Understanding the causes of skincare product pilling

Abstract

Background

Skincare and makeup “pilling” is an unsightly and undesirable phenomenon whereby skincare such as moisturizers or foundation ball up to form flakes on the skin. To date, the causes of skincare product pilling have not been studied. This study aimed to examine the relationship between skin physiology and pilling potential of sunscreen and foundation (the two products most reported by consumers to cause pilling). This study also examined the effects of product application methods on pilling.

Materials and methods

528 female volunteers from Guangzhou, China, aged between 20 and 49 years, underwent various clinical skin assessments, followed by three steps of product layering. Pilling was assessed after each product application step.

Results

217 volunteers (41%) experienced pilling. The majority of pilling (n = 655 events) occurred following sunscreen application, while only a few pilling events (n = 35) occurred with foundation. Foundation improved pilling caused by sunscreen in 98.9% of cases. Volunteers experiencing pilling with both sunscreen and foundation had significantly lower facial skin hydration and oiliness, higher pH, and smoother skin texture (P < 0.05). Two application methods, rubbing of products in circular and linear motions, yielded the highest numbers of pilling events.

Conclusion

This study has provided the first insights into the causes of pilling. Sunscreen is a promoter of pilling, while foundation may resolve sunscreen‐induced pilling in many cases. Skin physiology, particularly drier, smoother skin with higher pH, and product application methods are likely contributing factors to this undesirable phenomenon.

1. INTRODUCTION

Skincare and makeup “pilling” is an unsightly and undesirable phenomenon whereby skincare such as moisturizers or foundation ball up to form flakes on the skin (Figure 1). Market research has shown that consumers are bothered by product pilling as they need to remove the pills; pilling was found to be especially bothersome in the morning when people are usually more rushed for time. Consumers also believe that product pilling affects product absorption and hence product efficacy. ¹ To date, there is no peer‐reviewed literature on the causes of pilling; hypothesized causes include an incompatibility between a person's skin properties and the product ingredients; an incompatibility between multiple products applied together; the inherent skin properties of a person which increase pilling potential; and incorrect skincare application methods. Product layering is also thought to increase pilling potential; market research has shown that the average Chinese consumer layers approximately seven products in one application. ¹

FIGURE 1.

Examples of the different appearances and grading of skin pilling.

It is generally accepted that certain products, particularly sunscreen and foundation, are more likely to lead to pilling. Sunscreen and foundation are thought to be a significant promoter of pilling. This may be caused because their ingredients generate a film on the skin that has the potential to be broken by the rubbing of the skin during subsequent product application.

To understand the causes of product pilling, this study aimed to: (1) determine the relationship between skin physiology and pilling potential; (2) determine the relationship between product application method and pilling potential; and (3) determine the effects of sunscreen and foundation on pilling potential since these two products are most reported by consumers to cause pilling. The study was conducted on Chinese female volunteers with varying skin physiologies.

2. MATERIALS & METHODS

Institutional Review Board approval was given by the Shanghai Ethics Committee for Clinical Research (Approval ID: SECCR 2024‐21‐01, project ID GZ23500002). This study was conducted between July 2022 and February 2023.

2.1. Participants

528 female volunteers from Guangzhou, China, aged between 20 and 49 years, completed the study (Table 1). Volunteers satisfied inclusion and exclusion criteria. Inclusion criteria were healthy non‐smoker female volunteers of age 20–49 years, with Fitzpatrick Grade II and IV; must have experienced pilling with skin care products; should use face skincare products twice a day; willing to provide written informed consent; willing to sign a photography/video release; willing to cooperate and participate by following study requirements for the duration of the study. Exclusion criteria were: the presence of skin diseases such as psoriasis, vitiligo, eczema, atopic dermatitis and cancer; volunteers who have gone through chemical peel, mechanical dermabrasion, microdermabrasion, non‐ablative laser or fractional laser resurfacing of the face and neck, radio frequency and injectable fillers in the last 2 months; had facial reconstruction (plastic surgery); started, changed or stopped a hormonal treatment (hormonal contraception, hormone replacement therapy), DHEA‐based treatment in the past 3 months; volunteers who exposed themselves to the sun or to artificial UV light for prolonged periods in the last 2 months.

