Particle Health

A digital health company, payer, or value-based care organization that needs clinical data from across the U.S. health system but cannot afford the 12+ months and millions of dollars required to negotiate, integrate, and maintain direct connections to each health information network. Particle's navigation splits traffic across four markets (value-based care, payers, health systems, developers) while the homepage hero speaks to none of them specifically. The word "innovate" gestures at developers. The logo bar shows One Medical. The product pages pitch payer risk stratification. The site cannot decide who is reading.

Clinical data API. Health data interoperability. An existing shelf that the buyer knows to shop. Particle sits alongside Redox, Health Gorilla, Zus Health, and (as of the antitrust dispute) Epic's own Payer Platform. The investor pitch calls Particle "the Plaid for healthcare," which is the right analogy for a VC deck but not for a buyer who has never heard of Plaid. The homepage uses two different registers: developer language ("API platform," "RESTful," "FHIR R4") and business language ("actionable clinical insights," "innovative solutions"). Neither is how the VP of Clinical Operations at a payer describes her problem. She thinks: "I need complete patient records from every network, and I don't have 18 months to get them."

Health data interoperability is in the early-to-mid pragmatist phase in 2026. TEFCA is live. The 21st Century Cures Act information-blocking rules have teeth. The visionary buyers (the ones excited by "destroying the fax machine") already bought years ago. The buyers remaining are pragmatists comparing vendors on coverage, data quality, compliance posture, and implementation speed. They don't want to "innovate." They want to ship.

Copy implication: The hero's innovator-phase language ("Transform medical records," "fuel innovative solutions," "actionable clinical insights") is addressing a buyer who already bought. The pragmatist landing on this page in 2026 needs coverage numbers, data quality metrics, named references, and a compliance checklist. Not aspiration.

The chain is intact. Every attribute maps to a value that maps to a named segment. Particle has real, differentiated substance: the single-API aggregation, the data processing pipeline, the 12-week implementation. The problem is that none of this specificity reaches the hero. The hero says "transform medical records into actionable clinical insights." The substance says "one API, every network, 12 weeks, 320 million patients."

"One API call, every health network in the country, clinical data back in weeks instead of months."

Defensible, specific, and falsifiable. Either one integration pulls from all three national networks or it doesn't. Either you're live in 12 weeks or you aren't. This is the entire reason a product lead at a digital health company or a VP of Clinical Ops at a payer would take a first meeting. It does not appear in Particle's hero, subhead, or CTA.

Founded 2017 by Troy Bannister (CEO turned CSO; 15 years in healthcare, clinical researcher, EMT, Rock Health Top 50) and Dan Horbatt (CTO; ex-Google, ex-Palantir, ex-Microsoft). Jason Prestinario became CEO in May 2023 (ex-Komodo Health GM, ex-Palantir, Stanford MS). The Palantir DNA runs deep: two of three key leaders built large-scale data infrastructure at one of the world's most demanding data companies before applying it to healthcare.

• Jul 2019: $2.3M Seed led by Collab Fund (angels from Flatiron Health, Clover Health, Plaid)
• Apr 2020: $12M Series A led by Menlo Ventures
• Jul 2022: $25M Series B led by Canvas Ventures (Rebecca Lynn called Particle "the Plaid for healthcare")
• Dec 2024: $10M Series B-II from PruVen Capital and Menlo Ventures
• Total raised: ~$49.3M
• 2022 growth: 450% ARR increase and 300% customer increase year-over-year (Series B announcement)
• Coverage: 320M+ patient records, 160,000+ health systems/practices/clinics, 90% national patient discovery rate, 105 records per patient per search
• Named customers: One Medical, Reveleer, xCures, Cureatr, Upperline Health, Yuvo, OSH

