“9 out of 10 agreed”: consumer perception tests explained

“9 out of 10 women agreed their skin felt more hydrated.” You’ve seen the line, usually in small print with an asterisk. It comes from a consumer perception test, and it’s one of two very different ways skin care companies back up a claim. The other is measuring the skin itself. Here’s what each one tells you, and why the asterisk matters.

Two kinds of evidence

Skin care claims lean on two families of evidence, and the EU’s technical guidance on cosmetic claims (a best-practice document used by European regulators) names them plainly: experimental studies and consumer perception tests, used alone or together, often with published research on the ingredients.

Experimental studies measure something about the skin. That could be an instrument reading, a trained assessor grading redness or flaking on a scale, or photographs analyzed under fixed lighting.

Consumer perception tests, in the EU document’s words, “evaluate consumers’ perception of product efficacy and cosmetic properties based on parameters that they can observe or feel.” People use the product, then answer a questionnaire.

Neither is fake. They answer different questions: “what happened to the skin?” versus “what did people notice?” Trouble starts when an answer to the second is dressed up as an answer to the first.

What the instruments actually do

A few devices show up again and again in skin care studies.

A corneometer measures skin hydration indirectly. A small probe is pressed on the skin and reads its capacitance, an electrical property that rises as the outer layer holds more water. A 2022 manufacturer-funded study in Skin Research and Technology describes it as a well-established method that gives “a single dielectric value (arbitrary units)” integrating hydration to a depth of about 45 micrometres, the very top of the skin.

A TEWL meter measures transepidermal water loss: how much water vapour escapes through the skin per hour. Higher loss means a weaker skin barrier (the outer layer that keeps water in and irritants out). A 2018 pilot study in Skin Pharmacology and Physiology notes that intact forearm skin in its participants ran around 4 to 9 grams per square metre per hour, that irritants can push this to two or three times baseline, and that the values “can be influenced by a wide range of factors.”

Colorimeters read redness; pH meters read surface acidity; cameras under standard lighting track wrinkles or pores.

The conditions matter as much as the device. In the 2022 study, participants sat for 30 minutes in a room held at 21°C and 50% humidity before anyone took a reading, and measurements were made by trained staff. Skip that, and a cold walk from the parking lot changes the numbers.

What a perception test actually does

A perception test is a structured survey. Participants use the product, usually at home, for days or weeks, then rate statements: “my skin feels softer,” “my skin looks less red,” often on a five- or seven-point scale. Agreement is counted, and the count becomes the number on the box.

The EU document sets out what a good one looks like: a “statistically representative sample of the target population” with clear inclusion criteria, proper statistical analysis, a written report with protocol and results. And it flags the weak point directly: “A critical point for the validity of consumer tests is the wording of the questionnaire.”

That last line is the whole story. “Did your skin feel moisturized right after applying?” will get near-universal agreement for any cream with oil or glycerin in it. “After four weeks, was your dry skin better than with your usual product?” is a much harder question, and a much more useful answer.

Why companies use both

Instruments are precise but narrow. A corneometer can show a hydration bump 30 minutes after application; it can’t tell you whether the cream felt greasy, whether you’d use it again, or whether the change was big enough to notice. Perception tests capture exactly that, and for a product people buy because of how it feels, that’s not trivial.

The 2022 study is a decent example of the two working together. The moisturizer produced significantly higher corneometer readings than its base cream at 30 minutes and 24 hours, and, separately, 64% of participants rated it more effective against dry skin than the base cream (22% preferred the base). Instrument and opinion pointed the same way, which is what you want to see.

There’s also a practical reason: the 2018 pilot study points out that older ways of judging skin irritation “relied upon subjective visual assessment,” which can’t detect changes “before changes are seen by the naked eye.” Instruments catch what eyes and questionnaires miss. Questionnaires catch what instruments can’t ask.

What a survey can’t do

A perception test cannot prove a product changed your skin. The US Federal Trade Commission’s health products guidance says it bluntly: anecdotal evidence, “including surveys of consumer experiences, are never sufficient to substantiate claims about the effects” of a product, because reported experiences “may be attributable to a placebo effect or other factors unrelated to the product.”

That’s a US rule for health claims, but the logic applies to any cosmetic claim that sounds like an effect. In Canada, the Competition Bureau’s guidance on performance claims asks for tests done under controlled conditions that keep subjectivity to a minimum. A questionnaire is subjectivity by design.

The EU document draws the same line for specific phrases. “Dermatologically tested,” it says, implies testing on humans under a dermatologist’s supervision, and “consumer self-perception studies are not appropriate to support such claims.” Likewise “clinically tested” means testing “according to a clinical protocol or in a clinical setting,” not a mailed questionnaire.

So when a label says “9 out of 10 agreed” next to “clinically proven,” check whether those are two different studies or one study wearing two hats.

Reading the number

When you see “X out of 10” or “87% agreed,” ask:

• Agreed with what, exactly? The question is the result. “Felt hydrated” and “dry patches gone” are different claims.
• How many people? 9 out of 10 could literally be ten people. Good tests run in the dozens to low hundreds.
• Compared with what? Did they also use a plain cream, or their usual product? A preference is more telling than a rating in isolation.
• After how long? Immediately after applying, or after four weeks of daily use?
• What about the 1 in 10? Neutral, or actively worse? Surveys usually count “agree” and “strongly agree” together and don’t mention the rest.
• Was anything measured? If the company ran a corneometer or TEWL study too, the perception result has backup. If the survey is all there is, the claim is about feelings.

What to make of it

A good perception test is honest evidence about experience: whether a product feels nice, absorbs well, and seems to help. That matters, because a cream you hate using won’t get used. A good instrument study is honest evidence about the skin: whether hydration went up, water loss went down, redness faded. The best claims have both, and say which is which.

If you’re managing a skin condition rather than shopping for a feel, eczema (atopic dermatitis, the itchy inflamed rash), persistent dryness or irritation, a survey number is the wrong kind of evidence to lean on. If it isn’t better after two weeks, see your doctor or pharmacist.

What to remember: “9 out of 10 agreed” measures opinion, not skin. Instruments and trained graders measure skin. Both are used because both matter, but a survey can’t prove an effect, and its number is only as good as the question behind it. Find the question, the sample size and the comparison, then decide how much the asterisk is worth.