TABLE 1.

Volunteer profile of the 528 Chinese women from Guangzhou, China, who took part in this study.

Age group	n	% of n
20‐29	175	33.1%
30‐39	178	33.7%
40‐49	175	33.1%
Fitzpatrick grade
II	5	0.9%
III	492	93.2%
IV	31	5.9%
Perceived skin type
Normal	56	10.6%
Oily	27	5.1%
Dry	85	16.1%
Combination tending to be oily	169	32.0%
Combination tending to be dry	191	36.2%

2.2. Products used in the study

To understand if ingredients have an impact on pilling, we developed 10 simple product formulations which were used as a base to be layered with sunscreen and/or foundation. The composition of each of the 10 products is provided in Supplementary Table S1. The tested sunscreen reference number is 5P 80071273 and the tested foundation reference number is SI 80039689.

2.3. Product application method

Volunteers were allowed to apply products with their hands using any method they wanted. The method(s) used by volunteers were recorded.

2.4. Skin assessments

Volunteers’ skin properties were measured prior to any product application (Figure 2). Volunteers cleansed their face with makeup remover and then with cleanser and acclimatized in a controlled environment (temperature 21°C ± 1°C and humidity 50% ± 10%) for at least 30 min (Figure 2, step 2). Skin parameters measurement sites are shown in Supplementary Figure S1. Measurement sites were the left and right forehead, and upper and lower cheek. A total of 12 measurement points were taken in total for each equipment except Colorface. Various skin parameters were then measured, as listed below.

FIGURE 2.

Study design.

2.4.1. Sebum (oiliness)

Sebum at the skin's surface was measured using a Sebumeter ® SM 815 (Courage & Khazaka). A probe covered with a sebum‐sensitive clear film was applied to the skin and the total quantity of sebum taken up on the film (µg/cm²) was measured. Each measurement site was assessed once.

2.4.2. Hydration

The level of cutaneous hydration of the superficial layers of the epidermis (stratum corneum) was measured using a corneometer CM 825 (Courage & Khazaka). The measurement method was based on determining the capacitance of the dielectric medium. Measurements were reported in arbitrary digital data units from 0 to 125. Three successive measurements were taken at each site and the mean value was calculated.

2.4.3. Skin pH

Cutaneous pH measurements were taken with a Courage & Khazaka PH 905 PC Skin‐pH‐meter fitted with an Ingold® electrode. Each measurement site was assessed three times, and the mean value was calculated.

2.4.4. Skin texture

Facial skin texture measurements were taken using C‐Cube® from PIXENCE. The size of image acquisition was 12 × 16 mm and the magnification was X60. Three parameters were measured: Sa, representing average roughness; Sq, representing root mean square roughness; and Sdr, representing developed interfacial area ratio.

2.4.5. Skin sensitivity

A cutaneous stinging was performed at the end of the study to assess skin sensitivity. An occlusive filter paper immersed in 10% lactic acid solution was applied to one nasolabial fold, while another filter paper immersed in water (control) was applied to the other fold. At 30 s, 2.5 min, and 5 min post‐application, subjects were asked to score the intensity of the burning/stinging/itching sensations using the 4‐point ordinal scale: (0 = none, 1 = slight, 2 = moderate, 3 = severe). Volunteers whose cumulative stinging score was ≥ 3 will be considered as “positive stingers.”

2.4.6. Desquamation

Skin desquamation was assessed prior to any product application. Dry skin scales were collected using a Corneofix F 20 and analyzed using Visioscan® VC20 Plus.

2.4.7. Water evaporation from the skin

Trans‐epidermal water loss (TEWL) was measured using a Tewameter TM Hex^® (Courage & Khazaka). Each measurement site was assessed for TEWL once.

2.5. Test sample application and pilling ratings

Randomization of applied products was performed (Figure 3). Following each application, pilling intensity (if any) was graded by researchers according to the scale shown in Figure 1. A rating scale of 0−5 was used to grade pilling, where 0 indicated no pilling, 1 indicated slight, and 5 indicated extreme pilling.

FIGURE 3.

Products were selectively applied to one of 12 sites on volunteers’ faces. Numbers in red and blue were the application sites for phase 1 and 2, respectively.

