Skincare has a strange talent for turning discomfort into reassurance. A product prickles, warms or stings, therefore it has reached the important bit and begun earning its keep. If nothing happens, you can find yourself looking at the bottle wondering if you’ve bought the correct thing.
The sensation is persuasive because it arrives at the perfect time. You apply the product, wait at the sink, feel the first sharp flicker and receive what appears to be immediate confirmation. Most outcomes you actually bought the product for are far less obliging. Acne lesions, pigmentation and changes in keratinisation answer over days or weeks. Collagen has never been known for giving swift feedback.
A sting does confirm that your skin detected an exposure. An ingredient may have reached sensory-active tissue, changed the local chemical environment or opened an ion channel. This is information about the encounter. It is not a measurement of the intended result.
A product can sting while it’s working, but the sting cannot tell you that it is working, how well it is working or whether the same result could have arrived without the discomfort.
Your Skin Can Report the First Second
Sensation and efficacy run on different timetables. Sensory nerves can respond within seconds to heat, acidity and chemicals acting on channels such as TRPV1, the receptor made famous by capsaicin and the burn of chilli. Keratinocytes can also generate calcium and inflammatory signals when exposed to certain molecules. Your face is perfectly capable of reporting that something’s up.
What it cannot report in that moment is the future. There is no receptor for ‘this will clear the acne by week eight’, or ‘collagen synthesis has begun, please remain at the mirror’. The outcomes depend on a longer chain of receptor binding, gene regulation, cell behaviour and time.
The two processes may start with the same ingredient, which is why the folklore feels plausible. The mistake is treating a fast sensory output as a progress bar for a slower biological one. A stronger sting shows a stronger sensation. The rest still has to be measured on its own terms.
One Molecule Can Hold Two Conversations
Retinoids make the separation unusually clear. Their therapeutic effects are classically associated with nuclear retinoic-acid receptors, which alter gene transcription and influence cell differentiation. Their capacity to provoke burning and hypersensitivity can travel through another route. This helps explain why a retinoid may be both effective and irritating without making the irritation the source of the effect. It is the same logic behind asking why azelaic acid stings rather than deciding that the sting itself is treatment.
In 2013, Shijin Yin, Jialie Luo, Aihua Qian and colleagues tested natural and synthetic retinoids against TRPV1. Several activated the channel in engineered cells and in mouse sensory neurons. Retinoids also produced pain-related responses in animals, which were reduced when TRPV1 was blocked with an antagonist or removed genetically.
This is a rather elegant piece of biological untangling. A molecule can speak to nuclear receptors about longer-term cell behaviour and to an irritant channel about what your nerves should report now. Turning up the second conversation does not prove that the first has become more productive.
The work was mechanistic, largely using engineered cells, isolated mouse neurons and animal models. It was not a trial showing that the amount a human face stings predicts a topical retinoid’s benefit. The study was supported by public and academic grants, including the US National Institutes of Health and the National Natural Science Foundation of China, and the authors declared no conflicts of interest. Its value here is the separation of pathways, not a rule for dosing your retinoid by feel.
Sometimes the Active Isn’t Making the Fuss
A formula contains more than the ingredient advertised on the front. Solvents, preservatives and pH all meet your skin too, although they rarely get featured in the marketing ad. The active has a habit of taking the credit for everyone in the bottle.
D-G Li, H-Y Du, Gerhard Schmaus and colleagues showed this with phenoxyethanol, a preservative rather than a treatment target. They screened 243 Chinese women and deliberately enrolled the 60 who reported phenoxyethanol discomfort. In a split-face experiment, the researchers applied a formulation containing 1% phenoxyethanol to the nasolabial folds and compared it with the same exposure plus a TRPV1 antagonist. Burning and itching were lower on the antagonist-treated side. In HaCaT keratinocytes, phenoxyethanol also produced a calcium influx that the antagonist blocked. This is one reason supporting formula ingredients deserve attention, even when the front label has supplied a more famous suspect.
The study tells us something narrow and helpful: an ingredient with no therapeutic role can generate the sensory signal people often award to the active.
The commercial context is worth keeping in the room. Gerhard Schmaus and Imke Meyer were affiliated with Symrise, and the TRPV1 antagonist being tested had clear cosmetic value. The paper declared no conflict of interest, but the company affiliation gives the soothing side of the experiment a commercial stake. That does not make the split-face result disappear. It does mean the selected cohort, the model system and the interest in the antagonist should travel with the finding, rather than be left in the cloakroom.
