I had accumulated an eleven step routine over several years, added one product at a time, each for a reason that made sense at the time.
I dismantled it over about three months and ended at four. This is a description of the process rather than advice, and anybody with a diagnosed condition should be talking to a dermatologist rather than reading this.
Why I could not tell what was working
The problem that started it.
With eleven products applied daily, attributing any change to any one of them is impossible.
Skin also varies for reasons unrelated to what is applied — season, sleep, stress, hormonal cycle, diet, water hardness — which produces variation that is easy to attribute to a product.
And the timescales are long. Meaningful change from most actives takes weeks to months, which is longer than most people persist before adding something else.
So I had a system I could not evaluate, built from decisions I could not test.
How I dismantled it
One at a time, with a gap.
Remove one product, wait four weeks, note any change. Then the next.
This is slow and it is the only method that produces information. Removing several at once tells you nothing.
The results were unflattering to most of what I had been using. Removing eight products produced no change I could detect.
Two removals produced a noticeable deterioration and were reinstated.
What survived
Four things.
A gentle cleanser, because the alternative left my skin uncomfortable.
A moisturiser, which had an immediate and obvious effect when removed.
Sunscreen in the morning, which is the one with the strongest evidence behind it and which I would keep regardless of whether I could perceive a difference, because the effect it is preventing is not visible on a four week timescale.
And one active, used a few nights a week, which produced a change I could see when removed.
What did not survive
Toners, which I had never had a clear reason for.
Several serums with ingredients I could name and whose effect I could not detect.
Essences and mists, which appeared to do nothing measurable.
An eye cream, which turned out to be a moisturiser at a higher price in a smaller container.
And an exfoliant used more often than it should have been, whose removal improved matters, which was the clearest single result.
The over-exfoliation finding
Worth expanding because it was the most significant.
Several of the products I was using contained active ingredients doing similar work, and the combined effect was more than I had intended.
The symptoms — sensitivity, irritation, a tight uncomfortable feeling — I had been treating by adding soothing products rather than by removing the cause.
That is a recognisable pattern and it is common. A routine that grows by addition can end up treating problems it created.
Why routines grow
The mechanism is worth naming.
Each product is marketed against a specific concern, and identifying a concern is easy if you look closely enough at your own face.
Content about routines is extremely popular and demonstrates elaborate regimens, which normalises the length.
And there is no natural stopping point, since no product announces that you have enough.
The industry has no incentive to suggest a floor, and the floor is where most of the benefit sits.
What I would suggest to somebody starting
Cleanser, moisturiser, sunscreen. That is a complete routine and for a lot of people it is sufficient.
Add one active if there is a specific thing you want to change, introduce it slowly, and give it twelve weeks before judging.
Change one thing at a time, always, because otherwise you learn nothing.
And be sceptical of any product whose promised effect you would not be able to detect if it were removed, which is a surprisingly demanding test.
The cost of the experiment
Worth stating since it was not free.
Three months of deliberately removing things included a period where my skin was worse, which was the point and was still unpleasant.
Anybody doing this should probably not start before an occasion that matters, and should be prepared to reinstate something if the deterioration is significant rather than persisting on principle.
The other cost was the products I stopped using, which were largely unfinished and which I gave away rather than throwing out.
What I would do differently
Remove the cheapest and least evidenced things first rather than working through in the order I had added them, which would have reached the useful conclusions faster.
The order question
Worth a note since it comes up constantly.
The general convention is thinnest to thickest, so that lighter preparations are not blocked by heavier ones.
How much this actually matters is less established than the confidence with which it is stated, and for a short routine it matters very little.
The one ordering point with reasonable support is that sunscreen goes last in a morning routine, since it is intended to form a film on the surface.