Before-and-after imagery is the most persuasive asset an aesthetic practice owns and the most regulated. Two rule sets apply and they do not agree with each other.
What the platform says
Meta's health and wellness standards restrict transformation imagery. Side-by-side before/after comparisons for weight loss and for anti-aging and wrinkle treatments are the clearest prohibitions. Close-ups designed to generate negative self-perception — pinching fat, zoomed-in wrinkles — are out. The interpretation has been widening toward implied transformation, where a product shown alongside an idealised body is treated the same way as a literal split image.
The underlying principle is Meta's rule against content that generates negative self-perception to promote a health product.
What the states say
The opposite, more or less. California's B&P § 651 makes it unlawful to use a photograph that does not accurately depict the results of the procedure advertised, or that has been altered in any manner from the actual subject — and requires that a model be labeled as a model, in a prominent location, in easily readable type.
New York and Florida approach the same problem through professional standards and consumer protection law, and arrive somewhere similar: the image must not create a false or unjustified expectation.
The platform restricts whether you may show results. The state governs what the image must be if you do. Satisfying both means moving the results off the ad.
The structure that works
The ad sells the consultation. A clinician on camera, the room, the process, a clear offer. No transformation imagery at all.
The landing page carries the results. Properly consented, unaltered, labeled, with realistic framing. Your state's rules apply here and Meta's do not.
This is not a compromise. It is better advertising. Ads leading with a transformation attract people shopping for a transformation and comparing yours against everyone else's. Ads leading with a practitioner attract people choosing a clinic, and they close better.
If you do publish results, the standards
Consent in writing, specific, revocable. And specific about channel — "you may post this" and "you may run this as a paid ad targeting my neighbors" are different permissions and should be recorded separately.
Fixed conditions. Same camera position, same lighting, same distance, same expression. Two photos of one face under different lighting can show a result that did not occur, which is an image that does not accurately depict the result.
No retouching whatsoever. Including the retouching your phone applies automatically. Turn it off. This is a genuine practical problem and it is worth solving with a dedicated camera and a fixed setup.
Label anything that is not a patient. Stock imagery of a model requires the disclosure, prominently, in readable type. This is the single most common violation we find in California practices.
Say what is typical. An unusual result presented as a normal one is the definition of an unjustified expectation.
If your practice is running results imagery in ads, it is worth checking what the consent covers and whether anything has been retouched. Both are quick and both are the sort of thing that only looks urgent afterwards.