Healthcare has no shortage of data.
It has a shortage of usable data.
Clinical records, molecular tests, pathology images, treatment histories, and outcomes often live in separate systems. Tempus is trying to connect them, use the resulting data to guide care, and sell insights back to life-science companies.
That creates an appealing flywheel.
More tests create more data. More data improves the models. Better models can make the platform more useful to physicians and drug developers.
At $84.57, the market is paying for that flywheel before mature profits have arrived.
What Tempus actually controls
Tempus controls a growing multimodal clinical dataset, a diagnostics operation that adds new information, and software and licensing products that make the information useful.
The combination matters.
A diagnostic lab alone can become a volume business.
A data vendor alone can struggle to keep its information current and differentiated.
Tempus uses the lab to refresh the data asset and the data asset to improve the value of each test and license.
In second-quarter 2026 results, revenue rose 22% to $382.5 million. Oncology volume grew 31%, Data Licensing and Modeling revenue grew 36%, and the company signed roughly $200 million of new Data and Applications licenses.
Adjusted EBITDA turned positive at $8 million.
Why does that matter?
The flywheel is producing growth across both the data source and the higher-margin monetization layer.
It is only beginning to produce operating profit.
The data advantage can deepen - or get expensive
Tempus says it is connected to roughly 65% of U.S. academic medical centers and 55% of U.S. oncologists, with more than 45 million research records.
Scale helps because rare diseases, uncommon biomarkers, and treatment-response patterns require large datasets.
But healthcare data is not a frictionless internet dataset.
Privacy rules, consent, reimbursement, assay validation, data rights, and clinical adoption all determine what can be collected and how it can be used.
Tempus is also buying capabilities. The proposed Personalis acquisition carries an enterprise value of roughly $1.5 billion and follows other transactions including Ambry Genetics and Paige AI.
Acquisition can accelerate the flywheel.
It can also hide whether the core engine is improving.
What $84.57 requires
Current earnings do not support a conventional earnings multiple, so I use 2031 revenue per diluted share and a 11.5% required annual return.
At a 7 times terminal sales multiple, the checked price requires about $20.80 of 2031 revenue per diluted share.
With a future diluted count around 230 million to 250 million, that implies approximately $5 billion of 2031 revenue.
Against 2026 guidance of $1.595 billion to $1.605 billion, the company needs roughly 25% annual revenue growth while expanding margins and controlling dilution.
That is possible.
The expectations are demanding.
Three possible outcomes
Conservative: about $24 today
Tempus reaches $3 billion of 2031 revenue, ends with 260 million diluted shares, and receives a 3.5 times sales multiple.
Diagnostics grow, but reimbursement and acquisition integration keep margins modest. The dataset remains useful without producing software-like economics.
Base case: about $68 today
Tempus reaches $4.5 billion of revenue, ends with 245 million diluted shares, and receives a 6.5 times sales multiple.
This requires low-to-mid-20% growth, sustained positive adjusted EBITDA, and successful integration of acquired platforms.
The base case sits below the checked price.
Upside case: about $128 today
Tempus reaches $6.5 billion of revenue, limits dilution to about 235 million shares, and receives an 8 times sales multiple.
This assumes the data platform becomes a standard across oncology diagnostics, clinical decision support, and biopharma discovery.
That is the outcome today's premium is leaning toward.
Debt, dilution, and acquisition risk
Tempus ended June with $820.7 million of cash and marketable securities after completing a $460 million zero-coupon convertible-note offering due 2032.
Zero interest sounds free.
It is not free if conversion enlarges the share count after the stock rises.
Second-quarter net income of $5.6 million also included $98.5 million of unrealized gains on marketable securities, while stock compensation and related payroll taxes were $55.6 million.
Why does that matter?
The headline profit did not yet represent mature operating economics.
Investors should watch adjusted EBITDA, but they should also put stock compensation, acquisition consideration, and future conversion back into the ownership math.
What to watch out for
- Oncology test volume, reimbursement, and gross profit - because diagnostics feed the data asset and fund the platform.
- Data Licensing and Modeling growth - because higher-value reuse is central to the margin thesis.
- New license value converting into recognized revenue - because bookings are not cash.
- Personalis, Ambry, and Paige integration - because acquisitions should deepen the platform without obscuring organic growth.
- Stock compensation, convertible debt, and diluted shares - because per-share value can diverge from company growth.
- Adjusted EBITDA and operating cash flow - because data scale must eventually reduce the need for outside capital.
What could change the thesis
The thesis would strengthen if organic growth remains above 20%, Data and Applications outgrows diagnostics, adjusted EBITDA scales rapidly, and acquisitions increase revenue per diluted share.
The thesis would weaken if growth becomes acquisition-dependent, reimbursement weakens, stock compensation stays elevated, or the company repeatedly raises capital before cash generation matures.
The disconfirming evidence would be more data, more revenue, and no path to more owner earnings.
Bottom line
Tempus is building one of the more compelling data flywheels in healthcare.
Strategic asset: valuable.
Growth: strong.
Profitability: early.
Financing and integration risk: material.
Valuation: above my base case.
At $84.57, TEM requires the dataset to become a durable economic network, not merely an impressive collection of clinical information.
The practical principle: data becomes a moat only when every new record lowers the cost or raises the value of the next decision.
Sources
- Tempus second-quarter 2026 results
- Tempus financial information
- Tempus investor relations overview
- Tempus SEC filings
- Federal Reserve H.15 interest rates
- Damodaran U.S. industry betas
- Damodaran implied equity risk premium
Scenario values are analytical estimates, not company guidance or price targets. Educational only - not financial advice.