Before and after photos: the rules and the risks

The compliance work here is not paperwork around the edges. A gallery published without proper consent does not stop being a problem when you take it down, because the images were already indexed and shared.

I am not a lawyer and this is not legal advice. Your consent form should be written or reviewed by your attorney, and your state board's advertising rules are the ones that govern you.

What does the consent actually have to cover?

Marketing use, specifically and in writing, signed before you shoot.

A general treatment consent or a clinical photography consent is not the same thing. Photos taken for the record are one purpose. Publishing them to attract new patients is another, and the patient has to agree to the second one knowingly.

The elements to work through with your attorney usually include what will be shown, where it may appear, how long the permission lasts, whether the patient is identifiable, whether their name or any detail appears alongside, and how they withdraw permission later.

Verbal consent is worth nothing the day it is disputed, and it will only ever be disputed on a bad day.

The enforcement record shows what the absence of a written policy costs. In the settlement the HHS Office for Civil Rights announced with New Vision Dental, a California practice, in December 2022, OCR found three separate failures: impermissible disclosure of protected health information on a public platform, a Notice of Privacy Practices missing required content, and no implemented policies or procedures at all concerning the release of protected health information on social media. The practice paid $23,000 and accepted a two year corrective action plan, and the remedies included taking down every social media post going back to 1 January 2014 and issuing breach notices.

Read that third finding again. The regulator penalized the missing policy, not only the posts. A practice with a shelf full of signed consent forms and no written rule about who may publish a patient image is carrying the same gap.

Withdrawal deserves more thought than it gets. If a patient asks you to take their images down, you need a process that reaches every place they exist: website, social, ad creative, and anything a vendor runs for you. The advice breaks down exactly here, because most practices can produce the consent and cannot produce the inventory. Know where your images live before somebody asks.

Then handle the file itself. Strip location metadata. Check the frame for anything identifying: a name badge, a chart, a reflection, a tattoo. That detail is almost never the part you were looking at.

Stripping the geotag costs you nothing in search terms, whatever a vendor has told you. Presence of geotagged photos scored 23 in Whitespark's 2026 Local Search Ranking Factors survey, in which 47 local search experts scored 187 factors, which makes it the single lowest ranked item on the entire list. It is the most-sold local SEO myth in the category and it is dead last.

One more clause to watch. If you discount treatment in exchange for permission to publish, and that patient also writes you a review or a testimonial, the FTC's Endorsement Guides at 16 CFR section 255.5 treat the discount as a material connection that has to be disclosed clearly and conspicuously, and the rule applies regardless of whether you required the endorsement in return. Keep the photo release and the review ask on separate pieces of paper and separate conversations.

What makes a before and after misleading?

Anything that changes between the two frames other than the treatment.

Different lighting, angle, distance, a wider smile, makeup in one and not the other. Each one makes the result look better than it was, and each one turns a marketing image into an advertising problem.

Shoot both frames the same way. Same camera, same distance, same light, same head position, same expression. If your after photo is prettier because the photography improved, you have not documented a result, you have documented better photography.

Two more lines. Do not publish manufacturer-supplied images as though they are your work, and never present a case as a typical result. Claims about what patients generally get are clinical claims, and they belong to your medical director and your state board.

If a case involved several treatments, say so. A single-treatment claim built on a combination result is the one that gets challenged.

Med spas carry a sharper version of this problem, and it is the part of the category most likely to draw a regulator. Per AmSpa's state by state legal summaries and the analysis published by the law firm Quarles and Brady, every US state requires a good faith examination by a licensed practitioner before treatment, and that examination is the only legitimate mechanism for determining whether a given patient is a candidate. A gallery captioned in a way that anticipates the outcome, "this is what we can do for you," is copy that pre-empts a legally required medical act. Show the case, describe what was done, and let the consultation decide the rest.

Where can I use them, and where can I not?

Your own site is the most permissive place. Advertising platforms are the most restrictive.

Some ad platforms limit or prohibit before and after imagery in paid creative, and the rules differ by platform and change over time. Check the current policy for each platform before you build a campaign around a gallery, because a rejected ad account is a bigger setback than a rejected ad.

Your state board's advertising rules sit on top of all of it, and dental and medical boards do not read the same. That is a question for your attorney or your board directly, not for a marketing article.

