Barnholt in 2005, was the architect of the strategic clarity that would define Agilent's next two decades. A PhD chemist who had joined HP in 1977 and risen through the chemical-analysis business, Sullivan understood something that Wall Street kept missing: the life-sciences and chemical-analysis instruments business was fundamentally different from the electronic-test business, not in the physics of measurement but in the economics of the customer relationship.
An oscilloscope buyer purchases a box. A chromatography buyer purchases a system — the instrument, the column, the detector, the software, the method, the validation, the service contract, and a decades-long relationship with the vendor's application scientists who help develop and troubleshoot analytical methods. The instrument is the razor. The columns and reagents are the blades. And the methods — the specific protocols validated by regulators for testing drug purity or food safety or environmental compliance — create switching costs that make the customer relationship almost geological in its permanence. A pharmaceutical company that has validated an analytical method on an Agilent HPLC with an Agilent column and Agilent software, and has submitted that method to the FDA as part of a drug application, does not casually switch to a Waters instrument. The cost of revalidation alone can run into the hundreds of thousands of dollars. The regulatory risk of changing an approved method is existential.
Sullivan saw this. He saw that Agilent was carrying two fundamentally different business models under one corporate roof — a cyclical electronic-test business and a recurring-revenue life-sciences business — and that the market was giving Agilent the multiple of the worse business. In 2014, he executed the move that would define Agilent's modern identity: spinning off the electronic test-and-measurement division as Keysight Technologies, an independent public company.
The Keysight spin was clean and decisive. On November 1, 2014 — exactly 15 years after HP had spun off Agilent — Agilent spun off Keysight, completing a two-generation arc of corporate mitosis. What remained was a pure-play analytical-instruments and life-sciences company with approximately $4 billion in revenue, gross margins above 50%, and a clear strategic identity for the first time in its existence.
We are creating two highly focused companies, each with the ability to allocate capital, develop products, and pursue strategies specifically tailored to their respective markets.
McMullen's Machine
Mike McMullen took over as CEO in March 2015, and the contrast with his predecessors is instructive. Barnholt was an operator who managed a crisis. Sullivan was a strategist who clarified an identity. McMullen was a builder — a chemical engineer by training, a former president of Agilent's Chemical Analysis Group, a man who thought in terms of installed-base penetration, aftermarket attach rates, and the lifetime value of a laboratory customer. Where Sullivan had defined what Agilent would be, McMullen set out to define how it would compound.
His framework was disarmingly simple: "Build and Buy." Build the organic product pipeline — invest in next-generation instruments, expand the consumables portfolio, develop the software platforms that would tie instruments to data to workflows. And buy the businesses that would fill gaps in the portfolio, extend Agilent into adjacent workflows, and — crucially — increase the proportion of revenue that was recurring.
The M&A program that followed was disciplined, mid-market, and relentlessly focused on the aftermarket. McMullen did not chase transformative mega-mergers. He acquired businesses that deepened Agilent's presence inside the laboratory: Dako (clinical diagnostics, pathology), Cobalt Light Systems (Raman spectroscopy), Lasergen (sequencing-by-synthesis chemistry), ProZyme (glycan analysis), Resolution Bioscience (liquid biopsy), and BioTek (microplate readers and imaging) — among others. The deals ranged from tens of millions to roughly $1.1 billion, and they shared a common logic: each one either expanded Agilent's consumables revenue, added a software or services layer, or moved the company deeper into a regulated workflow where switching costs were structurally high.
The BioTek acquisition in 2023, which brought microplate readers into Agilent's cell-analysis portfolio (alongside ACEA Biosciences, acquired in 2018), illustrates the pattern perfectly. Microplate readers are not glamorous instruments. They are the workhorses of drug-discovery laboratories — measuring absorbance, fluorescence, and luminescence in 96-well and 384-well plates, tens of thousands of times a day. The instruments are relatively inexpensive. But the reagent consumption per plate, multiplied across thousands of plates per year, multiplied across thousands of laboratories, creates an annuity stream that compounds quietly in the background while analysts obsess over instrument shipment timing.
