Okara
Start free
Back to Blog

By · Published November 8, 2026 · 22 min read

OpenEvidence Valuation and Revenue: How It Grew to $15B

OpenEvidence hit a $15B valuation in September 2026 on a reported $300M run rate. See every funding round, revenue by year and the 9 moves behind it.

OpenEvidence is now worth $15 billion.

In September 2026, the AI medical search company raised $250 million in a round led by Andreessen Horowitz and Byers Capital. Business Insider first reported the $15 billion valuation on September 24. Founder and CEO Daniel Nadler later confirmed the raise and its lead investors in an email to Becker's Hospital Review. The company did not put out a press release.

OpenEvidence does not publish detailed revenue figures. In July 2026, PYMNTS, citing The Information, reported that it was generating about $300 million in annualized revenue, roughly $25 million a month, and running at breakeven on cash flow. That was double the roughly $150 million run rate it ended 2025 with, according to a Bank of America note covered by Investing.com.

The usage numbers are just as unusual. As of September 2026, 1.12 million license-verified U.S. clinicians were using OpenEvidence, Fierce Healthcare reported. Nadler told Reuters that U.S. clinicians consulted it 42 million times in August 2026 alone, in a story republished by The Hindu.

The short reason it grew: OpenEvidence gave doctors a medical answer engine for free, let pharmaceutical advertisers pay for it, and licensed the journals doctors already trusted instead of scraping them.

The loop looks like this:

Free for verified doctors → answers built on licensed journals → doctors recommend it to other doctors → more usage → more ad revenue and more content partners → hospitals sign on.

Here is how it happened.

OpenEvidence's growth in one table

DateMilestone
2021Daniel Nadler self-funds the company after selling his first startup, Kensho
2022OpenEvidence is founded by Nadler and Zachary Ziegler
2023Raises a friends-and-family round
July 2024Supports about 358,000 verified physician consultations in the month
February 2025Raises a $75M Series A led by Sequoia at a $1B valuation; signs NEJM content deal; used by more than a quarter of U.S. doctors
June 2025Signs a multi-year content agreement with the 13-journal JAMA Network
July 2025Raises a $210M Series B at $3.5B co-led by GV and Kleiner Perkins; 8.5M monthly consultations; launches DeepConsult
August 2025Launches Visits, a free AI scribe, and claims a perfect USMLE score
October 2025Raises a $200M Series C led by GV at $6B; about 15M consultations in September; announces Veeva partnership
December 2025About 18M consultations in the month; exits the year at a reported ~$150M revenue run rate
January 2026Raises a $250M Series D at $12B co-led by Thrive Capital and DST Global
February–May 2026Sutter Health, Mount Sinai and Cedars-Sinai roll it out across their systems
March 10, 2026Handles 1M clinical consultations in a single day
April 2026Used by about 65% of U.S. doctors; OpenAI launches a free rival, ChatGPT for Clinicians
July 2026Reported at ~$300M annualized revenue; reportedly weighs, then sets aside, offers at a ~$20B valuation
August 202642M consultations by U.S. clinicians in the month
September 2026Raises $250M at $15B led by a16z and Byers Capital; signs MSK and Anthropic partnerships
October 2026Settles its lawsuit with rival Doximity

1. Nadler treated medicine as an information-overload problem

OpenEvidence is Daniel Nadler's second AI company.

His first, Kensho, built AI tools for Wall Street and was incubated by Goldman Sachs, Nadler said in a Talks at GS interview recorded in July 2026. S&P Global bought Kensho in 2018 for what CNBC put at $700 million, the figure OpenEvidence also uses. Forbes has reported the price as $550 million.

The idea for OpenEvidence came during COVID. Nadler told Goldman that he read how the COVID literature had gone from zero papers to around 30,000 in a matter of months. To him, it looked like the same problem Kensho had tackled in finance: too much information for any one expert to track.

He compared doctors to analysts with an impossible coverage list: an oncologist is expected to know the side effects, interactions and sequencing of around 100 cancer drugs. He described the product he wanted as a "Bloomberg Terminal for physicians."

He self-funded the company in 2021 and raised a friends-and-family round in 2023, CNBC reported. His co-founder, Zachary Ziegler, was a Harvard PhD student in AI and became the company's chief technology officer.