Test sample application was performed in two phases. Volunteers applied a fixed dosage (2.0 ± 0.1 mg/cm²) of product within fixed area of their face (application sites are shown in Figure 2, step 3) using their preferred application method(s). The approximate application area size was 5 cm × 2.5 cm. After the products had been absorbed by the skin the application sites were each split into two areas. The tested sunscreen (reference number 5P 80071273) was applied to one area and the tested foundation (reference number SI 80039689) was applied to the other (Figure 2, step 4). After these products were absorbed onto the skin, the foundation was applied on top of sites with sunscreen (Figure 2, step 5). If pilling was observed, pilling was gently removed before the next layer was applied.

The study was performed in two phases, where six products were tested, and followed by a second phase where the next six products were tested. Between these two phases, volunteers cleaned their face with makeup remover and then with cleanser and were acclimatized in a controlled environment (temperature 21°C ± 1°C and humidity 50% ± 10%) for at least 30 min.

2.6. Statistical methods

Chi‐square test was used to assess whether there is a significant difference in the number of pilling grades between the different products (significance was considered when P < 0.05 was seen, and the chi‐square distributions of the number of pilling grades between products are significantly different). Chi‐square testing was performed using the “chi2_contingency” function of the “SciPy. Stats” library in Python. Variability in the mean values of pilling grades between products was assessed by independent samples t‐test and the Mann‐Whitney U nonparametric test was performed. When using the independent samples t‐test, the default was that the two sets of samples were drawn independently from two normally distributed totals and had the same variance. To make the results robust, we supplemented them with the Mann‐Whitney U nonparametric test. This test does not depend on the distributional assumptions of the data and is suitable for non‐normally distributed data or for small sample sizes. Significance was considered when P < 0.05 was seen for both tests. In experiments, the “ttest_ind” and “mannwhitneyu” functions of the “scipy.stats” library in Python were used to calculate the independent samples t‐test and the Mann‐Whitney U nonparametric test, respectively.

3. RESULTS

3.1. Pilling events following product layering

Approximately 41% (n = 217) of volunteers experienced pilling some stage of the product application process outlined in Figure 2. In step 3 (Figure 2) the total number applications were 6336 (12 facial sites x 528 volunteers) and the total number of pilling events was 61 (0.96%) (Table 2). In step 4, the total number applications were 5280 for sunscreen on top of the base formula (10 × 528) and the total number of pilling events was 655 (12.41%). In step 4, the total number applications was 5280 for foundation on top of the base formula (10 × 528) and the total number of pilling events was 35 (0.66%). In step 5, the total number applications were 5280 (10 × 528) and the total number of pilling events was 61 (1.16%).

TABLE 2.

The total number of pilling events observed at each application step.

	Number of pilling events	Frequency
Step 3	61	0.96%
Step 4 (sunscreen on top of base product)	655	12.4%
Step 4 (foundation on top of 10 base products)	35	0.66%
Step 5 (on top of base product + sunscreen)	61	1.16%

3.2. Skin parameters of volunteers experiencing pilling following sunscreen application

3.2.1. Sunscreen as monolayer on clean skin

A total of 59 volunteers experienced pilling when sunscreen was applied alone to clean skin; this group of volunteers had drier skin (lower facial sebum and moisture; P < 0.05), higher skin pH (P < 0.05), and a smoother skin texture (lower C‐Cube Sa and C‐Cube‐Sq; P < 0.05) (Table 3).

TABLE 3.

Skin parameters of volunteers experiencing pilling when sunscreen was applied alone to clean skin.

	Tewameter	Corneo‐meter	Moistmeter D‐ S15	Moistmeter D‐ M25	Sebumeter	Skin‐pH‐meter	C‐Cube‐Sa	C‐Cube‐Sq	C‐Cube‐Sdr	Desquamation index
Non‐pilling (n = 469) volunteers	14.734	59.653	40.075	35.468	17.422	5.035	18.673	24.117	1.073	31.900
Pilling (n = 59 volunteers)	14.063	54.887 *	38.842 *	34.215 *	11.305 *	5.117 *	16.946 *	22.026 *	0.996	31.224

*P < 0.05.