McMullen's most consequential strategic bet, though, may be the one that is hardest to see from the outside: the investment in Agilent CrossLab, the company's services and consumables business. CrossLab is not a product line; it is a business model — a platform that bundles instrument maintenance, compliance services, asset management, consumable columns and reagents, and workflow-optimization consulting into multi-year enterprise agreements. It is, in essence, Agilent's attempt to transform itself from a capital-equipment vendor into a laboratory-infrastructure-as-a-service provider. CrossLab grew to roughly $1.6 billion in revenue by FY2024, with margins that exceed the corporate average and retention rates that approach 95%.
The beauty of CrossLab is that it grows when our customers' labs grow, and it doesn't decline when their capital budgets tighten. It's the most resilient part of our business.
The Three Kingdoms
Modern Agilent is organized into three reporting segments, and the architecture reveals the strategic logic more clearly than any mission statement.
Life Sciences and Applied Markets Group (LSAG) is the flagship — the instruments business. Gas chromatographs, liquid chromatographs, mass spectrometers, spectroscopy systems, cell-analysis instruments, and the software that runs them. LSAG accounted for roughly $3.1 billion of FY2024 revenue, or about 48% of the total. These are the machines that sit on laboratory benches around the world, and their placement drives the entire downstream economics of the company. LSAG's gross margins are in the mid-50s, pressured by the capital-equipment nature of the business and the R&D intensity required to maintain technology leadership.
Agilent CrossLab Group (ACG) is the aftermarket engine. Consumables (columns, sample-preparation products, reagents), services (instrument maintenance, compliance, enterprise asset management), and the software platform that ties laboratory workflows together. ACG generated approximately $1.6 billion in FY2024 — about 24% of total revenue — with gross margins well above 60%. This is the segment that gives Agilent its durability. When pharmaceutical companies freeze capital budgets, they do not stop buying columns. When a government regulator mandates a new pesticide-screening method, the testing laboratories need new consumables, not necessarily new instruments.
Diagnostics and Genomics Group (DGG) is the youngest and most volatile segment, built largely through acquisition — Dako's pathology business, the nucleic-acid solutions business, and the genomics portfolio. DGG generated roughly $1.8 billion in FY2024, or about 28% of total revenue, with margins that have been compressed by the normalization of COVID-related demand (Agilent's nucleic-acid products were used in PCR-test manufacturing) and the structural challenges of the clinical-diagnostics market. DGG is the segment where Agilent's strategic ambitions are most exposed — it is a bet that the same installed-base economics that work in analytical chemistry can be extended into clinical pathology and genomics. The jury remains out.
🔬
Agilent's Three Segments
FY2024 revenue and operating profile
| Segment | FY2024 Revenue (est.) | % of Total | Key Products | Margin Profile |
|---|
| Life Sciences & Applied Markets (LSAG) | ~$3.1B | ~48% | LC, GC, MS, spectroscopy, cell analysis | Mid-50s gross |
| Agilent CrossLab (ACG) | ~$1.6B | ~24% | Columns, reagents, services, software | 60%+ gross |
| Diagnostics & Genomics (DGG) |
The Column Is the Strategy
To understand Agilent's competitive position, you have to understand what a chromatography column actually does and why its economics are so extraordinary.
A chromatographic column is a tube — typically stainless steel or fused silica, ranging from a few centimeters to several meters in length — packed with stationary-phase material (silica particles, polymer beads, or carbon) whose surface chemistry has been precisely engineered to interact differently with different molecules. You inject a mixture into one end. A mobile phase — a solvent or gas — pushes the mixture through the column. Different molecules interact differently with the stationary phase, so they emerge from the other end at different times. This is separation. The detector at the far end identifies and quantifies what comes out.
