What copay program analytics should tell a brand team
Copay program analytics is the analysis of copay card and patient assistance claims to show whether a program is getting the right patients onto therapy and keeping them there at a cost the brand can defend. Done well, it answers three questions: who is using the program and why, how much of the money actually reaches patients rather than payer programs such as accumulators and maximizers, and what the program does to abandonment, time to therapy and persistence.
Most manufacturers already receive monthly reports from their copay vendor. Those reports count claims, redemptions and spend. They rarely link copay activity to specialty pharmacy, hub and claims data, so they cannot say whether a patient who hit their card maximum in April came back for a refill in May. That link is where the value is, and it is what this guide covers: the KPIs with plain definitions, the data you need, how to spot accumulator and maximizer claims, a worked example, the regulatory guardrails, and a step-by-step build.
Greenwolf Tech Labs builds patient services and specialty pharmacy analytics for pharma and biotech brands. Copay data sits naturally alongside the SP and hub feeds we already integrate.
Why copay analytics moved up the agenda
Where copay programs leak money, and why most reports miss it
A copay program is one of the largest controllable lines in a specialty brand's budget, and it leaks in predictable places. Accumulator programs let the copay card pay the patient's share but stop that money counting toward the patient's deductible, so the card runs out mid-year and the patient faces the full deductible. Maximizer programs set the patient's cost share to drain the card's annual maximum evenly across the year, so the brand pays the maximum on almost every enrolled patient. Pharmacy-level misuse, duplicate enrolments and patients enrolled after they no longer qualify add more.
Standard vendor reports show these effects only as higher spend per patient. They do not show which plans, PBMs or pharmacies drive it, or what happened to the patient afterwards. The IQVIA Institute's 2024 report found that patients starting new therapies abandoned 98 million prescriptions at the pharmacy in 2023, and that aggregate patient out-of-pocket costs rose by $5 billion that year. Copay analytics that cannot connect a card claim to a later abandonment or discontinuation will miss the outcome the program exists to change. We explain the measurement difference in prescription abandonment vs reversal.
The copay program KPIs that matter
Enrolment and activation
Enrolled patients, activated cards and the share of commercially insured new starts that use the program. Low activation among eligible patients means the offer is not reaching the point of dispense.
- Activation rate = activated / enrolled
- Penetration = card users / eligible commercial starts
- Split by SP, prescriber segment and region
Cost per patient and per fill
Average manufacturer payment per claim and per patient per year, against the program's design. Rising cost per patient with flat volume is the first sign of maximizer activity.
- Average benefit per claim
- Annual cost per patient
- Share of patients hitting the annual max
Accumulator and maximizer exposure
The share of patients and spend flagged as likely accumulator or maximizer, by plan and PBM. This is the number finance and market access will ask for first.
- Flagged patients and spend
- Top plans and PBMs by exposure
- Trend by month
Time to first fill
Days from referral or enrolment to first dispense for card users against non-users, using SP and hub dates. Affordability support should shorten it.
- Referral to first fill, median
- Card users vs non-users
- By payer and SP
Abandonment after card exhaustion
The share of patients who abandon or delay a fill within 60 days of hitting their card maximum. This links program design directly to lost therapy days.
- Abandonment within 60 days of max
- Gap days after exhaustion
- By plan design
Persistence and adherence
Persistence at 6 and 12 months and proportion of days covered for card users compared with similar patients who did not use the program.
- Persistence at 6 and 12 months
- PDC for card users vs matched non-users
- Discontinuation reasons from SP data
How to spot accumulators and maximizers in claims data
Accumulator signature
The card pays large amounts in the first months of the plan year, reaches its annual maximum early, and the patient's own out-of-pocket then jumps. Look for patients whose card benefit is exhausted well before year end followed by a sharp rise in patient pay or a gap in fills.
Maximizer signature
The patient copay on each claim is set near the card's annual maximum divided by the number of fills, so the card pays almost the same high amount every month and is drained exactly by year end. The patient often pays nothing, which hides the cost from patient-facing metrics.
Ordinary high deductible
A genuine high deductible plan also produces large early claims, but the copay falls to a normal coinsurance level once the deductible is met, and the card is not drained in a regular pattern. Separating this from accumulators avoids over-flagging.
The data a copay analytics model needs
The minimum is the copay vendor's claim-level file: claim date, NDC, pharmacy, the patient's plan cost share, the amount the card paid, the amount the patient paid, and a program or patient identifier. Most vendors can supply this weekly or monthly. Add three more sources and the analysis changes from counting to explaining:
- Specialty pharmacy and hub data for referral, benefits verification, prior authorisation, ship and discontinuation dates. Our page on specialty pharmacy data analytics covers how these feeds are ingested and standardised.
- Payer and plan reference data so that thousands of plan identifiers roll up to PBMs and parent payers. Without it, accumulator exposure cannot be reported by PBM. See payer hierarchy in pharma.
- De-identified patient tokens so the same patient can be followed from copay claim to SP status to claims without sharing identities. We explain the approach in patient tokenization for specialty pharma.