3.2.2. Sunscreen on top of base formulation

Volunteers who had experienced pilling after sunscreen was applied on top of a base formulation had lower TEWL (P < 0.05; Table 4).

TABLE 4.

Skin parameters of sites experiencing pilling when sunscreen was layered on top of a base product.

	Tewameter	Corneo‐meter	Moist Meter D‐ S15	Moist Meter D‐ M25	Sebu‐meter	Skin pH‐meter	C‐Cube‐Sa	C‐Cube‐Sq	C‐Cube‐Sdr	Desquamation index
Non‐pilling (n = 4625 experiences)	14.663	58.844	39.919	35.334	16.310	5.043	18.518	23.926	1.072	31.551
Pilling (n = 655 experiences)	14.357 *	58.629	39.880	35.172	14.873	5.057	18.694	24.139	1.111	32.008

^* P < 0.05.

3.3. The effect of sunscreen layering on pilling

Participants were found to experience pilling with sunscreen in one of three distinct ways (Table 5). As such, the volunteers were separated into three groups (Group A, Group B, and Group C) and their skin compositions were analyzed. Group A volunteers experienced product pilling when sunscreen was applied on top of a test product, but not when sunscreen was applied alone (Table 5, Group A). Group B volunteers were those who experienced pilling with sunscreen alone, but not if a test product was applied prior to sunscreen (Table 5, Group B). Finally, Group C volunteers were those who experienced pilling with sunscreen, both as a single product and in conjunction with test products (Table 5, Group C).

TABLE 5.

The presence of pilling was evaluated following sunscreen +/‐ skincare test product application in 528 Chinese female volunteers.

Test product ID	Group A (Does not pill with sunscreen as a single layer/application, but experience pilling when applying test samples [first layer], then with sunscreen as second layer) (n of volunteers)	Group B (pilling with sunscreen as single layer, but when applied with test sample, they no longer experience pilling with sunscreen [as second layer]) (n of volunteers)	Group C (pilling with sunscreen as single layer, still pill after applying test samples) (n of volunteers)
P441901	23	39	20
P441902	29	42	17
P441903	25	49	10
P441904	36	38	21
P441905	42	42	17
P441906	45	39	20
P441907	59	37	22
P441908	50	38	21
P441909	70	34	25
P441910	73	29	30

Subgroup analysis across the three groups found a significant difference across several skin parameters between the groups following the application of some of the test products trialed (Supplementary Table S2). Skin parameters were most different between Groups A and B in areas where test products 2, 7, 9, and 10 were used. Skin parameter differences were seen in hydration, oiliness, texture, and pH.

3.4. Skin parameters of volunteers experiencing pilling following foundation application

3.4.1. Foundation on top of base formulation

Volunteers who experienced pilling when foundation was layered on top of a base formulation had drier skin (lower facial sebum, hydration, and moisture; P < 0.05), higher skin pH (P < 0.05), and smoother skin texture (lower C‐Cube Sa and C‐Cube‐Sq; P < 0.05) (Table 6).

TABLE 6.

Skin parameters of volunteers experiencing pilling when foundation was layered on top of a base product.

	Tewameter	Corneo‐meter	Moist Meter D‐ S15	Moist Meter D‐ M25	Sebumeter	Skin pH‐meter	C‐Cube‐Sa	C‐Cube‐Sq	C‐Cube‐Sdr	Desquamation index
Non‐pilling (n = 5245 experiences)	14.627	58.856	39.925	35.325	16.180	5.044	18.550	23.966	1.077	31.613
Pilling (n = 35 experiences)	14.230	52.982 *	38.403 *	33.680 *	8.886 *	5.131 *	16.930 *	21.947 *	1.031	30.762

*P < 0.05.

3.4.2. Foundation on top of base formulation and sunscreen

Volunteers who experienced pilling when foundation was layered on top of a base formulation plus sunscreen had a lower TEWL (P < 0.05), higher sebum levels (P < 0.05), higher skin pH (P < 0.05), smoother skin (lower C‐Cube Sa and C‐Cube‐Sq; P < 0.05), and less scales (lower desquamation index; P < 0.05) (Table 7).

TABLE 7.

Skin parameters of volunteers experiencing pilling when foundation was layered on top of a base product plus sunscreen.