1. The Epic lawsuit is both the biggest risk and the biggest positioning opportunity they're not using. Epic cut off Particle's access in 2024. The antitrust case is in discovery. If Particle wins, it reshapes healthcare data access. If it drags on or loses, Particle's access to Epic-sourced data remains uncertain. Either way, the fight itself is proof that Particle matters enough to threaten a monopoly. Not a word about it on the homepage.
2. TEFCA is a double-edged sword. TEFCA promises standardized network access, which is what Particle was built to solve. As TEFCA matures, the "build direct" alternative gets easier. Particle's "TEFCA in a Box" product is an attempt to stay ahead, but the category's future depends on whether TEFCA reduces the integration pain enough to make intermediaries optional.
3. The payer-vs-developer identity split is unresolved. The homepage hero speaks to developers ("API platform," "Try It Right Now"). The product pages speak to payers and VBC orgs (risk stratification, care gap closure, readmission prevention). Particle cannot be the developer API company and the payer insights company in the same breath. The buyer profiles are different. The sales cycles are different. The copy is trying to serve both and serving neither.
4. The multi-product positioning trap. Particle now has four distinct products (Snapshot for on-demand records, Signal for event alerting, Workbench for analytics, Navigator for care coordination) sold to four different segment pages. When a company has multiple products, it doesn't know what to position. The homepage has to decide whether it's positioning the company ("Particle is the clinical data layer") or a specific product ("Snapshot pulls patient records in seconds"). Right now it does neither. Trying to position four products on one homepage waters down all four. The answer is usually to position the company around the shared infrastructure (one API, every network) and let individual product pages carry the segment-specific positioning.
5. The "treatment purpose of use" overhang. The entire Epic dispute centered on whether payer queries qualify under "treatment" purpose of use on health information networks. This governance question remains unresolved industry-wide. It shadows every payer deal Particle closes until network governance catches up.

A cold reader landing on Particle's homepage encounters "Transform medical records into actionable clinical insights." This could be a data analytics company. It could be an EHR vendor. It could be a clinical decision support tool. Nothing in the hero tells you this is a clinical data API, that it connects to health information networks, or that it solves an integration problem. The category cue is missing entirely.

The site tries to speak to four audiences (VBC, payers, health systems, developers) and the hero has not decided which one is reading. The "Try It Right Now" CTA and developer portal suggest a PLG motion aimed at engineers. The "Contact Us" CTA and "Let's connect" suggest enterprise sales aimed at executives. These are two different go-to-market strategies jammed into one hero.

The awareness stage is miscalibrated. "Innovate" is innovator-phase language. The buyers shopping for clinical data APIs in 2026 are pragmatists comparing Particle against Redox and Health Gorilla on coverage, data quality, price, and compliance. They don't need to be inspired. They need to be convinced.

The homepage hero targets the shared value every segment recognizes: replacing months of integration work with one API call. VBC orgs, payers, health systems, and developers all experience the same pain. The hero speaks to that pain in language that transcends segment jargon.

• Names the enemy in the first sentence. The enemy is the multi-month integration slog every buyer has lived through. The reader recognizes her own pain before finishing the first line.
• The punch lands last. "One API call" sits at the end where English readers weight it most. The contrast (months vs. one call) is the entire value proposition in eight words.
• The subhead stacks three load-bearing numbers. 320 million records, 90% discovery, 12 weeks. Each number does work the current hero's adjectives cannot.
• Named customers live inside the subhead. One Medical, Reveleer, and xCures are specific enough to mean something to a buyer evaluating clinical data vendors. They earn their position next to the claim.
• Networks are named, not gestured at. "Carequality, CommonWell, and eHealth Exchange" tells the solution-aware buyer exactly what shelf this sits on. The current hero never names a single network.
• CTAs promise rewards, not meetings. "Query your first patient free" is a specific, no-commitment action that proves the product works. "See the coverage map" gives the researcher exactly what she needs before a call.

For buyers actively comparing Particle against Redox, Health Gorilla, Zus, or building direct. They know the category. They need differentiation.