The pitch to doctors stayed simple. In its January 2026 funding announcement, Nadler said that keeping up with only the new evidence in the top 10 medical journals and recent guideline changes would take a doctor nine hours a day.

2. OpenEvidence licensed the journals instead of scraping them

This is the move that most separates OpenEvidence from general-purpose chatbots.

Around 2022, publishers were suing AI companies for copyright infringement. Nadler told Goldman he was reading Walter Isaacson's biography of Steve Jobs at the time, specifically the chapter on how Apple convinced record labels to sell music through iTunes instead of losing it to piracy.

He decided to do the same with medical publishers. He offered nonprofit journals and medical societies a deal to become their official AI partner, with OpenEvidence's success tied to theirs. Some, he said in the Goldman interview, became shareholders.

The deals came quickly once the company had funding:

  • New England Journal of Medicine: announced alongside the Series A in February 2025, CNBC reported.
  • JAMA Network: a multi-year agreement in June 2025 covering JAMA, JAMA Network Open and 11 specialty journals, according to the JAMA Network's announcement.
  • NCCN and the American College of Cardiology: both named as official AI partnerships in the Series D release.
  • Cochrane Library: listed as an official partner in the company's March 2026 announcement.
  • American Diabetes Association: named by NBC News among the specialty groups supplying treatment guidelines.

The licensing gives OpenEvidence the full text, not just abstracts. Nadler argues that many clinical questions are answered in a paper's methods section or its survival charts, which general models either can't see or have to guess at.

It also gave doctors a reason to trust the answers. As Nadler told CNBC in February 2025, "Trust matters in medicine, and the fact that it's trained on The New England Journal of Medicine... the result is a black-and-white difference in terms of accuracy."

There is a limit to this moat. NEJM editor-in-chief Eric Rubin told NBC News that the journal has similar licensing agreements with other tools.

3. It was free for verified doctors and closed to everyone else

OpenEvidence never charged doctors.

Clinicians sign up with their National Provider Identifier, the ID number the U.S. government issues to healthcare providers, and then ask unlimited questions for free. "Our commitment is that core OpenEvidence will always be free for users," Nadler told NBC News in May 2026.

Free removed the biggest obstacle in health tech: hospital procurement. Selling software to health systems usually means long pilots, security reviews and budget cycles. STAT noted that OpenEvidence reached its valuation by going directly to clinicians and avoiding that process.

It also opened the door to doctors outside big academic centers. "Most health care in America isn't happening at billion-dollar hospitals in New York or San Francisco," Nadler told CNBC. "It's happening in small practices that don't have IT departments or budgets for expensive software."

Verification mattered too. Because every user is a credentialed clinician, OpenEvidence could offer advertisers a rare audience.

Sequoia partner Pat Grady, who led the Series A, summed up the approach to CNBC: "This is a consumer internet company masquerading as a health-care business."

4. Doctors sold it to other doctors

OpenEvidence's main growth channel was word of mouth inside hospitals.

Nadler explained the mechanism to CNBC in February 2025: "When one doctor pulls out their iPhone and looks at something, other doctors can see that. Their natural question is, 'What's that?'"

By January 2026, Nadler said 95% of new users heard about OpenEvidence from another physician, CNBC reported.

Monthly sign-ups kept climbing:

Share of U.S. doctors rose with it. The company said it was used by more than a quarter of U.S. doctors in February 2025, CNBC reported. By July 2025, it said more than 40% of U.S. physicians logged in daily, in its Series B announcement. By April 2026, about 65% of U.S. doctors, roughly 650,000 people, actively used it, the company told NBC News.

5. Pharma advertisers paid the bill

If doctors don't pay, someone else has to.

OpenEvidence was one of the first AI startups to rely on advertising for revenue, CNBC reported, with companies paying for promotions such as video ads in its app. Nadler said the model allows faster adoption and wider use than a subscription.

The advertisers are mostly pharmaceutical and medical device companies. The company's advertiser page pitches "query-relevant, full-screen interstitial engagements" shown when a clinician asks about the advertiser's disease area.

