Every Subscription Start Has to Beat a Free Coupon: A GoodRx Subscriptions Growth Teardown

Executive Summary

Industry Trends

GoodRx isn't selling entertainment, it's selling relief from healthcare costs, to people who are by definition price-sensitive and often anxious at the moment of purchase. That changes which parts of the standard subscription-growth playbook are usable and which are radioactive, and it's the lens for every finding below. Three structural shifts define the environment here.

First, the cash-pay prescription market is being reshaped by GLP-1 economics and manufacturers going direct. Weight-loss drugs created an enormous cash-pay category almost overnight, and manufacturers responded by building direct channels and consumer pricing programs rather than leaving access to insurers. GoodRx positioned itself as an access layer for exactly this: its Pharma Direct segment (manufacturer-funded direct pricing programs) grew 76% YoY to $61.6M and is now 31% of total revenue, with 135+ consumer direct pricing programs live.12 The subscription business and this trend feed each other: GoodRx for Weight Loss is a subscription wrapped around GLP-1 access, with an introductory $199/month cash price for Ozempic and Wegovy negotiated with the manufacturer.7

Second, the retail pharmacy footprint is shrinking under the funnel. Management named retail store closures explicitly as a headwind on transaction volume.2 A discount card is only worth what the counter it's redeemed at is worth, so every closed pharmacy thins both the coupon business and the "free medications at participating pharmacies" promise Companion runs on. It also pushes the market toward delivery and e-commerce, where GoodRx says 6,000 pharmacies now use its e-commerce and direct contracting capabilities.2

Third, subscription cancellation and health-data privacy are both regulatory live wires, and GoodRx sits on top of both at once. The FTC's "click-to-cancel" rule was finalized in 2024, vacated by the Eighth Circuit in July 2025 on procedural grounds, and the FTC restarted negative-option rulemaking in early 2026 while continuing enforcement under ROSCA, with state auto-renewal laws active in parallel.10 GoodRx also carries its own scar tissue: a 2023 FTC enforcement action, the first ever under the Health Breach Notification Rule, over sharing users' health data with advertising platforms.11 For a subscriptions growth role specifically, that history means the personalization and lifecycle mechanics that power growth everywhere else have to be engineered inside a much stricter data perimeter here.

The Company

GoodRx's own framing for why Companion exists, from launch-day coverage. Aaron Crittenden, GoodRx's president of Rx Marketplace, described the underlying services as running on a rented network, paying for access to existing provider relationships rather than building from scratch, deliberately scoped to areas of high, recurring utilization. He was explicit that it isn't an insurance product: "We're not building an insurance product... We're trying to build something that can be ancillary to insurance. Or if you were uninsured it would give you a good backstop on some things, not everything." He tied the launch to the broader environment patients are in: "Coverage is getting worse, costs are going up, you've got ACA roll-offs, it's a mess."13

Their Plans

Working reference assembled from goodrx.com's membership and care pages plus the Companion launch release, captured August 2026, since pricing pages change over time and the findings below cite specific numbers from them.46 Prices and eligible medications subject to change.

The subscription portfolio, side by side

What changed from Gold to Companion (May 27, 2026)

Marketing copy, as positioned

Companion leads with "Cut costs, not care" and "From everyday prescriptions to unexpected care," and sells with concrete dollar anchors rather than percentages: $180 average off dental procedures, $1,300 average off LASIK, $650 average off wellness labs, 250+ free medications.4 Two disclaimers do heavy lifting across every page: "GoodRx is NOT insurance," and membership savings can be used alongside insurance but not on the same transaction. GoodRx Care prices visits at $39–70 without membership and $19 with it, 7 days a week, across UTIs, skin, refills for chronic conditions, and weight loss.6

Competitor Analysis

Research-only for now, positioning and business model, not a live product walk. GoodRx's competitive set splits into three groups that pressure the subscription from different directions: free discount cards that erode the reason to pay at all, flat-fee pharmacy programs that compete on price simplicity, and vertically integrated telehealth brands that compete for the exact condition-subscription buyers GoodRx is now going after.9