• Names the enemy as the category default. "Most clinical data APIs connect to one network" is the status quo the in-market buyer already suspects. Confirming it earns trust and creates the opening for the punchline.
• Differentiates on the axis the buyer is already evaluating. Network breadth is the decision criterion for the solution-aware buyer. This variant leads with the dimension that matters most in a shortlist comparison.
• Stacks the data quality pipeline into the subhead. 90% dedup and 17% gap rate are specific enough to mean something to a data team. No competitor publishes numbers this granular. These are hard to copy.
• Networks are listed by name, not category. Naming Carequality, CommonWell, and eHealth Exchange individually tells the buyer "we connected to each one" rather than "we have broad coverage" (which is what everyone says).
• CTAs serve the comparison mindset. "Compare coverage side by side" tells the buyer she'll get an honest comparison. "Read the data quality whitepaper" gives the technical evaluator something to bring back to her team.
• Passes the swap test. No other vendor can claim simultaneous connection to all three national networks plus state HIEs through a single API. Redox, Health Gorilla, and Zus each have different network profiles.

For VBC teams and payers not yet shopping for a clinical data API. They're still stitching together ADT feeds, faxing records, or manually chasing charts. They haven't framed the integration problem as solvable.

• Reframes the problem the buyer didn't know she had. "40% of the clinical picture" is a claim that makes the out-of-market buyer realize her status quo is worse than she thought. Rational drowning happens in the first sentence.
• Doesn't sell a product. Teaches something the buyer will repeat. "The other 60% is in a network you're not connected to" is a sentence a VP of Clinical Ops would say in her next leadership meeting. That's the goal.
• Uses a single surprising number. 105 records per patient per search is unexpected and concrete. It tells the unaware buyer something she didn't know was possible.
• Single CTA shows, doesn't tell. "See what 105 records per patient looks like" is a specific, visual promise that makes the buyer curious without requiring a meeting.
• Names the status quo without naming a competitor. "A network you're not connected to" is the problem. Not Redox. Not Health Gorilla. The buyer's own disconnection is the enemy.

• Mentions the Epic lawsuit. The lawsuit is a positioning asset, but weaponizing it in the hero would make the copy about a legal fight instead of the buyer's problem. It belongs in a trust section or press page, not the first thing a cold reader sees.
• Uses any wallpaper word. No "innovative," "seamless," "holistic," "actionable," "empower," "leverage," or "transform."
• Tries to be for all four audiences simultaneously. The homepage rewrite speaks to the shared value. The campaign variants are tuned by awareness stage, not segment.
• Claims anything Particle hasn't already published. Every number, customer, and network name is already on their public site or in published press.
• Writes prescriptive strategy. The rewrites demonstrate what the existing substance could sound like. They don't tell Particle what to build.

"Is the category cue the one the buyer already shops?"

The homepage never names a category. "Product-ready insights" is not a shelf anyone shops. A buyer comparing clinical data APIs (the actual category) gets no signal that Particle belongs on that shelf. The closest the page comes is the subhead's mention of "API platform," but it's buried behind "simple, elegant" and doesn't register as a category cue. The buyer who just read Redox's hero (which says "healthcare data platform" in the first sentence) and then lands here can't tell if Particle competes with Redox or sells something else entirely.

"Does the copy confront what the buyer does today?"

No. The status quo for most buyers is one of three things: (1) building direct integrations network-by-network over 12+ months, (2) stitching together 6+ ADT feeds from different vendors for partial coverage, or (3) going through Epic's Payer Platform because it's the path of least resistance. None of these alternatives is named or confronted. The buyer cannot feel the contrast between "what I do today" and "what Particle offers" because the hero doesn't acknowledge that "what I do today" exists.

"Are the non-copyable differentiators on the page?"

They're on the page, but not in the hero. The substance (single API to all three national networks, 90% dedup, imputation from 50% to 17% gaps, 12-week contract-to-value, free developer portal) lives in product pages, documentation, and a "Ready-Made Partnerships" card below the fold. These are the non-copyable differentiators. But a buyer scanning for 4 seconds sees "transform medical records" and "actionable clinical insights." She leaves before reaching section 2.

"Does the copy lead with blunt, published numbers?"