That is a valuable placement. A doctor searching for treatment options for a specific condition is close to a prescribing decision. Sacra, a research firm, estimates that OpenEvidence's ad rates have reached more than $1,000 per thousand impressions while it sells only about 5% of its inventory, though the company has not confirmed these figures.

The model has a precedent in medicine. "Advertising has long been part of how clinicians access medical research through journals, and this represents a similar model in a digital format," Cedars-Sinai's chief health informatics officer told Fierce Healthcare.

Doctors don't seem bothered so far. Many clinicians interviewed by NBC News said the ads were unobtrusive or nonexistent.

The ad business ramped quickly once the company started selling. OpenEvidence said in its January 2026 announcement that it reached $100 million in annual revenue less than a year after building out its commercial team.

6. It kept adding free tools that pulled doctors back in

OpenEvidence started as a search box. It became a set of tools for most of a doctor's day, all still free to verified U.S. clinicians.

  • DeepConsult (July 2025): an AI research agent that compiles a report from hundreds of studies while the doctor sees other patients. The company said each run costs more than 100 times a standard search, in its Series B announcement.
  • Visits (August 2025): an AI scribe that transcribes patient encounters, drafts notes and adds relevant evidence, according to the company's launch post.
  • Medical coding (March 2026): automatic billing code suggestions, Fierce Healthcare reported.
  • Doctor dialer: an AI-integrated calling feature that takes on rival Doximity's core business, per the same Fierce report.
  • Voice Mode (May 2026): hands-free spoken answers for doctors on rounds or in the emergency department.

In August 2025, the company also announced that its AI scored 100% on an official USMLE sample exam, the U.S. medical licensing test. Its own methodology notes say it excluded questions with images and corrected two items in the dataset. Nadler told Fierce Healthcare that a USMLE score is not the best way to evaluate the product in clinical use.

Each new tool gave doctors another reason to open the app, and gave the company more moments to show ads.

7. Shadow usage turned into hospital deals

Many doctors used OpenEvidence on their personal phones, some without their employer's approval, NBC News found.

"There's this whole growing area of shadow AI," Mount Sinai's head of AI, Dr. Girish Nadkarni, told NBC News. "I think it's time to bring tools like OpenEvidence to the surface."

That pressure from doctors turned into enterprise deals in 2026:

  • Sutter Health and Mount Sinai announced plans to integrate it into their electronic health records in early 2026, with Mount Sinai's announcement coming in March, according to NBC News.
  • Cedars-Sinai rolled it out across its system in May 2026, letting clinicians pull in a patient's record when they ask a question, Fierce Healthcare reported.
  • Penn Medicine made it available to 10,000 clinicians across its seven-hospital system, according to Fierce Healthcare.
  • Memorial Sloan Kettering announced in September 2026 that it would embed OpenEvidence in its Epic workflow and put its OncoKB cancer genomics database into OpenEvidence, per MSK's announcement.

The demand came from clinicians, not IT departments. Mount Sinai's chief clinical innovation officer told Fierce that when he asked a senior physician when the deal should happen, the answer was "Yesterday." All three of the early health systems skipped pilots and went straight to enterprise-wide deployments.

The Sutter, Mount Sinai and Cedars-Sinai deployments run on the ad-supported model. Nadler told Fierce that a non-ad-supported enterprise version, "similar to Anthropic's business model," is in the works for large systems that want it.

8. It raised money fast, but on its own terms

OpenEvidence went from a $1 billion valuation to $15 billion in 19 months.

RoundDateAmountValuationLead investors
Series AFebruary 2025$75M$1BSequoia
Series BJuly 2025$210M$3.5BGV, Kleiner Perkins
Series COctober 2025$200M$6BGV
Series DJanuary 2026$250M$12BThrive Capital, DST Global
Unnamed roundSeptember 2026$250M$15BAndreessen Horowitz, Byers Capital

The Series A was its first institutional round and brought total funding above $100 million, CNBC reported. The company announced the Series B itself. Nadler confirmed the Series C valuation to Fierce Healthcare. The company announced the Series D in January 2026.

All told, OpenEvidence has raised nearly $1 billion, Becker's reported in September 2026. Investors besides the leads include Nvidia, Coatue, Blackstone, BOND, Craft Ventures and Mayo Clinic.

