What we can show you — and what we cannot.
The reasoning behind this house is spread across a science page, a story, six formulation pages and a journal. This page gathers it in one place, says where each part lives, and is equally plain about the parts that do not exist yet.
Two effects, on two clocks.
Every Zero Lines formulation is made to do two things at once: an immediate effect on how the skin looks, and a lasting one that consistent use is designed to build underneath. The industry sells the first and implies the second. We name which is which, every time, and the rest of this page is about how far each of them is actually supported.
The immediate effect
This is a claim about appearance, and only about appearance: applied to an expression line, the Line Corrector is designed to make the surface read smoother within minutes and hold that look through the day. It is a change in the way the skin’s own surface sits and catches light.
What supports it: it is the kind of claim a person can judge for themselves, on the day, in a mirror. What does not support it: any figure. Nothing here has been quantified in a trial, because the formulations are not on sale and no such trial exists.
It does not act on muscle. Nothing sold without a prescription does, and the journal has a piece that works through exactly why the topical answer to a line made by muscle is a doctor rather than a jar.
The lasting effect
This one is slower and harder to sell. Consistent use is designed to support the skin’s own condition — barrier, hydration, the structure that gives skin its resilience — over weeks and months rather than days.
What supports it: published literature on the ingredient classes in the formulations, which is the reason each was selected. That is evidence about peptides, ceramides and collagen fragments as classes. It is not a completed study of a finished Zero Lines product, and the two are not the same thing. The science page sets out the rationale and the questions page answers the same point in the words people actually use.
The honest limit of the lasting effect is that you cannot see it on the first morning, and nobody — including us — can tell you in advance how much of it your own skin will show.
Why you will not find a figure here.
Most competitor pages carry a percentage and a number of days. Where the small print says what was actually measured, it is usually consumer self-assessment: a group of people were given a product, used it for a few weeks, and were asked whether their skin felt better. That is a real thing to measure, and it is not what most readers assume they are reading. It is worth looking for the footnote, on any page, including this one.
This house has never sold anything to anybody. There is no group, no questionnaire and no result, so there is no figure we could publish that would mean what a figure appears to mean. We would rather have the empty space than borrow the appearance of rigour.
The same rule covers the laboratory: no percentage, no panel result, and no substantiation language until a completed study exists with a reference a reader can go and read. It is the second of six refusals the house wrote down before it wrote its product pages, and it has cost us more copy than any of the others.
Efficacy percentages published by this house, anywhere, to date.
Consumer panels run, because there are no consumers yet.
Testimonials on the wall. That page stays open until it can be honest.
A machine reads it before you do.
Good intentions do not survive a rewrite. The failure this house actually had was not somebody deciding to overstate a result; it was an ordinary marketing adjective sitting in a heading on the flagship formulation page, surviving a full rewrite because nobody was looking for it.
So a claim checker now reads this site before anything goes up. It is a script kept beside the pages, some three hundred lines of it, and it reads every page in the repository except three: the terms, the privacy notice and the cookie notice have to be able to use the words a legal document uses, including saying that a cosmetic is not intended to cure anything. It fails on a single hit, and a page it objects to is rewritten rather than argued with. It works in two tiers.
- Language never acceptable in this house’s own voice — substantiation claims, professional endorsement, promises of permanence, any percentage attached to an outcome, favourable comparison with a prescription medicine, or offering a cosmetic as suitable for a named condition.
- Blocks
- Language acceptable when the journal explains the wider field, and not acceptable when the house describes itself. An article about what the industry’s phrases commit a manufacturer to has to be able to name them.
- Depends
- The structured data — the machine-readable block search engines quote — read as the house’s own voice, because that is exactly what it is.
- Both tiers
What it has caught
A house that publishes only its wins is asking to be taken on trust. These are real findings, on this site, each of which had already been written, reviewed and shipped:
- An unsupportable adjective in a heading on the flagship formulation page. It reads as ordinary marketing, which is exactly why it survived a full rewrite: nobody was looking for it.
- Claims that had been deleted from the visible copy of a page and were still living in that same page’s structured data, where only a crawler would ever see them. The checker had never read structured data until the day those were found. It does now.
- A step in that structured data which told search engines the day cream would protect the wearer from the sun. Its own page says four times that it carries no SPF and that a sunscreen goes on top of it.
