Send twenty finished images from whoever retouches your catalogue now. We read four things across the set and send you the numbers — your frame dimensions, your background values, your crop logic, your edges at full resolution. No pitch, no quote, nothing to install. If your current supplier measures clean, that is what the reply will say.
Most arguments about retouching quality are arguments about taste and cannot be settled. These four can be, on a batch you have already paid for.
| What is measured | What you get back | What it means when it drifts |
|---|---|---|
| Frame dimension | The pixel width and height of every frame you sent, listed | Two or three dimensions in one batch means the crop was decided per image rather than against a specification |
| Background value | Background brightness sampled in the same corner of each frame | A set drifting between 252 and 255 passes frame by frame and reads as grey patchwork on a category page |
| Crop and margin logic | Whether the product occupies the same proportion of the frame and sits on the same baseline | Correct dimensions with inconsistent margins still reads as a jumbled grid |
| Edge treatment | A full-resolution look at the hardest material in what you sent | Clean on the easy frames and soft on the difficult ones is where a cheap cut-out and a drawn path separate |
The first two are arithmetic. The second two need eyes, which is the part you cannot script. All four are things you could check yourself — the method and the command are published on the catalogue consistency standard, so you can verify how we measure before you send anything.
The usual offer in this industry is five free images, and we publish one too. It tests our standard, which is useful only once you have already decided to look for someone else.
An audit tests something more immediately useful: the thing you are paying for now. It uses your own files, returns numbers you can check against your own eyes, and answers a question you may not yet have put into words — why does the category page look patchy when every individual image looked fine on approval?
Three situations where buyers run this and find something:
None of those people are shopping for a retoucher. All of them have a problem worth measuring.
Measurement catches drift. It does not catch taste, and anyone who says otherwise is selling something.
A garment can be cut to exactly the right dimensions, on exactly the right white, with every frame identical — and still have a collar rebuilt in the wrong shape. The four checks tell you whether a supplier works to a standard. They do not tell you whether that standard is any good. Where the measurements come back clean, the reply will say so plainly, and the honest conclusion is that there is no supplier problem to solve.
We also measure ourselves on the same four, and publish the result. Twenty-four finished garments across seven material problems — tailoring, ribbed knitwear, sheer silk, denim, moulded activewear, waxed cotton and lace. Every frame 2600 × 3400. Background brightness 255 on all of them. No variance. Those numbers are on the standard page with the method, so you can hold us to the same test you are asking us to run.
If your own agreement with your current supplier restricts sharing finished files, check it before sending. We would rather you did not send anything than breach a contract to do it.
Each tests a different thing, and they are meant to be used in order rather than all at once.
| Step | What you send | What it tests | Cost |
|---|---|---|---|
| Retouching audit | Up to 20 finished frames from your current supplier | Whether you have a problem at all | Free |
| Free test edit | 5 of your unretouched images | Our standard, on your products | Free, returned in 6 hours |
| Production fit test | 50 to 100 frames of a live job | Whether we hold that standard at volume, inside your pipeline | Published rate, then taken off your first invoice |
Below roughly 500 images a month, a high-volume studio is usually the wrong choice, including this one. The audit is still free and the answer will still be honest.
Because the measurement is the argument. A studio that cannot hold one dimension and one background value across a set has a process problem, and the numbers say so without anyone needing to be persuaded of anything. If your current supplier measures clean, the honest answer is that you do not need to change supplier — and that is what you will be told.
Yes. The method and the command are published on the catalogue consistency standard. The first two checks take about twenty minutes on two hundred frames with any tool that reports pixel dimensions and pixel values. The audit exists because most people would rather not spend the twenty minutes.
The difficult ones, not the clean ones. Drift shows across a range, so a spread of several products from one category tells you more than twenty frames of a single item. If there is a hard material in the batch — lace, sheer fabric, fur, wire, anything translucent — include it.
No. It is useful context and the field is optional. The measurement does not depend on it, and no supplier is ever named by us.
Only partly, and this is worth being clear about. Measurement catches drift, not taste. A set can be perfectly consistent and still be poorly judged. Where that is what we see, the reply will say so rather than hiding behind the numbers.
That is the most common case and it is fine. Plenty of buyers run this to settle an internal argument, or to have something concrete to take back to the supplier they already use. There is no follow-up sequence attached to it.
No. We correct and refine — we never generate. Every image is retouched by hand, frame by frame, and that includes frames a client generated themselves before sending them to us. Your files never pass through a generative tool, and they are never used as training data for any model.
Up to twenty finished frames from whoever retouches your catalogue now. We reply within one business day.