More Irritation Does Not Buy More Effect
The inverse example comes from a randomised, investigator-masked trial led by Pengfei Tu, G-Q Li, X-J Zhu and colleagues. One hundred and fifty people with grade II–III acne at three Chinese clinics used adapalene 0.1% gel or tretinoin 0.025% gel once daily for eight weeks. The researchers counted inflammatory, non-inflammatory and total lesions, while rating erythema, burning, itching, scaling and dryness separately.
Both groups finished with average lesion-count reductions of 69 to 74 per cent. More than 70 per cent of participants in each group were judged to have complete clearance or marked improvement. Irritation was generally mild, but it was more common and more severe with tretinoin. The easier treatment had not forgotten to work.
This was a comparison of two retinoids and two gels, not a test in which the same formula was made more or less painful. It also did not ask whether one person’s burning score predicted that person’s lesion response. The trial measured clinical improvement and irritation as separate outcomes, and the findings show why they must remain separate: a similar acne result arrived with different sensory costs.
The Sting Can Still Tell You Something
Stinging isn’t empty information. It can say a great deal about the conditions receiving the product. You can apply the same formula twice, to what appears to be the same face, and receive two quite different reports. Dry weather, a recent exfoliation, vigorous cleansing or an already irritated barrier can change the threshold even when the bottle has behaved accordingly.
Dong-Mei Ding, Ying Tu, Mao-Qiang Man and colleagues studied 292 Chinese women with self-perceived sensitive skin. After a standardised lactic-acid sting test, higher sting scores were associated with higher transepidermal water loss and lower stratum-corneum hydration. The scores also tracked self-reported stinging, itching, tautness and scaling, though not burning or visible redness.
The relationships were cross-sectional and partial. The test used a deliberate acid challenge in a specific group, so it cannot turn one uncomfortable serum into a diagnosis of barrier damage or sensitive skin. Higher TEWL does not disclose the whole barrier, either. What the study adds is a plausible reason the same exposure can become more pronounced when hydration and permeability have changed. The work was funded by public programmes in Yunnan Province, and the authors reported no conflicts of interest.
What to Do at the Sink
A mild, brief tingle that fades without redness or a worsening reaction tells you the exposure was felt. It doesn’t earn the product extra efficacy points. You still judge it by the outcome it promised, over the timescale on which that outcome can reasonably appear.
Persistent or escalating burning, marked redness, swelling, blistering, a spreading rash or pain changes the question. The sensible response is to stop treating the discomfort as evidence of progress, and to check in with a trained medical professional.
A product can work without telling you it’s working. It can sting while working, and it can sting because the formula or the skin is objecting. The sensation cannot adjudicate between them.
Stinging records an encounter, not its value.
References
Yin, S., Luo, J., Qian, A., et al. (2013). Retinoids activate the irritant receptor TRPV1 and produce sensory hypersensitivity. Journal of Clinical Investigation, 123(9), pp. 3941–3951. https://doi.org/10.1172/JCI66413
Li, D-G., Du, H-Y., Gerhard, S., et al. (2017). Inhibition of TRPV1 prevented skin irritancy induced by phenoxyethanol: a preliminary in vitro and in vivo study. International Journal of Cosmetic Science, 39(1), pp. 11–16. https://doi.org/10.1111/ics.12340
Tu, P., Li, G-Q., Zhu, X-J., et al. (2001). A comparison of adapalene gel 0.1% vs. tretinoin gel 0.025% in the treatment of acne vulgaris in China. Journal of the European Academy of Dermatology and Venereology, 15(Suppl. 3), pp. 31–36. https://doi.org/10.1046/j.0926-9959.2001.00010.x
Ding, D-M., Tu, Y., Man, M-Q., et al. (2019). Association between lactic acid sting test scores, self-assessed sensitive skin scores and biophysical properties in Chinese females. International Journal of Cosmetic Science, 41(4), pp. 398–404. https://doi.org/10.1111/ics.12550
This article contains no affiliate links, sponsored content or product recommendations. Scientific claims are grounded in the peer-reviewed literature cited above. The phenoxyethanol study included two Symrise authors and tested a commercially relevant TRPV1 antagonist. The adapalene comparison was funded by Galderma International, which had a commercial interest in adapalene. Those connections do not invalidate the studies, but they are included in the text because they change how the evidence should be weighed.
