Telehealth analytics: attribution, LTV and 6 core metrics

Telehealth brands often spend on Google, Meta, and TikTok with no reliable way to answer two basic questions. Which channel produces patients who stay the longest? And what is the actual revenue per patient by source?

Without attribution (the ability to trace a paying, retained patient back to the ad, campaign, and channel that acquired them), every marketing dollar is a guess. You might be pouring money into a channel with high checkout conversion and terrible 90-day retention while starving one that brings lower volume and much better LTV.

Why telehealth attribution is harder than e-commerce

In standard e-commerce the conversion happens at checkout, and the analytics story is short: click, purchase, revenue. In telehealth the journey is longer, and the value arrives over months or years of subscription payments.

• The conversion is a subscription start, and its true value isn't known until the patient churns
• Checkout may span several sessions, with the ad click on Monday, intake on Wednesday, and payment on Friday
• HIPAA limits which patient data can be passed to advertising platforms for optimization
• The most important metric, LTV by source, takes months to calculate, while ad spend decisions happen weekly

The 6 telehealth analytics metrics to track

1. Patient acquisition cost (CAC) by channel

Total spend on a channel divided by the number of patients acquired from it. This is the most basic metric, and it means little without LTV next to it. A $300 CAC from Meta might be better than a $150 CAC from Google if Meta patients retain twice as long.

In the operator programs behind our CAC benchmarks, GLP-1 programs see CAC of $200-$500+, depending on competition and geography. Other verticals usually cost less to acquire: about $150-$350 for TRT and men's health, $100-$300 for mental health, and $80-$200 for dermatology. For a sustainable telehealth business, the target CAC is whatever produces a 3:1 or better LTV:CAC ratio.

2. LTV:CAC ratio by channel

This ratio decides whether a channel deserves more budget, because it tells you whether the channel produces profitable patients. A ratio of 3:1 or higher indicates healthy unit economics. Below 2:1, you're likely losing money on that channel once cost of care, platform fees, and operations are counted.

Calculate LTV as (average monthly revenue per patient) × (average subscription lifetime in months) × (gross margin percentage), then divide by each channel's CAC. Doing that well needs clean attribution from the ad click through months of subscription payments.

3. Checkout conversion rate by device and source

What share of patients who start your checkout complete payment? Break it down by device (mobile and desktop) and by traffic source, and expect real variation. Mobile checkout usually converts lower than desktop, and organic traffic converts better than paid.

4. Time to first fill

For medication programs, the time from checkout to the patient receiving their first shipment is one of the clearest measures of the experience. Every day of delay is a churn risk. The benchmark for medication programs is under 5 days from checkout to delivery.

5. 30/60/90-day retention by cohort

Cohort retention tells you whether your patient experience is getting better over time. If your February cohort retains better at 90 days than your November cohort, the changes you made in between are working. Track it by acquisition channel and plan type, and split out patients who added an upsell.

6. Revenue per patient (RPP)

Total revenue a patient generates over their lifetime, including subscription payments, add-ons, lab panels, and upsells. Paired with CAC, it gives you the whole unit economics picture. Track it by channel to see which sources bring in the most valuable patients.

Building the attribution pipeline

Clean attribution needs data to flow from the first ad click through months of subscription payments, and each step depends on the one before it.

1. Capture UTM parameters and click IDs when the visitor lands. Save them on your server or in a first-party cookie your server sets. Safari caps cookies set by JavaScript at 24 hours when the visitor arrives from a link carrying a click ID, and session storage clears when the tab closes.
2. Pass the attribution data through checkout. When the patient completes payment, attach the stored data to the patient record in your CRM and billing system.
3. Fire server-side conversion events. Send conversion data back to the ad platforms server side, through Meta's Conversions API and Google's enhanced conversions or offline conversion imports. Health brands should confirm which events each platform accepts, since Meta limits lower-funnel events for some health and wellness advertisers. Server-side events are more reliable than browser-based pixels, especially with ad blockers and Safari ITP in play.
4. Connect billing events to attribution. Every subscription payment, plan change and churn event should trace back to the original acquisition source.
5. Build LTV reports by source. Add up revenue by acquisition source at 30, 60, 90 and 180 days to see how each channel really performs.

HIPAA and analytics: what you can and can't track

HIPAA limits telehealth analytics in ways standard e-commerce never has to deal with. You can't send protected health information (PHI) to advertising platforms or most analytics tools unless a signed BAA and a HIPAA permission cover it, or the patient has given written HIPAA authorization. Properly de-identified data isn't PHI.

• Never send patient names, email addresses, or health conditions to ad platforms directly
• Hashing doesn't de-identify an email or phone number, so a conversion event that carries a hashed identifier from a patient checkout is still PHI. Send one to an ad platform only under a signed BAA or the patient's written HIPAA authorization, and confirm the platform offers a BAA at all (Google doesn't for Google Analytics)
• Any analytics tool that processes PHI, such as a CRM that stores health data alongside analytics, needs a BAA
• Google Analytics and similar tools should receive only de-identified, non-PHI events
• For patient-level data, consider HIPAA-eligible analytics platforms. PostHog and Segment offer HIPAA-eligible plans with BAAs available, though configuring them properly takes real work

Where to start

Operators who know their LTV by channel, their retention by cohort, and their true CAC after churn can tell which channels to fund. Without those numbers, budget decisions rest on checkout counts from the ad dashboards.

Build attribution into your checkout and portal from day one, starting with UTM capture and server-side conversion events. Calculate LTV for each source as well as the overall average. Keep PHI out of ad platforms and general analytics tools, and do patient-level analysis in a tool covered by a BAA.