How to scale Meta ads for a GLP-1 telehealth brand

To scale Meta ads for a GLP-1 telehealth brand, raise budget in steps only after an ad set leaves the learning phase (about 50 results a week). Combine similar ad sets, broaden to adult audiences with no health signals, optimize for the deepest event Meta still receives, and upload creative that was checked against Meta’s health rules first.

Meta publishes no budget percentages. The ones in most guides come from agencies, and this page labels them that way. Each lever below is matched to Meta’s own wording. Policy checked on October 27, 2026.

Why GLP-1 is harder to scale on Meta than most categories

A GLP-1 brand scales slower on Meta because four health limits sit on top of the normal auction, and each one takes away a lever that other advertisers use freely:

• Prescription drug authorization: Rx ads need Meta’s authorization, and Meta says “you must maintain an active LegitScript Certification” to keep it.
• An age floor: Meta restricts weight-loss ads “from targeting users under 18 years of age.”
• Restricted events: A data source in a restricted category can lose “specific mid and lower funnel events.”
• Flagged audiences: Custom or lookalike audiences that reference a health condition “will be flagged and cannot be used for running ad campaigns.”

When a GLP-1 ad set is ready to scale on Meta

A GLP-1 ad set is ready to scale when it has left the learning phase and its cost per result has held at your target for about a week. Meta says ad sets usually exit learning “after about 50 results in the week after the ad set’s last significant edit.”

Check four things before you touch the budget:

• The Delivery column reads Active. “Learning” and “Learning limited” both mean wait.
• Cost per result has stayed at or under your target for 7 days.
• The Last significant edit column, which Meta says “shows when you last made a change that could reset the learning phase”, is more than a week old.
• Nobody plans another edit this week. Meta warns that editing during learning means “you reset learning and delay our delivery system’s ability to optimize.”

Meta says Learning limited “isn’t a penalty.” One cause it lists is “an infrequent optimization event”, and restricted lower-funnel events make that likelier in a GLP-1 account.

Meta’s scaling levers in a GLP-1 account

Every Meta scaling lever still works for a GLP-1 brand, each with its own health limit:

Vertical scaling: how much to raise a Meta budget

Vertical scaling means putting more budget into ad sets that already work, and Meta gives no percentage for how big each raise should be. Its guidance stays qualitative: “Use realistic budgets”, avoid “frequent budget changes”, and expect that “Your cost per result may increase because your budget is significantly higher than usual.”

The percentages in most guides come from agencies, and your own CPA can overrule them:

• CausalFunnel (February 23, 2026): “Increase budgets by 10–20 percent” and “Wait 48–72 hours before adjusting again.”
• HOC Digital Solutions (April 24, 2025): “Raise the budget by around 20% every 48–72 hours, as long as CPA or ROAS holds.” A new budget takes about 15 minutes to apply, and Meta recommends that you “wait 15 minutes after changing your budget to change your bid.”

Horizontal scaling inside Meta’s health limits

Horizontal scaling means adding audiences, ad sets or campaigns instead of raising the budget on one ad set. In a GLP-1 account every new audience has to stay 18+ and carry no reference to a health condition. The audiences Meta still allows after its targeting cuts, and how Advantage+ audience treats age and location, are covered in Meta targeting for GLP-1 ads.

Flagged custom or lookalike audiences can’t be used in new campaigns. For a flagged lookalike, Meta says you “will need to resolve the issue with the underlying custom audience.” Build every audience and seed list from data with no health-condition information in it. Meta also points to “Advantage+ detailed targeting” to widen a detailed or lookalike audience.

Split carefully, because Meta lists “running too many ads at the same time” among the causes of Learning limited. Meta’s advice points toward a few broad ad sets under one campaign budget. It says that with Advantage+ campaign budget “it’s important to analyze results at the campaign level, rather than at the ad set level.”

For its automated Advantage+ sales campaigns, Meta claims businesses see “a 20% lower cost per result on average.” Its footnote says the figure comes from “an experiment run globally across verticals during Sep 9, 2025 and Sep 14, 2025 on 10,107 campaigns using the Sales objective,” and calls it a result meant “to give you an idea of possible performance.” It doesn’t show how a health advertiser with restricted lower-funnel events performs, so run your own test against your target before you move budget.