• SingleCare, Optum Perks, and the free-card field. Structurally the most dangerous, the way free pre-installed tools are for any paid app: free, no enrollment, and good enough for the occasional filler. SingleCare is accepted at 35,000+ pharmacies and sometimes beats GoodRx's coupon price on a given drug on a given day. Every dollar of value Companion gates behind $14.99 has to be value a free card, including GoodRx's own, can't match. That's why the free-med list and the non-pharmacy benefits (telehealth, dental, vision, labs) matter: they're the parts a coupon card structurally can't copy.
• Amazon Pharmacy RxPass. The closest structural comparable to Companion: $5/month flat for a defined list of eligible generics, on top of Prime, with delivery built in. Cheaper than Companion, but narrower (no telehealth, dental, vision, or labs; excludes controlled substances and brands) and gated behind Prime. RxPass is the benchmark a price-sensitive shopper will find in one search, so Companion's pitch has to be legibly "more than meds" or the $10/month gap does the arguing against it.
• Mark Cuban Cost Plus Drugs. Competes on radical price transparency (cost plus a published markup), mail-order only, no subscription required. Weak on urgency and same-day pickup, strong on trust, and trust is the axis GoodRx can least afford to lose on. Cost Plus is less a funnel competitor than a narrative competitor: it defines what "honest drug pricing" looks like in the public conversation Companion sells into.
• Hims & Hers and the vertical telehealth brands. The direct competitors for Weight Loss, ED, and Hair Loss subscriptions, with years of head start on lifecycle marketing, bundled med-plus-visit pricing, and brand permission in exactly these categories. GoodRx's counterpositioning is visible in its own press releases: transparent pricing, FDA-approved meds only, no compounded alternatives, no upsell kits.8 The open question for the subscriptions portfolio is whether GoodRx's pharmacy-network and price-comparison DNA converts into condition-sub retention the way the verticals' brand DNA does.

1. The subscription's biggest competitor is GoodRx's own free product, and the paywall doesn't do the math that would beat it

What I saw. Every GoodRx price search returns a free coupon, and since the pivot, the same surfaces now carry the Companion pitch, subscriptions became "the primary call to action across our homepage and price pages" this year.2 So the conversion moment is a side-by-side: a coupon price the user can have right now for $0, next to a $14.99/month membership whose headline claim is a list ("250+ medications are FREE, with hundreds more under $10").4 Whether Companion wins that comparison depends entirely on the user's own basket: someone filling two or three generics from the free list monthly saves real money; someone filling one occasional script is better off with the coupon, and GoodRx's own positioning admits this. The page doesn't compute which one the viewer is.

Why it matters. The funnel logically starts at "qualified exposure," and here is what that should mean operationally: an exposure where the viewer's basket makes the membership mathematically better. Generic list-claims convert the already-convinced; the marginal subscriber, the one growth has to win, needs the breakeven shown for their meds at their pharmacy. This matters double because the audience the funnel converts from is shrinking, MACs fell 12% YoY,1 so subscription growth can't keep leaning on raw traffic. Conversion quality per exposure has to carry more of the load each quarter.

2. Activation is a pharmacy-counter event, and the 7-day trial usually ends before it happens

What I saw. Companion's trial is 7 days, cut from Gold's 30.45 But Companion's core value is redeemed at a pharmacy counter, on the member's refill schedule, not the app's. Prescriptions run on 30- and 90-day cycles: a member who just filled their meds last week has no natural reason to stand at a counter during their 7-day trial at all. The membership's "aha" moment, handing over a card and paying $0, is an event the trial window mostly can't guarantee.

Why it matters. This is a structural problem I've dug into before in another subscription teardown, value the trial window can't demonstrate on schedule, and the benchmark data is better than intuition here: RevenueCat's 2026 dataset across 115,000+ apps shows 17–32 day trials converting at a 42.5% median versus 37.4% for 5–9 day trials, and over half of trial cancellations happening on Day 0, before the product has done anything.12 Subscription start, activation, and retention are the three levers here, and activation has a physical-world dependency most subscription products don't have. Whoever owns this funnel needs activation defined as the first fill, measured inside the trial, and engineered for, not assumed.

