Ecommerce OS · Telehealth7 min readUpdated November 2, 2026

Ad Attribution for Telehealth Companies

How a telehealth brand measures which ads bring approved, retained patients: MER, blended CAC, surveys and holdout tests, without health data.

Nick Bonnassies, CMO at Tagada · checked against 10 sources

Article summary

TL;DR

Don't judge telehealth ads on the purchases Meta, Google or TikTok report. A checkout is not a patient: some are declined, some never refill. Measure in your own data (approved patients, captured revenue, second refill), watch MER and blended CAC, test big channels with holdouts, and send the platforms only a generic paid event.

Meta can restrict health and wellness data sources, up to blocking events such as Purchase, and this applies to the Conversions API too (Stape, March 2026).
Meta's Business Tools Terms (effective November 3, 2025) ask advertisers not to share data based on health information, so clinical milestones stay out of ad events.
A platform credits the order when it starts, even if approval or fulfillment later fails, as Bask Health notes (February 2026).
No published source gives a telehealth CAC or ROAS benchmark with a stated method (checked November 2, 2026): build your own baseline.
MER (total revenue ÷ total ad spend) doesn't depend on which platform claims a sale.

Want to go further? Ask

On this page

This guide is part of our playbook on the best ecommerce platform for telehealth companies. The general guide, for any business: attribution modeling. Clean purchase data comes first: server-side tracking for telehealth companies.

Why attribution breaks in telehealth

A checkout is not a patient
Between the payment and the first shipment, a clinician can decline the case. In a hold-then-capture model, the card hold is then released and no revenue exists.
The value is in refills
A patient who stops after one month and one who stays a year look the same on the day of the ad click.
Meta sees less
Health and wellness restrictions can strip details or block Purchase events1.
You shouldn't send it more
Approvals and prescriptions are medical facts about a person; they don't belong in ad events2.
Clinical delay
Review and shipping take days, longer than many default attribution windows.

Measure the patient, not the checkout

Follow each new patient through five steps. The ad platforms see the first one at best; you see all five.

Step	What happened	Who sees it	Use it for
1. Checkout	Payment or card hold	Ad platforms (generic event) and you	Creative and campaign tests inside a platform
2. Capture	Card charged after approval	You	Real first-order revenue
3. Approval rate	Approved ÷ submitted cases	You	Whether a channel brings eligible patients
4. First shipment	Pharmacy ships	You	Cohort start
5. Second refill	Rebill succeeds	You	Early sign of retention

Fig. 1The platform sees one step, you see five

What the ad platform sees

1. CheckoutGeneric paid event

The checkout, at best

What you see

1. CheckoutPayment or card hold
then2. CaptureCard charged after approval
then3. Approval rateApproved ÷ submitted cases
then4. First shipmentCohort start
then5. Second refillRebill succeedsApproved

Real revenue, audience fit and retention

Steps 2 to 5 stay in your own data.

From the table above.

Example

Meta reports 100 purchases in a week. Your data shows 100 checkouts, 82 approved and captured, 80 shipped, and 52 second refills two months later. Judge the campaign on the 82 and, later, the 52, not the 100.

Fig. 2100 purchases for Meta, 52 patients still there two months later

Checkouts (what Meta reports)100 patients
Approved and captured82 patients
Shipped80 patients
Second refill, two months later52 patients

Source: Illustrative example from the note above

Judge on the 82, then the 52.

Illustrative example.

If a channel brings many checkouts but a low approval rate, its audience doesn't fit your clinical criteria. Fix the targeting or the landing page, not the bid.

The numbers that matter

Number	Formula	What it tells you
Captured revenue	Revenue actually charged, by day	The truth everything else is checked against
MER	Total revenue ÷ total ad spend	Whether the whole system is efficient
Blended CAC	Total acquisition spend ÷ new approved patients	What one real patient costs
Approval rate by channel	Approved ÷ submitted, per source	Audience fit
Second-refill rate by channel	Patients reaching refill 2 ÷ approved, per source	Retention by source
Platform-reported ROAS	As reported	Relative performance inside one platform only

Payback targets and LTV math are in LTV and CAC for a telehealth company.

Watch out

A platform ROAS that includes refills charged months later credits this month's ads for last quarter's patients. Separate first orders from rebills before you judge any campaign.