Which event to optimize for when Meta restricts health data

Optimize for the deepest conversion event Meta still receives that also reaches about 50 results a week per ad set. For a restricted category, Meta’s own advice is to “explore alternative events” and to consider “more upper-funnel events to maintain effective audience targeting and campaign optimization.”

Work it out in Events Manager:

1. Check which data source category Meta assigned and which standard events are restricted.
2. Of the events that still arrive, pick the deepest one that reaches about 50 a week per ad set at your planned budget.
3. Judge results on paid orders in your own system. The optimization event only steers delivery.

Custom conversions follow the same rule as audiences. A custom conversion that suggests a health condition gets flagged, and Meta warns: “Once the campaign is published, you cannot remove or select a different custom conversion.” Under full restrictions, Meta points to objectives that don’t rely on shared events: Awareness, Engagements and Traffic. For the Events Manager side, read what Meta’s health data limits do to telehealth tracking.

Creative volume on Meta without review problems

More budget on Meta needs more ads, and each new GLP-1 ad is checked against Meta’s health rules. Meta also calls new creative testing “essential.”

If you keep posting ads that break policy, Meta may treat “all ads from your Page, domain, ad account or other associated entities as lower quality.” Check each concept against the age, body-image and claims rules in our guide to Meta’s weight-loss ad policy before it’s uploaded, and add new ads in small batches.

How the GLP1.partners media team scales GLP-1 brands on Meta

The GLP1.partners media team runs Meta for GLP-1 brands on established agency accounts built for GLP-1 scale, with a dedicated Meta agency contact. In our experience, GLP-1 brands lose scale at the account level more often than at the ad level, so the account setup matters from the first campaign.

• The target is yours. You choose the number, such as cost per paid first order. Our weekly report shows spend and results against it, so you can see whether each step up paid off.
• Same Meta rules on our accounts. For prescription drug ads, the brand brings its own LegitScript certification, and our agency accounts are set up to run certified telehealth brands. Every lever and limit on this page applies to the campaigns we run.
• More channels after Meta. When Meta volume levels off at your target, we add Outbrain and TikTok, our own comparison sites and our affiliate network, each judged on the same target. The multi-channel plan is in our guide to scaling past the first channel.

So far that’s $38M in GLP-1 media managed and 27 GLP-1 brands scaled. How the accounts work is on our whitelisted Meta agency page.

A step-by-step plan to scale GLP-1 spend on Meta

You can run this plan on your own accounts.

1. Write down your target, such as cost per paid first order, and the order data you’ll judge it on.
2. In Events Manager, note which events are restricted, then choose the deepest one that still arrives and can reach about 50 results a week per ad set.
3. Merge similar ad sets. Try Advantage+ campaign budget and read results at campaign level.
4. Wait until each ad set shows Active and has held your target for 7 days.
5. Raise budget in steps. A common practitioner default is 10 to 20% every 48 to 72 hours. Change budget and bid at least 15 minutes apart, and stop when cost per result leaves your target.
6. Broaden next: 18+ audiences built without health-condition data, plus Advantage+ detailed targeting. Add a new ad set only for a new audience or angle.
7. Feed creative in small batches, each concept checked against Meta’s health rules before upload.
8. Review weekly against your own orders. When Meta stops adding volume at your target, plan the next channel.

FAQ

How long is the learning phase for Meta ads?

Meta gives no fixed number of days. Ad sets leave learning “as soon as they can deliver stably”, usually after about 50 results in the week after the last significant edit. Another significant edit restarts it.

What is horizontal and vertical scaling in Facebook ads?

Vertical scaling raises the budget on ad sets that already work. Horizontal scaling adds audiences, ad sets or campaigns.

How much should you increase a Facebook ad budget?

Meta publishes no percentage. Agencies often use 10 to 20% every 48 to 72 hours, a practitioner rule to test against your own cost per result.