The column is a consumable. It degrades with use. Depending on the application, a column might last for hundreds of injections or thousands, but eventually the stationary-phase chemistry breaks down, peak shapes broaden, resolution degrades, and the analyst replaces it. A single HPLC column costs anywhere from $200 to $2,000. A high-throughput pharmaceutical QC laboratory might consume dozens per month. Multiplied across Agilent's installed base — hundreds of thousands of LC and GC instruments worldwide — the column business alone generates hundreds of millions of dollars in annual revenue at gross margins that likely exceed 70%.
But the economics go deeper than simple razor-and-blade arithmetic. Columns are method-specific. When a pharmaceutical company develops an analytical method for testing drug purity — say, an HPLC method for assaying the active ingredient in a tablet — that method specifies the exact column (brand, dimensions, particle size, stationary-phase chemistry), the exact mobile-phase composition, the exact flow rate, the exact temperature, and the exact detector settings. This method is then validated according to ICH guidelines, submitted to the FDA as part of the drug application, and locked in place for the commercial life of the drug. The column specified in the method becomes, in effect, a regulatory mandate. Switching to a competitor's column would require method revalidation — a process that can take months and cost six figures. For a blockbuster drug, the risk of a failed revalidation, and the potential delay to commercial supply, makes the switching cost functionally infinite.
This is why Agilent's CrossLab business is so resilient, and why the company's competitive moat is deeper than a simple market-share analysis would suggest. The moat is not in the instruments per se — Waters, Shimadzu, Thermo Fisher, and Bruker all make excellent chromatographs and mass spectrometers. The moat is in the method — the validated, regulated, locked-in combination of instrument, column, software, and protocol that the customer cannot easily change. Agilent's strategy, under McMullen, has been to maximize the number of methods validated on Agilent systems and to expand the aftermarket revenue that flows from each validated method.
The China Variable
Every analytical-instruments company has a China story, and Agilent's is both more instructive and more uncomfortable than most.
China represented approximately 20% of Agilent's revenue through the early 2020s — a proportion that had grown steadily as Chinese pharmaceutical companies expanded, as environmental-monitoring mandates proliferated under Xi Jinping's early environmental reforms, and as Chinese universities built world-class analytical-chemistry departments. Agilent was the dominant Western brand in Chinese laboratories. Its local sales force was large, well-trained, and deeply embedded in the academic and pharmaceutical customer base.
Then the ground shifted. Beginning in 2023, the Chinese government — motivated by a combination of economic nationalism, geopolitical tension with the United States, and a desire to build domestic alternatives to Western scientific infrastructure — began steering procurement decisions toward domestic instrument manufacturers. Shimadzu, with its manufacturing footprint in Japan and closer cultural ties, was less affected. But American brands, Agilent included, faced a systematic chill: purchase orders delayed, tenders restructured to favor local suppliers, and a growing ecosystem of Chinese instrument companies — Shimadzu's local competitors, essentially — that were improving rapidly and competing on price.
Agilent's China revenue declined meaningfully in FY2023 and FY2024. Management has been measured in its public commentary, acknowledging "softness" and "extended deal cycles" without naming the geopolitical dynamics explicitly. But the implications are structural: the assumption that China would be a high-single-digit growth engine for Agilent for the foreseeable future — an assumption baked into many analysts' long-term models — now carries a significant asterisk.
The China question is not whether Agilent will lose all its Chinese business. It won't. The installed base is too large, the methods too deeply validated, the applications too technically demanding for domestic competitors to displace overnight. The question is whether China reverts to a growth contributor or becomes a slowly eroding legacy market — and whether the rest of Agilent's geographic portfolio (the Americas, Europe, non-China Asia-Pacific) can compensate for the lost growth.
We continue to see a more cautious spending environment in China, particularly in the pharma and academic segments. We're adjusting our expectations accordingly while maintaining our investment in the region.