The interesting part is a round it didn't take. In July 2026, OpenEvidence considered raising $200 million after receiving offers at around a $20 billion valuation, but was unlikely to proceed partly because it would dilute founders and shareholders, PYMNTS reported, citing The Information. The same source said the company had held acquisition talks with a large tech company.

Two months later it raised at $15 billion instead, below the floated $20 billion. Founder control matters here: Forbes reported in July 2025 that Nadler owned roughly 60% of the company.

Nadler has also said he does not plan to spend like a frontier lab. "We're not running this like a private equity portfolio company, but we're also not planning on burning billions of dollars over the next year," he told CNBC in January 2026. On an IPO, he said model companies like OpenAI and Anthropic should go public first.

9. It is reaching past search into pharma, global health and drugs

The next phase of OpenEvidence's growth goes beyond answering doctors' questions.

Pharma software. In October 2025, OpenEvidence and Veeva Systems announced a long-term partnership to build Open Vista, which will use AI to match patients with clinical trials and help drugmakers understand unmet needs, according to Veeva's announcement. That would sell software to pharma companies, not just ads.

Global expansion. In September 2026, OpenEvidence and Anthropic began offering a specialized version free to clinicians in about 100 low- and middle-income countries. Anthropic is providing Claude credits, engineers and funding for the effort, Fierce Healthcare reported. The core product is already free to clinicians in the U.S. and Europe.

Its own drugs. Nadler told Forbes the company plans to use data from its MSK and NCCN relationships to build a pipeline of cancer therapies, starting with rare cancers, as Refresh Miami summarized. He said he expected the first drug to enter clinical trials before the end of 2026, with three to five more candidates in 2027.

Clinical trials connect all three. Nadler told Goldman Sachs that 1 in 3 trials are discontinued, mainly because they can't find enough patients, and that OpenEvidence wants to "match supply and demand on clinical trials" through the doctors already using it.

How fast did OpenEvidence grow?

OpenEvidence's revenue figures are a mix of company statements, analyst estimates and press reports. Here is the clearest version:

"Annualized" means a recent month's revenue multiplied by 12. It is not audited annual revenue. At $15 billion, the latest valuation is about 50 times the reported $300 million run rate.

Usage grew even faster. These are company-reported counts of clinical consultations by verified U.S. clinicians:

That is more than a 100-fold increase in two years. On March 10, 2026, OpenEvidence handled 1 million consultations in a single day, the company said. As Nadler has pointed out to Fierce Healthcare, there are only about 1 million doctors in the entire U.S.

OpenEvidence's growth loop

OpenEvidence's growth comes down to one loop:

  1. A doctor gets a fast, cited answer from licensed journals, for free.
  2. A colleague sees it on the ward and asks what it is.
  3. More verified doctors sign up and use it daily.
  4. A large, verified audience lets OpenEvidence charge pharma advertisers high rates.
  5. Ad revenue funds more compute, more free tools and more content deals.
  6. More journals and medical societies sign on, because that's where doctors are.
  7. Better content and tools make the answers more useful, and doctors come back more often.
  8. Health systems formalize what their doctors already use, adding enterprise channels.

Most enterprise software in healthcare starts with the hospital and works down to the doctor. OpenEvidence started with the doctor and let hospitals catch up.

What could slow OpenEvidence down?

OpenAI is giving away a direct competitor. On April 22, 2026, OpenAI launched ChatGPT for Clinicians, free for verified U.S. physicians, nurse practitioners, physician assistants and pharmacists. It follows the same free, bottom-up approach and adds documentation tools, though NBC News noted it does not currently license the same journal content.

Incumbents are pitching "no ads." UpToDate, long the default clinical reference, launched its own AI tool. Its spokesperson told NBC News that about 2,000 hospitals and health systems had signed up as of April 2026 and that "healthcare AI must prioritize patient safety, transparency, and freedom from advertising."

Doximity is fighting for the same doctors. OpenEvidence sued Doximity in federal court in Massachusetts on June 20, 2025, according to the court docket, alleging improper access to its platform. On October 15, 2026, the companies told the judge they had reached a settlement, Law360 reported. Doximity has also claimed that its AI outperformed OpenEvidence in a safety study, per Investing.com.