3. The 50% monthly premium pushes annual commitment, but the ask lands before the product has proven a single fill

What I saw. Companion is $14.99 monthly or $119.88 annually, marketed as "$9.99/mo when paid yearly, save $60."4 That's a 50% premium on monthly billing, a strong push toward annual, and notably the annual effective price is exactly old Gold's monthly price, which makes the increase read gentler than it is. The annual option is presented at signup, next to a 7-day trial, to a buyer who hasn't yet redeemed a single fill.

Why it matters. Annual commitment is the right retention architecture for this product, it survives the low-engagement months between refills that kill monthly subscriptions. But asking for $119.88 upfront from a price-sensitive healthcare consumer, before any demonstrated value, is the hardest possible sequencing of that ask. The behavioral literature and my own experience both say the same thing: commitment asks convert best immediately after a value moment, not before it. And there's a signal worth reading in the disclosed numbers: subscription revenue growing 39% while plans grow 14% means ARPU is rising,12 some mix of annual adoption, the Companion price increase, and higher-priced condition subs, and which of those it is changes what the priority should be next. The billing-mix data answers it.

4. Gold's family plan died in the migration, and nothing replaced it, but the buyer is often a household

What I saw. Gold sold a family tier: $19.99/month covering up to five people (and pets). Companion has no family option, it's an individual membership, full stop.45 A Gold family of four that migrated faces either one membership covering one person, or four memberships at $59.96/month for what used to cost $19.99. Meanwhile the monthly price rose 50%, the trial shrank from 30 days to 7, and legacy members are typically ineligible to re-trial.5

Why it matters. The heaviest prescription users, the exact people for whom "250+ free generics" clears breakeven fastest, skew toward households managing multiple chronic conditions: parents, older adults, caregivers filling for a spouse or parent. Willingness to pay in this category is household-shaped rather than individual-shaped, and Companion's individual-only structure prices the best segment out of its best offer. There's also a nearer-term risk: the Gold-to-Companion migration is the single largest retention event in this subscription base's history, a simultaneous price increase, benefit restructure, and family-plan removal applied to the existing payer base. Whether that cohort held is the first thing I'd want to know in the job, and the retention question starts there.

5. Four subscriptions, no visible ladder between them

What I saw. GoodRx now sells at least four consumer subscriptions: Companion ($14.99/mo), GoodRx for Weight Loss ($39/mo introductory, $119/mo standard, meds separate), and Hair Loss (from $16/mo) and ED subscriptions through GoodRx Care.378 Publicly, they're presented as separate products with separate prices, separate signup flows, and no visible bridge: a Weight Loss subscriber doesn't get Companion's free-generics list for their other medications, and a Companion member pays the same condition-sub prices as anyone else. Nothing on the public surfaces suggests holding two is cheaper than holding each.

Why it matters. Condition-sub members are the highest-intent, highest-ARPU users in the entire base, a Weight Loss member at standard pricing pays 8x a Companion member, and GLP-1 users are disproportionately likely to be managing other medications, i.e. to be exactly Companion's best-fit buyer. That overlap is a cross-sell surface with essentially no acquisition cost, the kind of "member value" architecture worth ending on. It cuts the other way too: Companion's $19 telehealth visits are a natural on-ramp into condition subs. Portfolio architecture, what's a tier, what's an add-on, what bundles, is a decision I've owned before at DIRECTV, where curating an unmanaged add-on catalog down to four with explicit tier rules protected roughly $250M/year in add-on revenue, and the shape of the problem here is the same even if the scale isn't.