Pharma's Pendulum
Pharmaceutical and biotechnology companies account for roughly one-third of Agilent's revenue — the single largest end-market — and the relationship between Agilent's growth rate and pharma's capital-expenditure cycle is the most important variable in the company's near-term financial model.
The COVID era was extraordinarily good to Agilent's pharma business. Vaccine development, mRNA manufacturing, therapeutic-antibody production, and the vast expansion of QC testing capacity for COVID-related therapies drove an unprecedented investment cycle in analytical instrumentation. Pharma companies were not only buying instruments; they were accelerating laboratory buildouts, doubling QC capacity, and stockpiling consumables. Agilent's revenue growth hit 18% in FY2022. The CrossLab and consumables businesses were running at full capacity.
What followed was the hangover. By mid-2023, pharma capital budgets began normalizing. The large biopharma companies — Pfizer, Roche, Novartis, AstraZeneca — had overbuilt capacity during COVID and were now rationalizing. Biotech funding contracted as interest rates rose, squeezing the smaller companies that had been enthusiastic instrument buyers. Agilent's pharma-related revenue declined in the mid-single digits in FY2023 and continued to face headwinds into FY2024.
The structural case for pharma instrumentation demand remains intact — the drug-development pipeline is enormous, regulatory requirements for analytical testing are increasing, not decreasing, and the shift toward biologics and cell-and-gene therapies requires more sophisticated analytical workflows, not fewer. But the cyclical overlay is real. Pharma is Agilent's most important customer, and pharma's capital-spending cycles are long, lumpy, and driven by forces (patent cliffs, clinical-trial outcomes, FDA policy, biotech funding markets) that Agilent cannot control.
The strategic response has been to tilt the revenue mix toward consumables and services — the components of the pharma relationship that are less cyclical than instrument sales. If Agilent can grow CrossLab's share of pharma revenue from roughly 35% to 40% or higher, the company's overall revenue volatility declines meaningfully. This is the single most important operational priority McMullen has articulated, and it explains the M&A focus on consumables and software businesses.
The Software Layer
The least visible but potentially most consequential investment Agilent has made in the McMullen era is in software — specifically, in laboratory-informatics platforms that manage data, workflows, and compliance across the analytical laboratory.
Agilent's OpenLab platform is the company's attempt to build a software layer that spans the entire laboratory workflow: instrument control, data acquisition, data analysis, reporting, and integration with laboratory information management systems (LIMS). OpenLab competes with Thermo Fisher's Chromeleon, Waters' Empower, and a constellation of smaller LIMS vendors, and the competitive dynamics are intense. But the strategic logic is clear: if Agilent can become the operating system of the analytical laboratory — the software layer that ties instruments, data, and workflows together — the switching costs compound exponentially.
A laboratory running OpenLab across 50 instruments has made a systems-level commitment that is orders of magnitude harder to reverse than a single-instrument purchase decision. The data is in Agilent's format. The methods are validated in Agilent's software. The compliance records — audit trails, electronic signatures, 21 CFR Part 11 compliance — are all embedded in Agilent's platform. Ripping out OpenLab and replacing it with a competitor's software is a multi-year project with profound regulatory risk.
Agilent has also invested in cloud-based laboratory services — remote diagnostics, predictive maintenance, instrument-utilization analytics — that layer on top of the installed base and create new recurring-revenue streams. These are early-stage businesses, but they represent the logical extension of the installed-base strategy: once you own the instrument and the consumable and the software, you can begin selling insight.
The risk, as with all enterprise-software ambitions from hardware companies, is execution. Agilent is an instrument company. Software is a different discipline — different talent, different development cycles, different customer-success models. The history of industrial companies that have tried to "become software companies" is littered with expensive failures. McMullen has been careful to frame the software strategy as complementary to, not a replacement for, the core instruments business. Whether that discipline holds as the temptation to "platform-ize" grows remains to be seen.