The ad model has a ceiling and critics. Bank of America estimates the pharma advertising market at about $3 billion, growing 5% to 7% a year, according to Investing.com. A January 2026 Silicon Valley Bank report questioned whether ad and software models can justify valuations like OpenEvidence's, STAT reported. Nadler himself has said the ad model may not be the one the company takes into the future, according to NBC News.

Accuracy is still being tested. A December 2025 academic study, not yet peer reviewed, found OpenEvidence answered complex medical questions accurately less than 45% of the time, NBC News reported. Some health systems, such as MaineHealth, ask doctors not to enter patient health information into it.

The valuation already priced in a lot. OpenEvidence reportedly set aside offers at around $20 billion in July 2026, then raised at $15 billion in September. It now needs pharma software, enterprise contracts or drugs to grow beyond advertising.

Borrowing OpenEvidence's doctor-to-doctor word of mouth

OpenEvidence says 95% of new users heard about it from another physician, so the closest thing most startups can copy is showing up, usefully, where their buyers already compare notes. Okara's Reddit Agent finds high-intent Reddit threads in your niche and drafts replies you post yourself. Its LinkedIn Agent drafts posts that keep you visible to the professionals you want recommending you.

More startup growth stories

Frequently asked questions

What is OpenEvidence's valuation?

OpenEvidence is valued at $15 billion after raising $250 million in September 2026 in a round led by Andreessen Horowitz and Byers Capital. CEO Daniel Nadler confirmed the round, though the company did not issue a press release. The previous round, a $250 million Series D in January 2026, valued it at $12 billion.

How much revenue does OpenEvidence make?

OpenEvidence does not publish detailed revenue figures. The Information reported in July 2026 that it was generating about $300 million in annualized revenue, roughly $25 million a month, and was cash-flow breakeven. The company said it topped $100 million in annualized revenue in 2025, and Bank of America put its end-of-2025 run rate at about $150 million.

How much funding has OpenEvidence raised?

OpenEvidence has raised nearly $1 billion across five institutional rounds: a $75 million Series A (February 2025), a $210 million Series B (July 2025), a $200 million Series C (October 2025), a $250 million Series D (January 2026) and a $250 million round in September 2026. Before that, Nadler self-funded the company and raised a friends-and-family round in 2023.

How does OpenEvidence make money?

OpenEvidence is free for verified clinicians and makes money from advertising, mostly from pharmaceutical and medical device companies. Ads appear when clinicians search topics related to the advertiser's disease area. The company is also developing a paid, ad-free enterprise version for large health systems and building pharma software with Veeva.

Who founded OpenEvidence?

Daniel Nadler and Zachary Ziegler founded OpenEvidence. Nadler, the CEO, previously founded Kensho, an AI company sold to S&P Global in 2018. Ziegler, the chief technology officer, was a Harvard PhD student in AI. The company is based in Miami.

How many doctors use OpenEvidence?

As of September 2026, 1.12 million license-verified U.S. clinicians use OpenEvidence, including physicians, nurses, nurse practitioners and physician assistants. The company told NBC News in May 2026 that about 65% of U.S. doctors, roughly 650,000, actively use it. U.S. clinicians consulted it 42 million times in August 2026.

Which medical journals does OpenEvidence have deals with?

OpenEvidence has content agreements with the New England Journal of Medicine, the JAMA Network's 13 journals, the National Comprehensive Cancer Network, the American College of Cardiology, the Cochrane Library and the American Diabetes Association, among others. In September 2026 it added Memorial Sloan Kettering's OncoKB cancer genomics database.

Is OpenEvidence going public?

There is no announced IPO. In January 2026, Nadler told CNBC that foundation model companies such as OpenAI and Anthropic should go public first and application companies would follow. In September 2026, Business Insider cited a person familiar with the matter who said the company could be open to a sale, though OpenEvidence may prefer to stay independent.

Sources

Tags

Growth stories

Ready to grow?

Enter your website and let the AI CMO handle the rest.

Trusted by 150,000+ businesses · 7-day free trial