Direct-to-Patient Pharma Analytics: KPIs, Data and Build Plan

What direct-to-patient pharma analytics measures

Direct-to-patient pharma analytics measures whether a manufacturer's direct channel, its website, telehealth partners, cash-pay pharmacy and listings on aggregators such as TrumpRx, turns interest into patients who start therapy and stay on it, at a net revenue the brand can sustain. It links media and web data to consultations, prescriptions, fills and refills, then compares the direct route with the insured route through hubs and specialty pharmacies.

The channel has grown faster than the measurement behind it. Lilly launched LillyDirect in January 2024 and Pfizer launched PfizerForAll in August 2024. TrumpRx launched in February 2026, and by early May Pharmaceutical Commerce counted 17 manufacturers with pricing agreements on it; by June it listed more than 800 drugs after generics were added. Yet research from DigiPharma Insights and Freshpaint found that 83% of direct-to-patient pharma marketers struggle to connect marketing activity to prescription fills, and 55% cannot connect website visits to telehealth consultations.

This guide sets out the KPI at each stage of a direct-to-patient (DTP) funnel, the system that holds each number, a worked example with figures, the privacy limits that shape what you can measure, and a practical build plan. It extends our patient journey analytics service to the direct channel and sits within our specialty pharmacy analytics work.

The DTP funnel and the KPI at each stage

• 1. Reach and visit

  Who arrives at the brand's DTP page and from which campaign, search term or referral source.

  • Visits by source
  • Cost per qualified visit
  • Share from prescriber referral versus media

• 2. Eligibility and intent

  How many visitors start an eligibility check, benefit check or savings enrolment, and how many finish it.

  • Start rate
  • Completion rate
  • Drop-off by step

• 3. Consultation

  For programs with telehealth partners: booked, completed and prescribed consultations.

  • Booking to completion
  • Prescription rate per consult
  • Days from visit to consult

• 4. Route chosen

  Whether the prescription goes to the cash-pay pharmacy or through insurance with benefit verification and prior authorization.

  • Cash versus insured split
  • Reasons for switching route
  • Prior authorization approval rate

• 5. First fill

  Whether the patient receives the first dispense, and how long it takes.

  • First-fill rate
  • Days from prescription to fill
  • Abandonment by out-of-pocket cost

• 6. Persistence and refill

  Whether patients stay on therapy, which is where DTP economics are won or lost.

  • Refill rate at 30, 60 and 90 days
  • Proportion of days covered
  • Discontinuation reasons

• 7. Net revenue per patient

  What each retained patient is worth after cash price, fees, partner costs and media.

  • Net price by route
  • Acquisition cost per persistent patient
  • Comparison with the insured channel

Where each number comes from, and who owns it

The hard part of DTP measurement is that no single party holds the whole funnel. A typical program spreads it across five systems:

• Web and media analytics: visits, campaigns and form starts. On health-related pages these are constrained by privacy rules, covered below.
• Telehealth partner: bookings, consultations and prescriptions written. Most partners report aggregates by default; patient-level or tokenised data has to be negotiated into the contract.
• Pharmacy partner: cash-pay dispenses and refills. Pharmaceutical Commerce describes TrumpRx as aggregating pricing from partners including Amazon Pharmacy, GoodRx and Mark Cuban Cost Plus Drugs and referring patients to manufacturer-controlled purchasing channels, so the manufacturer's own pharmacy or partner feed remains the source of truth for fills.
• Hub and specialty pharmacy feeds: for patients who take the insured route, benefit verification, prior authorization and dispense status, usually with each pharmacy's own status codes.
• Claims or prescription data: the wider market, used to check whether DTP patients are new to the brand or simply moved from the insured channel.

The practical rule: write data rights into every partner contract before launch. You need the fields, the frequency, the identifier or token to link records, and the right to combine them. Linking patients across sources without exchanging identities is usually done with patient tokenization, and the insured side needs the same joins described in our guide to a unified patient view across hub, SP and claims data.

Why DTP teams cannot connect spend to fills

Three problems explain the 83% figure, and only one of them is technical.