Keep them off the Google Business Profile as well, and not for a compliance reason. Whitespark's 2026 survey added a set of new negative and suspension-risk factors, and two of them cover profile media directly: the presence of AI-generated photos or videos on the profile, and keyword stuffing the profile description. A profile is a place for pictures of the building, the room, and the team.

On your own site, put the images where the decision is being made. A gallery page collecting every case together is worth less than a small set of relevant cases on the treatment page itself, next to the price and the booking button. It also makes that page stronger.

That effect matters most for anything new. When you add a procedure, you have no reviews mentioning it and no history, so a documented case is most of your proof, which makes photos part of ranking for a procedure you just started offering.

It matters even more with no storefront. Mobile and traveling providers have no waiting room to photograph, so the work is the only visual evidence there is, and service area businesses rank differently from storefronts.

How do I shoot them so they are actually usable?

Standardize once, then never think about it again.

Pick a spot in the office with consistent light and mark it. Same background, same distance, same camera height, same phone or camera every time. Write the setup on a card and tape it inside the cabinet.

Take the before photo at every case that might qualify. You cannot go back for a before, and your best example is rarely the one you predicted.

Name the files something a human wrote, describing the treatment rather than the patient. Write alt text that describes the procedure and never the person.

Then make sure the inquiries the gallery generates actually land somewhere. Photos drive messages through social, forms, texts, and calls all at once, and practices routinely lose the ones that arrive on the channel nobody watches. Routing leads across four channels without dropping any is the unglamorous half of this.

How much is the gallery costing you in load time?

More than any other element on a treatment page, and the arithmetic takes two minutes.

Google's own analysis, reported by Daniel An and built on a neural network trained on bounce and conversion data with 90% prediction accuracy, states it plainly: as page load time goes from one second to 10 seconds, the probability of a mobile site visitor bouncing increases 123%. A separate Google Research sample of 11 million mobile domains on a 4G connection found 79% of pages were already over 1 MB and 25% could save more than 250 KB by compressing images and text alone.

Run yours.

Count the pairs. Say your gallery holds 14 pairs. That is 28 images.

Weigh them. Straight off a modern camera at 3.2 MB each, 28 images is 89.6 MB of image payload on one page.

Compress. At 120 KB each, the same 28 images come to 3.4 MB. That is a 96% reduction and nobody looking at the page can tell.

Then cut, because compression is not the whole answer. Move the six strongest pairs onto the treatment pages where the decision happens. Twelve images at 120 KB is 1.4 MB.

Then defer. Lazy load everything below the first pair and the phone downloads roughly 240 KB on arrival instead of 89.6 MB.

Check the result against the published Core Web Vitals thresholds rather than against how it feels on your desk: Largest Contentful Paint is rated good at 2.5 seconds or less, and all three vitals are assessed at the 75th percentile of real visits, not the average. Your office wifi is not the 75th percentile.

What to do this week

Pull your current consent form and read it as though you were the patient. If it does not mention marketing, advertising, or the website by name, it is not doing the job. Send it to your attorney this week.

Write the one page policy OCR found missing at New Vision Dental: who may publish a patient image, on which properties, on whose approval, and how a takedown request gets executed across all of them. It is a short document and its absence is what got cited.

Audit what is already published. For every image on your site and social accounts, confirm you can produce a signed consent for it. Anything you cannot, take down today.

Set up the photo station and write the setup card.

Move your best three cases from the gallery page onto the treatment pages they belong to, next to the pricing and booking sections. While you are there, check that the same pages handle the "do you take my insurance" question properly, since cost and coverage are the next two things a convinced visitor asks.

Then decide how you will know whether it worked. Gallery views on their own tell you nothing, which is why the dental marketing metrics worth tracking monthly are booking and consult numbers rather than page views.

Want a read on your gallery?

Email me at eric@seod.com with a link to your before and after page and I will send back three things: where the images should be moved to earn more bookings, which ones are slowing the page down, and where the alt text or file names are doing nothing for you. Takes me about ten minutes.

I will not review your consent form. That is your attorney's job. What I can tell you is whether the gallery sits where it can change a decision.

There is more on dental and med spa marketing in the library.