	Tewameter	Corneo‐ meter	Moist Meter D‐ S15	Moist Meter D‐ M25	Sebu‐ meter	Skin pH‐meter	C‐Cube‐Sa	C‐Cube‐Sq	C‐Cube‐Sdr	Desquamation index
Non‐pilling (n = 5219 experiences)	14.639	58.829	39.910	35.312	16.082	5.044	18.559	23.977	1.078	31.639
Pilling (n = 61 experiences)	13.394 *	57.828	40.318	35.464	20.377 *	5.109 *	16.875 *	21.833 *	0.926 *	28.922 *

^* P < 0.05.

3.5. Skin differences between volunteers affected by sunscreen versus foundation

A comparison of skin parameters was performed between volunteers who experienced pilling with sunscreen versus foundation. A total of 655 pilling events were observed following sunscreen application and 35 pilling events following foundation application (Table 8). Volunteers experiencing pilling with foundation had drier skin (lower facial sebum, hydration, and moisture; P < 0.05), higher skin pH (P < 0.05), and smoother skin texture (lower C‐Cube Sa and C‐Cube‐Sq; P < 0.05) than those experiencing pilling with sunscreen (Table 8).

TABLE 8.

Comparison of skin parameters between volunteers who experienced pilling following sunscreen versus foundation.

	Tewameter	Corneo‐ meter	Moist meter D‐ S15	Moist meter D‐ M25	Sebu‐ meter	Skin pH‐meter	C‐Cube‐Sa	C‐Cube‐Sq	C‐Cube‐Sdr	Desquamation index
Pilling following sunscreen (n = 655 pilling events)	14.357	58.629	39.880	35.172	14.873	5.057	18.694	24.139	1.111	32.008
Pilling following foundation (n = 35 pilling events)	14.230	52.982 *	38.403 *	33.680 *	8.886 *	5.131 *	16.930 *	21.947 *	1.031	30.762

^* P < 0.05.

3.6. The effect of foundation layering on pilling

Table 9 shows the number of pilling events once sunscreen was applied to a base product (step 2) and then followed by foundation (step 3). Several volunteers who previously experienced pilling with sunscreen no longer experienced pilling once foundation was applied on top.

TABLE 9.

The number of pilling events observed when sunscreen was layered onto a base product, followed by foundation.

Step 1: Base formula	Step 2: Sunscreen (number of pilling events)	Step 3: Foundation (number of pilling events)
P441901	43	9
P441902	46	14
P441903	35	8
P441904	57	12
P441905	59	2
P441906	65	1
P441907	81	0
P441908	71	3
P441909	95	6
P441910	103	6

All of the 59 volunteers experiencing pilling when sunscreen was applied to clean skin showed reduction in pilling intensity once a base formula + foundation was then layered on top of the sunscreen (Table 10). For example, 23 volunteers who experienced Grade 2 pilling with sunscreen had reduced (Grade 0 or Grade 1) pilling after base formula + foundation was applied. Two volunteers who had no prior pilling (Grade 0) experienced pilling when foundation was layered on top of base formula + sunscreen (Table 10).

TABLE 10.

Sunscreen promoted pilling in 59 volunteers, however the layering of base formula + foundation on top reduced pilling in these volunteers.

Pilling grade	Number of volunteers showed improved pilling after base formula + foundation were layered on top of sunscreen	Number of volunteers showed worse pilling after base formula + foundation were layered on top of sunscreen
0	0	2
1	14	0
2	23	0
3	9	0
4	9	0
5	4	0

Foundation also reduced pilling caused by base formula + sunscreen. A total of 655 pilling events were observed when base formula + foundation was applied. The vast majority (648 cases) of these pilling events were visibly improved after foundation was applied on top of the base formula + sunscreen (Table 11). Only four cases were seen to worsen; 51 new pilling events were observed when foundation was layered on top of base formula + sunscreen (Table 11)

TABLE 11.

Of the 655 pilling events seen when base formula + sunscreen were applied, 648 of these pilling events improved when foundation was applied on top.