The Oligopoly of Measurement
The global analytical-instruments market is one of the most concentrated oligopolies in industrial technology, and the barriers to entry are formidable — not because the physics is impossibly difficult (many university labs have built one-off mass spectrometers) but because the qualification and validation infrastructure that surrounds the instruments is effectively insurmountable for new entrants.
Four companies dominate: Agilent, Thermo Fisher Scientific, Waters Corporation, and Danaher (through its Sciex and Beckman Coulter businesses). Shimadzu and Bruker are significant players in specific segments. Together, these six companies control the vast majority of the global chromatography, mass spectrometry, and spectroscopy markets.
The oligopoly is sustained by several reinforcing dynamics. First, the regulatory qualification process: a laboratory that has qualified an Agilent instrument for GMP use has invested months and significant capital in the installation qualification (IQ), operational qualification (OQ), and performance qualification (PQ) protocols. Second, the method-validation lock-in described above. Third, the application-science expertise: Agilent employs hundreds of PhD-level application scientists who develop and publish analytical methods, work directly with customers on method development, and train laboratory personnel. This scientific field force is a competitive weapon that new entrants cannot replicate without decades of investment.
Fourth — and this is underappreciated — the service and support infrastructure. Analytical instruments are not consumer electronics. A mass spectrometer that goes down in a pharmaceutical QC lab can halt production of a drug. Response time matters. Agilent maintains a global service network with same-day or next-day response capabilities in most major markets, backed by a parts-and-consumables logistics chain that is itself a competitive moat.
The oligopoly is not static, however. Thermo Fisher, through its acquisitions of Life Technologies, Phenom-World, and PPD, has become an analytically diversified colossus with $44 billion in revenue — seven times Agilent's scale. Waters, under new CEO Udit Batra, has been aggressively investing in next-generation LC platforms and software. Danaher, the master of the operating-system playbook, continues to execute its Danaher Business System across its analytical portfolio. And the Chinese domestic competitors — Shimadzu Instruments (Suzhou), Fuli Instruments, Techcomp — are improving faster than the Western incumbents would like to admit.
In this industry, you don't just sell a product. You sell a validated workflow. And workflows, once established, are extraordinarily difficult to displace.
The Succession Question
In September 2024, Agilent announced that Padraig McDonnell would succeed Mike McMullen as CEO, effective March 2025. McDonnell, who had served as president of Agilent's largest segment (LSAG) and, before that, as the company's chief commercial officer, was an internal promotion — a signal of strategic continuity.
McMullen's tenure had been remarkable by any operational measure. From FY2015 to FY2024, he had grown Agilent's revenue from approximately $4.0 billion to $6.5 billion, expanded operating margins by roughly 500 basis points, executed more than a dozen acquisitions, and compounded the stock at approximately 20% annually. He had transformed a post-spinoff instrument company into a recurring-revenue-weighted laboratory-solutions provider. The "Build and Buy" framework had worked.
McDonnell inherits a company in a different position than the one McMullen inherited. The easy growth — the post-Keysight-spinoff multiple expansion, the COVID-era demand surge, the China boom — has been harvested. What remains is the harder, more operationally demanding work of compounding a $6.5 billion business at mid-to-high single-digit growth rates while maintaining margins and deploying capital efficiently. The new CEO's stated priorities — accelerating the shift to recurring revenue, deepening the software and digital platforms, expanding in biopharma and diagnostics — are extensions of McMullen's strategy, not departures from it.
The question is whether Agilent's organizational culture, which was built for the McMullen era's acquisition-driven growth, can adapt to a period that may require more organic innovation and fewer transformative deals. McMullen was a dealmaker and an integrator. McDonnell, by reputation, is more of a commercial operator — a closer of enterprise agreements, a builder of sales-force discipline. Different era, different tool.
The PFAS Tailwind and Other Secular Currents
Beneath the cyclical noise of pharma budgets and China procurement, several structural trends are quietly compounding demand for Agilent's instruments and consumables.