6. Trust is the constraint every growth mechanic has to run inside, and here it's enforceable, not aspirational

What I saw. Three facts frame every growth decision at GoodRx. "GoodRx is NOT insurance" appears as a standing disclaimer across the membership surfaces.4 The FTC's 2023 enforcement action against GoodRx, the first ever under the Health Breach Notification Rule, with a $1.5M penalty and a permanent ban on sharing health data for advertising, is public record.11 And subscription cancellation sits in an active regulatory environment: click-to-cancel finalized, vacated on procedural grounds, rulemaking restarted in 2026, with ROSCA enforcement and state auto-renewal laws running the whole time.10

Why it matters. The standard subscriptions-growth toolkit leans on behavioral retargeting, lookalike audiences off user data, urgency mechanics, and friction-shaped cancel flows. At GoodRx, the first two are legally constrained by a consent order, and the last two are exactly what ROSCA-era enforcement targets, in a category where the "purchase" is somebody's blood-pressure medication. That's not a limitation to work around, it's the design brief: the growth levers that remain, first-party personalization the user explicitly provides (Finding 1's breakeven math is built from a search the user typed), value made visible (Finding 3's savings receipts), and honest commitment sequencing, happen to also be the durable ones. Balancing growth with healthcare and privacy considerations is the real constraint; this is what that looks like in practice.

What's already working

Five things worth protecting exactly as they are, because a new growth PM's fastest way to lose a room is proposing to fix what's currently carrying the strategy.

• ARPU is rising alongside plan growth, not instead of it. Subscription revenue up 39% on plans up 14% means the growth isn't being bought with discounts, the CFO explicitly flagged revenue "outpacing the number of plans."2 Whatever mix of pricing and portfolio decisions produced that deserves understanding before anything upstream of it changes.
• Leadership already made the pivot call. Subscriptions as the primary CTA, marketing dollars redirected, cannibalization publicly framed as "a positive evolution."2 Most growth PMs spend their first year fighting for exactly this executive alignment; here it's the stated strategy, which makes shipping the higher-leverage move here, rather than persuading.
• Companion's benefit design attacks the free-card weakness directly. A defined free-med list plus telehealth, dental, vision, and labs is value a coupon structurally can't replicate, that's the right architecture against a free floor, even if the paywall doesn't personalize it yet (Finding 1 is about the presentation, not the design).
• Pharma Direct is a real second engine, and it feeds the subscriptions. 76% YoY growth to 31% of revenue,1 and the manufacturer relationships behind it are what make offers like $199 Ozempic possible inside GoodRx for Weight Loss.7 The subscription story gets stronger, not weaker, as this scales.
• The condition-sub machine is shipping fast. Weight Loss, ED, Hair Loss, and the Medicare GLP-1 bridge all launched inside roughly a year.278 Whatever product-and-clinical operating rhythm makes that cadence possible is an asset; Finding 5 is about connecting these products, not slowing their cadence.

How I'd measure it

North star. Retained subscription revenue: subscription revenue that survives its first renewal, rather than subscription starts. Starts are the easy headline number and the easiest one to inflate with a leaky trial; revenue that renews is the version of the same number that can't be gamed by conversion tricks that churn out 60 days later. It also matches how the company itself is starting to talk, the CFO's "revenue outpacing plans" framing is an ARPU-and-durability story, not a volume story.2

The guardrails I'd hold myself to. Two, tracked as their own lines and never blended. First, contribution cannibalization: every Companion convert stops generating coupon-transaction economics, so each cohort carries a net-revenue check, not just a subscription-revenue win. Second, the trust set from Finding 6: cancellation-flow completion, complaint and refund volume, support contacts per 1,000 members. A subscriptions business that has already met the FTC once doesn't get to learn this the hard way twice.

The funnel I'd instrument. Qualified exposure (basket clears breakeven, per Finding 1) → membership pitch view → trial start → activation, defined as first fill inside the trial (Finding 2) → trial-to-paid conversion → first renewal → 90-day retained revenue, with the annual-upgrade event (Finding 3) tracked as its own post-conversion stage. Plus the two portfolio edges: condition-sub cross-sell rate (Finding 5) and win-back conversion on the lapsed Gold cohort (Finding 4). The stage most subscription funnels under-instrument is the one that matters most here: the fill event happens at a pharmacy counter, and the instrumentation has to bridge app behavior and claims data to see it.

