These are evidence, not decoration
A results image next to a service description is not illustration. It is support for a claim about what the service achieves, and it is assessed on that basis. The same is true of a testimonial describing an outcome and of a star rating displayed prominently.
That reframing is the single most useful thing a business can take from this piece. Once these assets are understood as evidence, the requirements follow naturally: they have to be genuine, they have to be comparable, and they must not imply more than can be supported.
If an image is doing the work of a claim, it is held to the standard of a claim.
Reviews: genuineness is the whole issue
Consumer reviews are influential precisely because they are believed to be independent. Practices that undermine that independence are treated seriously, and the law in this area has been strengthened, with the Digital Markets, Competition and Consumers Act 2024 addressing fake and misleading reviews.
Practices to avoid entirely:
- Writing reviews, or having staff, friends or family write them.
- Commissioning reviews from people who have not used the service.
- Filtering so that only satisfied clients are invited to review.
- Suppressing negative reviews you have gathered, or publishing a selective subset without saying so.
- Incentives that are conditional on a positive review, or that are not disclosed.
- Presenting an average that is not what it appears to be.
What is legitimate: asking every client, at the same point, in the same way, without conditions attached to the content. Consistency is what makes the request fair, and it also produces a more representative picture, which is more useful to you.
| Practice | Position |
|---|---|
| Asking every client, at the same point | Fine, and produces a representative picture |
| Asking only clients you know were pleased | Not acceptable, it is selective solicitation |
| Offering an incentive conditional on a positive review | Not acceptable |
| Offering the same small incentive to everyone, disclosed | Care needed, disclose it and never condition it on content |
| Publishing a selected subset without saying so | Not acceptable |
| Writing or commissioning reviews | Not acceptable, and addressed by legislation |
| Replying to a negative review, factually | Fine, but disclose nothing about the client |
Source: Framework is this paper's own; obligations arise under consumer protection law, including the Digital Markets, Competition and Consumers Act 2024.
Not legal advice. The law on fake and misleading reviews has been strengthened and the area is actively enforced.
Responding to reviews, including bad ones
Two considerations, one commercial and one about data.
Commercially, a measured reply to a critical review is read by future clients as evidence of how you handle a problem, which is more informative than the complaint itself. Reply once, factually, without arguing.
On data, be careful not to disclose anything about the client. Confirming that someone was a client, or referring to their treatment, discloses personal information and potentially health information, in public, without their agreement. Keep replies general and move the specifics to a private channel. The wider framework is in client data in a beauty business.
Testimonials: what they can and cannot carry
A published testimonial is a claim you are making. Three practical requirements.
It must be genuine and you should hold evidence that it is. Keep the original and a record of permission to publish it.
Claims within it are yours. If a testimonial says a treatment removed a condition, you have made that claim. Editing to remove such statements is preferable to publishing them, and declining to publish is sometimes the right answer.
Typicality matters. Where a testimonial describes an unusually good outcome and is presented in a way that implies a normal result, that implication needs support. A qualifier does not fix a headline claim that is stronger than the evidence.
Results imagery: conditions matter as much as editing
Most misleading results images are not retouched. They are shot differently.
Differences in lighting, angle, distance, lens, posture, expression, makeup and even the time of day can produce an apparent difference that the treatment did not create. An image pair that varies in these ways overstates the result whether or not anyone intended it.
A fixed capture protocol solves this, costs nothing and improves your own clinical or practical record at the same time.
| Variable | Rule |
|---|---|
| Lighting | Same source, same position, room lights off |
| Distance and lens | Marked position, same focal length |
| Angle | Fixed, recorded, repeated |
| Posture and expression | Neutral, matched between frames |
| Makeup and product | Same state in both images |
| Interval | Recorded, and stated when published |
| Processing | Identical to both frames, nothing that changes the result |
Source: Working model used by this paper, not a measurement.
A protocol improves your own records as well as your marketing, and it is far easier to follow than to reconstruct.
On editing, the distinction drawn in art direction and photography applies: consistent exposure and colour handling applied identically to both images is presentation; anything that changes what the image appears to show about the result is evidence handling and is not acceptable in this context. Filters, smoothing and platform beautification effects are all capable of doing exactly that, sometimes without the user realising.
Consent for results imagery
An image of an identifiable client used in marketing is personal data, and where it shows a condition it is health information. The requirements are: a valid basis for this specific use, information given clearly, a record of what was agreed, and a workable route to withdraw that you can execute across every place the image has been published.
Two practical points that catch businesses out. Consent obtained for a website gallery does not cover a video posted to a platform. And consent can be withdrawn, which means you need to know where every image is published, including where content has been shared onward.
Platform rules are a separate constraint
Advertising platforms and social networks maintain their own policies about results imagery in this category, and those policies can be stricter than the law. A compliant image can still be rejected or an account restricted under platform rules.
Treat platform policy as an additional constraint rather than as guidance about legal requirements. The two are not the same, and satisfying one does not satisfy the other.
A workable process
For reviews: ask everyone, at the same point, in the same way, with no conditions. Publish what you receive. Reply factually and never disclose client information.
For testimonials: keep the original, hold permission, remove claims you cannot support, and do not present exceptional outcomes as typical.
For images: use a fixed protocol, record the interval and the conditions, apply identical processing to both frames, hold specific consent, and keep a list of where each image is published.
None of this is onerous once it is a routine. All of it is very difficult to reconstruct after a complaint, which is the reason to build it in from the start.