PFAS — per- and polyfluoroalkyl substances, the "forever chemicals" that have contaminated drinking water, agricultural soil, and human blood across the developed world — may be the most consequential regulatory tailwind Agilent has encountered since the Clean Air Act created the environmental-testing industry in the 1970s. The EPA's 2024 decision to set enforceable limits on six PFAS compounds in drinking water, at concentrations as low as 4 parts per trillion, requires analytical capabilities that only a handful of instruments can deliver: triple-quadrupole LC-MS/MS systems with the sensitivity to detect molecules at the parts-per-trillion level.
Agilent's 6470 and 6495 triple-quadrupole mass spectrometers are among the instruments qualified for EPA Method 533 and Method 537.1, the standard methods for PFAS analysis. Every municipal water utility in the United States — there are roughly 50,000 of them — will need access to PFAS testing capability, either through internal laboratories or commercial testing services. The commercial environmental-testing laboratories (Eurofins, SGS, Bureau Veritas) are already expanding their PFAS-testing capacity and purchasing instruments aggressively.
Beyond PFAS, the shift toward biologics and cell-and-gene therapies in pharmaceutical development is creating demand for new analytical workflows that did not exist a decade ago. Characterizing a monoclonal antibody requires a different analytical toolkit than characterizing a small-molecule drug — multi-attribute methods (MAMs), intact-mass analysis, glycan profiling, charge-variant analysis — and Agilent has invested heavily in building the instruments, columns, and software for these workflows.
The energy transition is another, less obvious driver. Battery manufacturers need to characterize cathode and anode materials for impurities at the parts-per-billion level. Hydrogen-fuel-cell developers need to analyze gas purity. Semiconductor manufacturers, racing to build domestic fabrication capacity, need ultra-trace-level contamination analysis. All of these applications run on chromatography and mass spectrometry. All of them need columns and reagents.
A Machine for Compounding
The simplest way to understand Agilent is as a compounding machine with three gears.
The first gear is instrument placement — putting a new chromatograph or mass spectrometer on a laboratory bench. This is the most cyclical part of the business, the most capital-intensive, and the lowest-margin. But each instrument placed creates a multi-decade revenue stream.
The second gear is the aftermarket — the columns, reagents, service contracts, and software subscriptions that flow from the installed base. This is higher-margin, more recurring, and less cyclical. It is the engine that converts volatile instrument sales into durable cash flow.
The third gear is the method and workflow lock-in — the regulatory and procedural infrastructure that anchors the customer to Agilent's ecosystem for years or decades. This is not a revenue line; it is a structural condition that protects the first two gears from competitive displacement.
McMullen's genius was recognizing that the second and third gears were underinvested and that the company's strategic priority should be maximizing the lifetime value of each installed instrument — not just through consumables and service, but through software, digital services, and enterprise-level relationships that transform transactional customers into annuity streams.
The result is a business that generates roughly $1.4 billion in annual free cash flow — a 22% free-cash-flow yield on revenue — and returns most of it to shareholders through buybacks and a growing dividend. The balance sheet carries modest leverage. The capital requirements are low relative to revenue. And the installed base, numbering in the hundreds of thousands of instruments, continues to consume columns and reagents regardless of whether a single new instrument ships this quarter.
In the spring of 2023, when Agilent's stock was falling and analysts were fretting about the pharma cycle and China, the CrossLab business grew. The columns kept shipping. The service contracts renewed. The methods stayed validated. The machine kept compounding.
In the basement of a pharmaceutical QC laboratory in Basel, or a food-safety testing facility in Shenzhen, or an environmental lab in New Jersey, an Agilent HPLC hums through its thousandth injection of the day, pushing solvent through a $400 column packed with 1.8-micron silica particles, separating molecules that no human eye will ever see, generating data that will be reviewed, approved, and archived according to methods that were validated years ago and will not change for years to come. The column will be replaced next week. Agilent will ship another one.