What the ad platforms should and shouldn't receive

Send	Keep in your own data
One generic paid event, value and currency	Approval, decline, prescription, shipment
Event ID shared with any browser event, for deduplication3	Treatment or product names that describe a condition
Hashed contact data where consented	Intake answers and survey answers
Click IDs captured on the landing page	Refill and cancellation events

Optimization will be weaker than in a store that sends everything. That is the cost of a regulated category: GoodRx paid a $1.5 million civil penalty in 2023 over health data shared with ad platforms5. Make up for it with better measurement on your side.

Ask new patients: the post-purchase survey

A one-question survey after checkout catches what tracking misses: creators, podcasts, word of mouth.

Ask one question: "Where did you first hear about us?"
Offer your channels as a short list, plus "Other".
Ask new patients only, once.
Never ask a health question in the survey, and keep the answers out of your ad platforms.
Compare with platform claims each month: if Meta claims 70% of new patients and patients name Meta 30% of the time, investigate.

Incrementality: test with the clinical delay in mind

A holdout test turns a channel off for part of the audience or a group of regions and compares results.

Pick one channel with enough spend, often Meta prospecting or branded search.
Hold out a share of the audience, or a group of states you serve.
Run long enough to cover the review and shipping delay, plus a few weeks.
Compare approved new patients, not checkouts, between the two groups4.
Discount or credit that channel's platform numbers by the result.

Compare approved new patients, not checkouts, and only hold out states your clinicians can treat.

Only hold out states where your clinicians can treat patients. State coverage is covered in is it legal to sell telehealth services in my state?.

Monthly routine

Step	What you do
1	Reconcile platform purchases with checkouts and captured payments
2	Split revenue into first orders and rebills
3	Compute MER, blended CAC and approval rate by channel
4	Read survey answers from new patients
5	Update second-refill rates for cohorts two months old
6	Plan one holdout test per quarter on your largest channel

Ad policies that shape what you can run are in Meta and TikTok ads for telehealth companies.

How Tagada helps

Attribution that ends at the paid order

Tagada keeps checkout, orders and tracking together, with payments on TagadaPay and the processors you already use: "Already on Stripe, NMI or Airwallex? We orchestrate those too."10

On the Tagada checkout, in the docs:

Server-side events to Meta (Conversions API) on paid orders, deduplicated with the browser event8
Click IDs captured at checkout (gclid, fbclid, ttclid and others) into the customer record, and server-side postbacks to trackers such as Voluum, RedTrack or Hyros8
Funnel conversion rate reported per funnel: the share of sessions that led to a paid first order over the last 30 days9

For a telehealth brand, check what each connected tool receives before you switch it on.

Building the whole telehealth brand? With Tagada Rx, "The card is only captured once a doctor approves", so captured revenue is approved revenue6, and a case timeline records each status up to delivery7. US patients only, by selection.

Start free

Updated on November 2, 2026

Nick Bonnassies

CMO at Tagada

CMO at Tagada. Before Tagada, Nick built direct-to-consumer brands that generated over $115M in sales, raised over $25M in crowdfunding and sold two of them. He also works as a fractional CMO on marketing operations and AI systems.

Frequently asked questions

How do I measure ad performance for a telehealth brand?

In your own data: captured revenue, approved patients, approval rate and second-refill rate by channel. Use MER and blended CAC to judge the whole budget, a post-purchase survey for untracked channels, and holdout tests for the biggest ones. Use platform ROAS only to compare campaigns inside one platform.

Why does Meta report more purchases than I have patients?

Meta counts the checkout, and some checkouts are later declined or refunded. It may also claim sales that other channels drove. Meta's health restrictions can hide events too. Reconcile its purchases with captured payments each month.

Can I send approved prescriptions to Meta as a conversion?

Don't. An approval or prescription is a medical fact about a person, and Meta's Business Tools Terms ask advertisers not to share health data. Send one generic paid event and keep clinical steps in your own reporting.

What is a good CAC for a telehealth company?

No published benchmark gives a telehealth CAC with a stated method (checked November 2, 2026). Compute blended CAC on approved patients, compare it with the margin a patient brings by the second or third refill, and track it month by month.

What is the difference between MER and ROAS?

ROAS is the revenue a platform attributes to itself divided by its spend. MER is your total revenue divided by total ad spend. MER can't be inflated by overlapping platform claims, which makes it the safer check for a telehealth brand.

Sources

Checked November 2, 2026

Tagada

Judge campaigns on approved patients, not checkouts

Funnels, one-click upsells, subscription billing and payment routing, on one customer record.

Start free See pricing

01Create your free account
02Connect your store or build a checkout
03Launch funnels, upsells and subscriptions