The subhead mentions "320+ million patients' medical records," which is strong. But it's the only number in the hero, and it's doing coverage work rather than outcome work. The numbers that would convince a pragmatist (90% discovery rate, 105 records per patient, 12-week implementation, 25% more events than ADT feeds, 90% deduplication rate) are buried across product and documentation pages. The hero asks the reader to believe "actionable clinical insights" on faith. The proof exists. The hero just doesn't use it.

"Does the homepage lead with the shared value the best-fit segments recognize?"

The shared value across all of Particle's segments (VBC, payers, health systems, developers) is the same: replace months of network integration with one API call. This is segment-transcending. But the hero doesn't express it. Instead, "transform medical records into actionable clinical insights" speaks to no segment specifically. The navigation splits into four market pages that each have their own hero, meaning the homepage hero has abdicated its job to interior pages. The homepage hero should do the one thing all four segments need to hear and then route by interest below the fold.

"Can a cold reader get it in four seconds?"

No. After four seconds, a cold reader knows: (1) this company has something to do with medical records, (2) there are 320 million patient records involved. She cannot answer: What problem does it solve? Who is it for? Why should I care? How is it different from the three other tabs I have open? The four-second test fails because "transform medical records into actionable clinical insights" tells her nothing she couldn't infer from the URL alone. It could be an EHR, a clinical decision support tool, or a data warehouse.

"Is the full adoption package visible?"

Particle actually has strong whole-product signals: HITRUST and SOC2 certification, HIPAA compliance, OAuth 2.0, SSO, MFA, audit APIs, a free developer portal, published documentation, multiple output formats (FHIR R4, C-CDA, flat JSON, Parquet), and a 12-week implementation timeline. These are visible on the page but mostly in the footer. An important distinction: HITRUST, SOC2, and HIPAA are objection handlers, not value drivers. They don't make someone buy Particle. They prevent a procurement team from blocking a deal that's already in motion. The pragmatist buyer expects them the way she expects running water in an office. They need to be easy to find (first scroll, not footer) so they clear the objection fast, but they should never be confused with the differentiated value. The differentiated value is the single API, the network breadth, and the 12-week implementation. The compliance posture is the thing that keeps the deal from dying after the buyer already wants it.

"Is the trust evidence overwhelming and visible?"

Trust evidence exists but is scattered. The logo bar shows One Medical, Reveleer, Yuvo, Upperline Health, and OSH (good). There are testimonials from One Medical, xCures, and Cureatr (good). There are award badges from CB Insights, Digital Health Awards, and Built In (moderate). But the trust evidence is not overwhelming or stacked near the claims it proves. One Medical's testimonial is generic ("promising solution"). The strongest trust signal (that Epic felt threatened enough to block Particle's data access, which is de facto validation from the market's dominant player) is completely absent.

"Is the compliance posture front-and-center?"

Particle's compliance credentials are real: HITRUST, SOC2, HIPAA, OAuth 2.0, MFA, SSO, audit APIs. They selected CommonWell as their QHIN for TEFCA. They have a "TEFCA in a Box" product. But the compliance story sits in a badge row near the bottom of the page with the heading "Secure & Compliant is our middle name" (which reads as a joke, not a trust signal). These credentials are objection handlers: they don't create demand, they prevent a deal from stalling in procurement. But they need to be easy to find precisely because the Epic dispute raised governance questions that every buyer's compliance team will ask about. The fix isn't to lead with compliance (that would confuse objection handling with value). The fix is to move the compliance badges to first-scroll position so the pragmatist's procurement officer can check the box in 10 seconds and get out of the way.

"Does the copy sound like the buyer's own words?"

The hero sounds like a marketing team writing for a pitch deck. "Transform medical records into actionable clinical insights" is builder language. The buyer's words sound like this: "We spent 9 months integrating with Carequality and we still don't have CommonWell." Or: "We're using 6 different ADT feeds and still missing 30% of discharges." Or: "We need patient records before the appointment, not three weeks after." None of these sentences appear anywhere on the homepage. The testimonials that do exist (One Medical, xCures, Cureatr) are generically positive rather than specifically painful. The voice of the customer is polite where it should be blunt.