- A named academic at a real university, quoted with a title, appearing nowhere else in the house’s writing. Every other quotation in the journal is the formulation team.
There is a fifth category worth admitting: the checker is also wrong sometimes. It objected to the day cream page for saying that no single morning is dramatic — a sentence written to lower an expectation, caught by a rule that was only looking at the adjective. A checker that fails honest writing is a checker somebody quietly switches off, so it now looks back for a negation and forward for the question mark that makes a headline a question before it objects.
This page was written to pass it, like every other. That is a constraint on how we may describe our own checker, and it seems fair that it should be.
One spring, and one missing document.
What is known
- Every formulation begins with water from a single spring in the Andorran Pyrenees, at just over 1,800 metres.
- It percolates through granite, limestone and gneiss, and reaches the surface after fifteen to twenty-five years underground.
- It carries zinc, selenium, copper and magnesium, and it is low in sodium — which matters, since a high-sodium source draws moisture the other way.
- The valley has no agriculture, no industry and no permanent settlement within five kilometres, so there is nothing in it that has to be filtered back out.
- Those trace minerals are cofactors: the elements the body’s own antioxidant and repair enzymes require in order to work. That is general skin biology, established long before this house existed, and it is a statement about the minerals rather than a finding about this spring.
The geology gets a fuller treatment in why the base of a formula matters and in what altitude changes about water.
What is not published
A certificate of analysis. There is no table of milligrams per litre on this site and there will not be one until there is a batch-release document behind it with the laboratory’s name attached. A number to the milligram implies a paper trail, and quoting one we cannot hand over would be the same offence as quoting a percentage.
What follows from that: the mineral profile is described qualitatively here, and anyone who needs the figures should ask us for them rather than assume the site is holding them back. It is not; they are not finished.
A reading, not a measurement.
The skin analysis is free, needs no account and no payment, and is open now rather than at launch. Sixteen questions and one photograph produce a written assessment across eight markers, which is sent to the address you give — it does not appear on the page, and the email carries a private link so you can keep it.
It is read by a machine rather than by a person: a large language model, currently GPT-4o. Your photograph is read once, is not stored, is never attached to your email address, and is not sent to you or to anyone else.
- Describes the appearance of your skin and suggests a routine
- Yes
- A medical diagnosis, or a substitute for a healthcare professional
- No
- A measurement — an instrument reading with a unit attached
- No
- A record. It does not remember you between visits, and does not track change over time
- No
- Sensitive to the light you stood in. Overhead bulbs carve shadow into every line and read as texture that is not there
- Yes
The assessment says so itself, at the top and at the foot: if something on your skin concerns you medically, please see a doctor.
The list we would rather you read.
Everything above is what a sceptic asked for. This is what a sceptic would find on their own, and it is shorter to publish it ourselves.
- A completed study of a finished formulation Nothing is on sale, so no such study exists. What exists is published literature on the ingredient classes, which is the reason each was chosen — a different and weaker thing, and we will not blur them.
- Not begun
- A consumer panel Nobody has used the six-step protocol for eight weeks and been asked what they thought of it, because nobody has been able to buy it.
- Not begun
- Safety and tolerance assessment under dermatological supervision Part of the programme every formulation must pass before it is sold — patch testing, instrumental measurement, self-assessment. It is not complete. It will be published with the method attached, and no animal is involved at any stage.
- In progress
- A certificate of analysis for the spring water See section four. Qualitative description until there is a batch-release document with a laboratory name on it.
- Not published
- Full ingredient lists Every active is named on its formulation page today. The complete lists, in the regulated format, go up when the formulations go on sale and appear on the packaging at the same time.
- Not published
- Vegan certification Most ingredients are plant-derived or synthetic. The peptide complexes are made by fermentation, and the growth medium may carry trace dairy-derived substrates, so the house does not describe the range as vegan and will not until a certificate says so.
- Not held
- Customer accounts of using the protocol The testimonials page is deliberately held open. No purchased reviews, no composite customers, and no voices before there are customers to have them.
- None
- A launch date We have not set one, and we would rather say that plainly than guess at one and be quoted on it later.
- Not set
If an item on this list changes, it changes here first. This page carries the date it was last reviewed for that reason, and for no other.
Ask us for the thing that is missing.
A question this page does not answer is a question worth putting to us directly — about the sourcing, the laboratory, the regulatory position, or anything above that reads as too convenient. It reaches the people who formulate, not a script.