Home How it works

How SkinLately works

The short answer

Photograph your face in the same light at the same time every day. Each photo is read for five cosmetic-level dimensions — hydration, texture, calmness, clarity and under-eyes — and rolled into one Glow Score. The score is compared only with your own earlier scores, so a routine that is working shows up as a trend rather than a feeling.

The loop

  1. Pick one spot and one time

    Indirect daylight beats every bulb in your house, and a window is the only light source that is in roughly the same place tomorrow. Pick the spot once. Pick the time once. Most people who quit progress photos quit because they were improvising both every day.

  2. Check the light before you trust the photo

    The Same-Light Check reads the actual brightness of what your camera sees and tells you whether it is usable. It is the same check that runs inside the app before a reading is saved — a photo that is too dark or blown out gets a lower confidence and a prompt to retake, instead of a confident-looking number that means nothing.

  3. Ten seconds, bare face, no filter

    Fill the frame, neutral expression, no makeup, no beauty mode. Front camera is fine. If your phone smooths skin by default, turn that off once and forget about it — otherwise you are tracking the filter.

  4. Read the five dimensions

    Every reading returns the same five, plus three small things to do today, matched to whichever two dimensions read lowest. The actions are ordinary skincare common sense — sunscreen, moisturiser, take a night off exfoliating, sleep — never a product, never a concentration, never a brand.

  5. Change one thing at a time

    This is the rule that makes the rest of it worth anything. If you start a new serum, a new cleanser and a new sleep schedule in the same week, no measurement on earth can tell you which one moved the number. The 30-day protocol is here.

  6. Look at Day 1 next to today

    Two photos, same light, honest dates. That side-by-side is the thing no product review can give you, and it is the only evidence that is actually about your face. You can build one right now with the Before/After Maker from photos you already have.

The five dimensions, in plain words

These are cosmetic descriptions of how skin looks on a given day. They are deliberately not clinical terms, and they are deliberately not the vocabulary of a skin condition.

Hydration — how comfortable and plump the skin looks, rather than tight or thirsty.
Texture — how even the surface reads — smooth versus bumpy or rough in patches.
Calmness — how even the tone is. Less pink and blotchy reads as calmer.
Clarity — how clear the surface is, without naming or judging anything on it.
Under-eyes — how bright or shadowed the under-eye area looks today.

Why “calmness” and not the obvious word. The obvious word is a medical one, and a tool that uses it starts sounding like it knows something about your health. It does not. Every label above describes an appearance, which is all a camera can honestly report on.

What the Glow Score is — and is not

It is a single 0–100 number combining the five dimensions, and it exists because one number is easier to watch move than five. It is a relative reference: useful against your own history, meaningless as an absolute.

  • Not a grade. There is no pass mark and no target.
  • Not a comparison with other people. There is no leaderboard, no friends feature and no percentile anywhere in the product, and there never will be.
  • Not a health measurement. Skin that looks calm today is not skin that is healthy, and we do not have any way of telling the difference.
  • Not stable across lighting. Change the light and the number moves. That is a property of the measurement, not a flaw we are hiding — we are running a study to say by how much.

The one thing it says about anything unusual

If a reading finds something outside what a cosmetic tool should be commenting on, it raises a flag and shows exactly this, with nothing added:

Worth a professional look. Something here is outside what a cosmetic tool should comment on — please show it to a dermatologist.

No name, no severity, no “this looks like”. The reason is simple: a phone camera plus a model is not in a position to tell you what something is, and a wrong guess in either direction — frightening you, or reassuring you — does real damage. So the product is built so that guess is not available to it.

Free, paid, and where the line sits

The two browser tools on this site are free, unlimited and account-free, permanently. In the app, the first scan is free; the daily scan, the trend, the five-dimension breakdown and the before/after album are the subscription. $2.99 a week or $29.99 a year, after a 3-day free trial. Cancel any time.

Payment happens inside the app, through Apple, and nowhere else. This website does not have a checkout, does not have accounts, and does not run a scan — the tools here are pure browser code, and the interactive demo runs on sample data.

Questions people ask at this point

How long does a scan take?

Ten seconds, most of which is you standing in the right spot. The reading itself is a couple of seconds.

What happens to my photo?

In the app: the photo is analysed and then deleted. Five numbers and a timestamp are what gets stored. The original stays on your phone, and one tap deletes everything. On this website there is no upload at all — the tools never send a pixel anywhere. The long version is on the privacy page.

What if I miss a day?

You get one free miss a month — the streak survives it and the app tells you it used it. Beyond that the streak resets, because a streak that never breaks is not measuring anything.

Can I use it without the daily habit?

You can, but the number stops meaning much. A single Glow Score is a reading with nothing to compare it to. The whole design assumes the second reading.

Does it work on every skin tone?

The five dimensions are relative measures — each reading is compared with your own previous readings, not with a reference face. That design avoids the worst failure mode of absolute scoring, but it does not make the underlying model tone-neutral, and we are not going to claim it does. Our Same-Light Study reports its consistency numbers split by skin tone for exactly this reason, and it will report them even if they are unflattering.