Pilling grade	Number of pilling events that improved after foundation was layered on top of base formula + foundation	Number of pilling events that worsened after foundation was layered on top of base formula + foundation
0	0	51
1	186	3
2	274	1
3	58	0
4	128	0
5	2	0

3.7. The effect of application method on pilling

A total of four application methods were used by volunteers: (1) patting product on facial skin; (2) rubbing products in a circular motion; (3) rubbing products in a liner motion; and (4) pressing products on facial skin. These four methods were used in a total of 29 combinations by volunteers. The 18 most frequently used combinations of application methods are shown in Figure 4. The number of pilling events recorded is presented in Figure 4. Rubbing of products in a linear motion was found to yield the highest number of pilling events (n = 266), followed by rubbing of products in a circular motion and then in a linear motion (n = 154).

FIGURE 4.

The number of pilling events observed across different product application methods.

3.8. The effect of application site on pilling

No significant difference in pilling risk was found among the six application sites used when sunscreen was layered onto a base product (data not shown).

3.9. Skin sensitivity and pilling potential

Based on the lactic acid stinging test, no correlation between skin sensitivity and pilling potential was found (data not shown).

4. DISCUSSION

This is the first study to objectively study the impact of skin physiology and skincare ingredients on pilling. These results provide insight into the skin properties that impact pilling potential.

The pilling experience of this study's volunteers was variable and could be categorized into three groups. Some volunteers experienced pilling when a product was applied to clean skin; some experienced pilling during product layering; and some experienced pilling during the first (single) application and reduced or no pilling upon product layering. These differential pilling experiences experienced by the volunteers may be due to differences in the individuals’ skin physiologies. Indeed, this study demonstrated that skin physiology may be a key player in pilling potential. Three different types of pilling experiences were observed in this study. Skin physiology differed between these three groups, with the main differences found skin pH, sebum levels, moisture, and skin texture. An additional factor seen to affect pilling potential was skincare product composition. The 10 different base products assessed yielded differing pilling results with the same volunteers, suggesting that product ingredients may affect the way sunscreen interacts with skin and alter pilling potential.

In this study, the most frequent cause of pilling was the application of sunscreen on top of a base product; 80% of the observed pilling events occurred when sunscreen was applied on top of a base product. This finding supports previous reports that sunscreen can be a significant inducer of pilling. Although we did not investigate the effect of individual ingredients within the sunscreen, we postulate that sunscreen may be a significant promoter of pilling. Ingredients such as zinc or titanium dioxide which are found in sunscreen may also be incompatible with many water‐based products, making sunscreen a product of high pilling potential.

The findings of this study suggest that the layering of foundation on top of sunscreen can reduce pilling. In volunteers who experienced pilling with sunscreen (either applied to clean skin or on top of a base product), the vast majority of pilling was either resolved or reduced when foundation was applied on top. The reason for this remains unclear, it may be due to some ingredients in the foundation that have hydrating effects, thus contributing to the reduction of pilling. We hypothesize that when the skin physiology is altered, pilling experience will also be altered.

The method of product application was found to potentially influence pilling potential. Out of the four application methods volunteers used in this study, the application of product via circular or linear rubbing motions were associated with the majority of pilling events. This is unsurprising as rubbing creates friction, which may lead to product pilling. This is supported by our observation that volunteers who experience pilling with sunscreen alone but not when sunscreen is used with a base product tended to have lower skin hydration.

This study has provided the first insights into the causes of pilling, finding that both skin physiology (such as hydration, oiliness, pH, and texture), skincare product ingredients, and product application methods can all contribute to this undesirable phenomenon. The findings of this study suggest that sunscreen may be a promoter of pilling, while foundation may resolve pilling in many cases. As there are no prior studies examining the pilling phenomenon, many questions remain on the topic. Further work is needed to delineate our hypothesis that alternating skin physiologies result in changes in pilling experiences. Further work is also needed to identify the ingredients that have potential pilling potential.

CONFLICT OF INTEREST STATEMENT

BLL and LR are employees of Clarins Groupe. YL and SL were employees of Eurofins (Guangzhou) Consumer Product Testing at the time of the study.

Supporting information

Supporting Information

LUA BL, RUAN L, LYU Y, LIU S. Understanding the causes of skincare product pilling. Skin Res Technol. 2024;30:e13828. 10.1111/srt.13828

DATA AVAILABILITY STATEMENT

Supporting data are available in section “Supporting information”.