1. Privacy limits on tracking. Health-related web pages are a regulated place to drop advertising pixels. HHS Office for Civil Rights guidance on online tracking technologies, partly vacated by a federal court in Texas in June 2024, put pixel use on health sites under scrutiny, and state laws such as Washington's My Health My Data Act add consent duties for consumer health data. Freshpaint's own research found only 1% of DTP marketers report a fully integrated, privacy-compliant patient data ecosystem. The answer is server-side, consented, minimised collection designed with your privacy and legal teams, not more pixels.
2. Partners that report in aggregate. If the telehealth partner sends a monthly PDF and the pharmacy sends a different monthly total, the funnel cannot be joined at patient level. This is a contract problem first.
3. No shared definitions. Marketing counts a conversion at form completion, the pharmacy counts a new patient at first dispense, and finance counts net revenue at invoice. Until one definition of a DTP patient and a DTP start exists, every report will disagree.

None of this needs an exotic tool. It needs a data model with agreed definitions, consented collection, and pipelines that load each partner feed and check it.

Worked example: a DTP funnel with real arithmetic

The figures below are illustrative, chosen to show how the funnel is read rather than to describe any real program. A brand's DTP page receives 40,000 visits in a month.

• 15% start an eligibility check: 6,000
• 40% of those complete a telehealth consultation: 2,400
• 75% of consultations end with a prescription: 1,800
• 1,100 prescriptions go to the cash-pay pharmacy and 700 go through insurance

Cash route: 85% receive a first fill, so 935 patients start. Insured route: prior authorization is approved for 60% (420) and 90% of those fill, so 378 patients start. Total first fills: 1,313, and the cash route looks far stronger.

Now add persistence at month three. If 55% of cash patients are still filling, that is 514 patients. If 70% of insured patients are still filling, that is 265. The cash route still leads on volume, but the gap in persistence changes the economics: a cash patient who stops after two months may earn less than an insured patient who stays a year, even at a lower net price after rebates. A dashboard that stops at first fill would have sent more media budget to the route that keeps patients for the shortest time.

The same funnel also shows where to act. A 40% consultation completion rate points at booking friction with the telehealth partner. A 60% approval rate on the insured side points at prior authorization support, which is the same time-to-therapy problem tracked in our guide to patient access KPIs. Abandonment by cost band, explained in prescription abandonment versus reversal, tells you whether price or process is losing patients at the counter.

How DTP measurement differs from hub and copay programs

• Copay card programs

  You see redemptions through the program administrator, but rarely the patient's journey before the pharmacy counter.

  • Good: redemption volume and cost
  • Weak: why patients never reach the pharmacy
  • Weak: link to media

• Hub and specialty pharmacy

  You see referrals, benefit verification, prior authorization and dispenses in detail, under data agreements built for it.

  • Good: status and time to therapy
  • Good: persistence through dispense data
  • Weak: anything before the referral

• Direct-to-patient

  You can see more of the journey than either, from first visit to refill, but only if partner contracts and consented collection allow it.

  • Good: end-to-end funnel if designed
  • Risk: privacy limits on web data
  • Risk: aggregate-only partner reports

How we build direct-to-patient pharma analytics

Common mistakes with DTP measurement

When a DTP analytics build is premature

If the program is a few months old and handles a few hundred patients a month, a monthly reconciliation of partner reports in a spreadsheet may be enough while you fix the contracts. If partners only supply aggregates, spend the effort on renegotiating data rights first; no model can join totals at patient level. And if your insured-channel reporting is still rebuilt by hand, fix that foundation first, because the most useful DTP question is always a comparison with the insured route.

The investment makes sense when DTP volume is material to the brand, more than one partner is involved, media spend is being allocated by channel, and leadership is asking whether the direct channel is adding patients or moving them. Adherence measures for the persistence stage are explained in PDC versus MPR, and the brand-level reporting that DTP results feed into is covered on our pharma commercial analytics consulting page. Our founder came up through pharma commercial operations consulting at ZS, and we build inside your own warehouse and BI tools under your data agreements.

Direct-to-patient pharma analytics questions