Composite read: Particle Health has real, defensible, differentiated substance. The single-API-to-all-networks architecture, the data processing pipeline, the 12-week implementation, and (perhaps most importantly) the market validation of being the company Epic tried to shut down. This is not a company with a positioning problem born of weak product. This is a company with a messaging problem born of a hero that trades in adjectives when it should trade in proof.

The positioning flow holds up at every step: the competitive alternatives are clear (build-direct, Redox, Health Gorilla, Zus, Epic Payer Platform). The unique attributes are real and non-copyable (single API to all three national networks, 90% dedup pipeline, 12-week implementation). The value those attributes enable is concrete (months of integration work eliminated, clean data without a data engineering team). The customers who care most are identifiable (VBC orgs stitching ADT feeds, payers drowning in messy C-CDAs, digital health startups burning runway on integrations). The market category is an existing shelf the buyer already shops (clinical data API). Every component of the positioning chain is intact. What's broken is that the hero communicates none of it.

One important distinction the current page misses: HITRUST, SOC2, and HIPAA compliance are objection handlers, not value drivers. They don't make someone choose Particle over Redox. They prevent a deal from dying in procurement. They should be easy to find (first scroll) but never confused with the differentiated value. The differentiated value is network breadth and data quality. The compliance posture is the table stakes that keep the door open.

The fix is not strategic (they don't need to pick a new category or new segment). The fix is structural: move the substance to the hero, name the enemy, stack the proof, distinguish value from objection handling, and let the product's real advantages do the talking.

• particlehealth.com — homepage hero verbatim, nav structure, logo bar, testimonials, product cards, CTAs
• particlehealth.com/about-us — founding story, team roster, mission statement, awards
• particlehealth.com/value-based-care — VBC use cases, Signal product details, coverage claims
• particlehealth.com/payers — payer use cases, Snapshot and Signal product positioning
• docs.particlehealth.com — technical documentation, network connections, data processing pipeline, output formats, eMPI details

• Particle blog — Series B announcement — $25M raise, 450% ARR growth, 300% customer growth, Canvas Ventures lead
• Collab Fund blog — Particle Health investment — seed round context, angel investors from Flatiron Health, Clover Health, Plaid
• Crunchbase — Particle Health company profile, total funding ($49.3M), investor list, Dec 2024 $10M B-II
• Tracxn — funding timeline, round details, valuation context

• Fierce Healthcare — Epic-Particle dispute (2024)
• STAT News — Epic must face monopoly claims (Sep 2025)
• MedCityNews — Judge advances Particle's antitrust lawsuit (Sep 2025)
• CNBC — Epic boots Particle Health (Apr 2024)
• TechCrunch — Can API vendors solve healthcare's data woes (Apr 2020)
• TechCrunch — Why this founder replaced himself as CEO (Jul 2023)
• Health API Guy — The Particle v. Epic Casebook (Oct 2025)

• One Medical — testimonial on particlehealth.com homepage (verbatim: "We're looking to enable better care for our patients by gathering their external health data before their first visit with us.")
• xCures — Zachary Kaufman, Director of Product Management testimonial on particlehealth.com
• Cureatr — Matt Benedon testimonial on particlehealth.com
• Reveleer, Yuvo, Upperline Health, OSH — named in homepage logo bar

• redoxengine.com — EHR integration platform, direct EHR connections
• healthgorilla.com — health information exchange, QHIN operator, lab data focus
• zushealth.com — shared health data platform, embedded UI components
• Epic Payer Platform — Epic's own payer data access product (central to antitrust dispute)
• Health API Guy — Evolution of Health Information Networks — competitive landscape context
• CB Insights — Particle Health alternatives and competitors

• Troy Bannister — founder, CSO; 15 years healthcare, Rock Health Top 50 in Digital Health
• Dan Horbatt — co-founder; ex-Google, ex-Palantir, ex-Microsoft, CMU
• Jason Prestinario — CEO since May 2023; ex-Komodo Health GM, ex-Palantir, Stanford MS
• Particle blog — CEO appointment announcement
• BusinessWire — Particle selects CommonWell as QHIN (Nov 2024)