All of this is commonly held in a warehouse such as Snowflake, Databricks or Azure, with dashboards in Power BI or Tableau. Patient-level work should run on de-identified, tokenised data inside your environment, with protected health information handled only under your Business Associate Agreement.
Copay data is one input to a wider patient services picture. For the KPIs around it, see KPIs for patient support hub programs and patient access KPIs. For the adherence measure that pairs with persistence, read PDC vs MPR, and for the payer side, our guide to market access analytics.
A worked example: cost per retained patient
Here is an illustrative calculation with assumed numbers, to show the method rather than a benchmark. A specialty brand has 3,000 commercially insured patients on its copay program and spends $18 million a year on it, an average of $6,000 per patient.
The analytics flags 900 patients (30 percent) as likely maximizer, each draining the full $12,000 annual maximum: $10.8 million. The remaining 2,100 patients cost $7.2 million, about $3,430 each. So 30 percent of patients account for 60 percent of spend, and much of that extra spend lowers the plan's cost rather than the patient's.
Now add outcomes. Suppose 12-month persistence is 62 percent for non-flagged card users and 64 percent for maximizer patients. The program is not buying extra persistence in the maximizer group; it is funding the plan. That points to specific actions: plan-specific benefit caps where your legal team agrees they are permitted, alternative support routes for affected patients, and a conversation with market access about those PBMs. The same frame, cost per retained patient-month by segment, also feeds the copay line of your gross-to-net waterfall.
Mistakes that distort these numbers:
- Reporting spend and redemptions without linking to fills, abandonment or persistence.
- Flagging every high early copay as an accumulator and overstating exposure.
- Rolling plans up to PBMs by name matching instead of a maintained payer master.
- Patient services, finance and brand each calculating card exhaustion differently.
- Changing program caps without a baseline to measure the effect against.
How to build copay program analytics in six steps
Agree the questions and definitions
With brand, patient services, market access and finance, define enrolment, activation, card exhaustion, abandonment and persistence once. Write the definitions down before any build.
Get the claim-level copay file
Ask your copay vendor for claim-level data with plan, pharmacy and paid amounts, on a fixed schedule and format, written into the contract if necessary.
Link to SP, hub and payer data
Join copay claims to SP and hub records through patient tokens, and roll plans up to PBMs and parent payers through a payer master.
Build the flags
Code the accumulator, maximizer and high-deductible rules, test them against a sample reviewed by your patient services team, and tune before reporting.
Publish the dashboards
One view each for brand, patient services and finance, on the same data and definitions, drillable to plan, PBM, SP and region.
Automate refresh and alerts
Load each file automatically, check for late or restated data, and alert when a plan's exposure or a card exhaustion rate moves outside its normal range.
Guardrails: Medicare, anti-kickback and patient privacy
Copay analytics sits next to some of the most regulated activity in pharma, so the scope needs to be clear before any build.
- Federal programs. Under the federal Anti-Kickback Statute, manufacturer copay cards are generally excluded for patients covered by Medicare, Medicaid and other federal health programs. Analytics should confirm that eligibility screening works and report any federally insured claims as exceptions, not treat them as a segment to grow.
- Medicare Part D redesign. The Inflation Reduction Act capped Part D out-of-pocket costs at $2,000 in 2025, and CMS set the 2026 cap at $2,100. That changes affordability for Medicare patients, who rely on foundations and free drug programs rather than copay cards, and it changes how brands compare commercial and Medicare persistence.
- State rules and litigation. Several states require insurers to count third-party copay assistance toward a patient's deductible in some circumstances, and federal rules on accumulators have been the subject of litigation. Treat plan-level policy as data that changes, and let your legal and market access teams interpret it.
- Privacy. Patient-level linking should use de-identified tokens, run in your environment and follow your HIPAA and Business Associate Agreement obligations. No patient data should be sent to consumer AI tools.
Vendor reports, an analytics vendor, or your own model
Copay vendor reports
Program administrators provide standard monthly reporting on claims, redemptions and spend, and some offer accumulator and maximizer flags. Useful and quick, but limited to their own data, so outcomes such as abandonment and persistence are hard to see.
Specialist analytics vendor
Some data and analytics firms offer packaged copay and affordability analytics combined with their own claims data. Fast to stand up, but the logic is theirs and the data often stays in their platform.
Your own model
Copay, SP, hub, claims and payer data joined in your warehouse, with flags and definitions you own and can audit. Takes longer to build but answers brand-specific questions and feeds GTN, forecasting and patient services reporting from one source.
Questions about copay program analytics
What is copay program analytics?
What is the difference between a copay accumulator and a copay maximizer?
How much copay assistance do manufacturers provide?
Can copay cards be used for Medicare patients?
Which KPIs should a copay program track?
What data do we need from our copay vendor?
How long does it take to build copay analytics?
Want to see what your copay program is really funding?
Tell us which copay vendor you use, how your SP and hub data arrive and what your team reports today. We will outline what linking them would show. Please do not send any patient information.
Part of our Pharma Commercial and Patient Analytics hub. Start with our specialty pharmacy analytics.