Where I'd start

1. 1 Instrument the two definitions everything else depends on: qualified exposure and activation. A measurement task, not a redesign, and it converts Findings 1 and 2 from hypotheses into a dashboard within weeks. If these are already defined internally, even better, the first move becomes reading them.
2. 2 Pull the Gold-to-Companion migration cohort. One query answers the biggest open retention question (Finding 4), sizes the family-plan hole with real churn data, and produces the win-back target list as a side effect.
3. 3 Ask for the experiment archive, then run the breakeven paywall as the first test. The trial-length economics (Finding 2), the annual-mix data (Finding 3), and any past paywall personalization tests may already answer half of this teardown. Whatever's left, the personalized breakeven pitch is the highest-leverage first experiment: it uses data the user already provided, it's privacy-clean, and it sharpens the exact moment the whole strategy turns on.

Ideas to test

Seven concrete, testable ideas, each grounded in something observable from outside plus the lever it maps to. None require internal data to propose; all need it to run.

1. 1 Breakeven paywall. Compute the member price for the exact drug/dose/pharmacy the user just searched and show the delta against the coupon on the same screen ("$0 with Companion, you'd save $23/month"). Hypothesis: trial starts rise among breakeven-positive users while unqualified starts (and their inevitable churn) fall (Finding 1).
2. 2 Counter-moment follow-up. After a free-coupon user redeems a fill, GoodRx knows what they paid; follow up with "with Companion, this fill would have been $0, that's ~$X/year on this prescription alone." The highest-intent conversion moment in the product is thirty seconds after a coupon just worked. Hypothesis: this converts meaningfully above any browse-time paywall (Finding 1).
3. 3 Fill-triggered trial clock. Anchor the trial to the first fill rather than signup (or auto-extend for trialers whose refill date falls just past day 7), with per-trialer fill caps as the abuse guardrail. Hypothesis: conversion conditioned on an in-trial fill is dramatically higher, and this design buys that condition for more trialers without the blanket-30-day abuse exposure (Finding 2).
4. 4 Savings receipt. A monthly "Companion saved you $X" recap, in-app and email, itemized per fill and visit, with the cumulative number surfaced at renewal. Makes invisible value visible in a product whose benefit is the absence of a charge. Hypothesis: recap viewers renew and upgrade to annual at meaningfully higher rates (Findings 2 and 3; the same restate-the-value mechanic as my Allianz "$100,000 medical bill" onboarding iteration, +3% conversion).
5. 5 Post-value annual upgrade. Trigger the $9.99/month annual offer immediately after the first documented savings moment, anchored to the member's own realized number, instead of (or in addition to) the signup-time pitch. Hypothesis: post-value timing lifts annual adoption without discounting deeper (Finding 3; the DIRECTV trial-timing mechanic, +15%, applied to commitment instead of acquisition).
6. 6 Household painted door. An "Add a family member" option shown to current members at test price points, measuring demand before any shared-billing engineering. Run the lapsed-Gold-family win-back as a parallel cell with its own offer. Hypothesis: household demand exists at a price that beats two individual memberships on retention-adjusted revenue (Finding 4).
7. 7 Condition-sub cross-sell placement. At the post-consult moment in Weight Loss / ED / Hair Loss flows, show the member's other medications priced under Companion ("your other 3 prescriptions: $0"). Hypothesis: condition-sub members attach Companion at high rates and the multi-sub cohort retains better, the data that decides whether bundle pricing is worth opening at all (Finding 5).

A note on scope

This was built entirely from GoodRx's public membership and care pages, its Q2 2026 earnings release and call transcript, its own press releases (Companion, Weight Loss, Hair Loss), third-party documentation of the Gold-to-Companion transition, public FTC and regulatory records, and cross-category benchmarks from RevenueCat's State of Subscription Apps 2026, which are priors about the subscription market, not measurements of GoodRx. I have not yet done a live screen-by-screen walk of the Companion signup, checkout, trial, or cancellation flows, or the condition-sub consult funnels, that's the natural next step, in the same format as my JamLoop and Net32 teardowns, and Findings 1 through 3 would all sharpen or change with those captures. No internal churn, cohort, billing-mix, entitlement, or experimentation data was available at any point. Every finding, priority, and metric above is a hypothesis built from outside evidence, not a confirmed diagnosis, and each one's first move in the job is a question or a query, not a redesign.

Notes