Face-to-Face Communication as A Tool to Support Second-Hand Fashion Sales: A Field Experiment at Fashion Week in Berlin

: We conducted a random allocation experiment at fashion week in Berlin in 2017, testing how face-to-face (f2f) communication affects sales of a fashion start-up focusing on second-hand. The experiment revealed that 11% of guests of an f2f event afterwards turned paying customers with an average basket size 11.8% higher than the overall sales event average. We add insights to research on entrepreneurial practice as well as on of ﬂ ine operations in the context of circular consumption in fashion, exposing the leveraging effect of f2f communication for customer acquisition and revenue of start-ups in the ﬁ eld of sustainable fashion.

In Germany, this location is usually considered to be the capital Berlin with its respective bi-seasonal fashion week equal to the more prominent and influencing fashion weeks in Milan, London, New York and Paris.
Finally, the relevance of f2f communication in fashion comes down to the consumer. Prior research indicates that sustainable fashion consumption is hampered by an attitude-behavior gap [22], which may partly be attributed to a gap in knowledge of sustainability pertaining to fashion [23]. Lundblad and Davies [1] report that sustainable fashion consumers intend to be perceived as not following the herd and appearing as individuals in their consumption choices. Thus, f2f communication may play a major role in closing the sustainability knowledge gap and corroborating the individualistic nature of second-hand fashion. We hypothesize that f2f communication is especially valuable when transferring tacit knowledge related to benefits of second-hand fashion as compared to new fashion. Referring to the advantages of f2f communication in communicating sustainability, the instant and open discourse can help consumers to overcome doubts and foster purchase behavior. We hypothesize that f2f communication supports second-hand fashion sales as supporting open exchange.
In order to study the effect of f2f communication on second-hand fashion sales, fashion week in Berlin was chosen as the site of study allowing for a compressed look at the fashion industry and its stakeholders [24].

Overview: Fashion Life Cycles
Bourdieu [25] acknowledged a hierarchy within fashion, with haute couture at the top followed by ready-to-wear (RTW). It depicts the stringent top-down development from runway collections to streetwear. Nowadays, haute couture as the classic catwalk show is less common and only presented in Paris. Retailers increase their receptiveness to the "newness" of fashion trends, trying to provide ever-new product palettes [26,27], leading to a presentation of RTW fashion at most fashion weeks, e.g., Berlin.
A typical fashion life cycle comprises four stages, as delineated in Figure 1 (modified from [28]). A new trend is introduced and adopted by fashion leaders usually via fashion shows. Comprehensively, the experiment in Berlin took place at the first stage. Instead of ending its life cycle by becoming "out" and making space for new fashion trends issued at another fashion week, second-hand fashion defies the regular fashion life cycle. Fashion leaders in terms of fashionable individuals re-introduce already produced and known fashion pieces into the trend life cycle. Bourdieu explicitly refers to vintage fashion as a subcategory of second-hand, challenging fashion hierarchy [25]. It is usually promoted as special and authentic, substituting contemporary fashion, as "a sign of individuality and connoisseurship" [26]. Though consumed nowadays, these items are not newly produced, proposing value of clothing items is maintained after the original production [29]. Second-hand fashion is part of streetwear and defies planned obsolescence of trends. Second-hand fashion is a sustainable offer since product life cycles are maximized and the need for new products diminished. In consequence, an increase in second-hand fashion sales can lead to a decreasing demand of newly produced fashion goods and thus, less resource usage in production.
In its denunciation of mass-design, second-hand clothing represents authenticity [30]. Yet, this inclination for authenticity, expressed through second-hand, has become a broader penchant, resulting in an upgrade from shabby second-hand stores to trendy ones [31].
Originally negligible, the idea of reinventing fashion through second-hand clothing developed into a mainstream phenomenon [32]. Nevertheless, original second-hand fashion, thus not recently produced items, is not represented at fashion weeks. This holds true also for the research environment at fashion week in Berlin. Fashion week in Berlin is a three-day event in January and July, presenting new designer collections and product lines via fairs, runway shows and showrooms/installations.
Although specific events during fashion week in Berlin, such as the Green Showroom, address sustainability issues, no events for alternative consumption patterns occur. This observation of the Berlin agenda is confirmed by other scholars who agree that various issues covering, e.g., labor conditions, environmental impact and other socio-economic topics, are addressed during fashion weeks but there is a lack of general understanding of the interrelation of these challenges [24]. According to Skov [33] the problem lies in the logical discrepancy between the need for prolonged use of fashion items with respect to circular consumption and the fashion logic of always new trends.
Fashion companies dealing with second-hand fashion need to shoulder this double burden. This commercial challenge is often compounded by the sourcing and logistics barriers faced by many companies in sectors of the circular economy that depend on reverse logistics. Against that backdrop, any measure that can alleviate or overcome the commercial double challenge could be relevant to the survival of these companies, including new insights into the applicability of standard business tools such as f2f. As second-hand fashion at fashion weeks addresses a relevant paradox of the fashion industry, VK, the company participating in the underlying experiment, was chosen due to its modern approach to second-hand fashion. This helped ensure that some other typical commercial challenges for circular businesses, such as a diminished consumer experience (e.g., in conventional charity stores or on conventional flea markets), were already addressed. A detailed introduction of the research object, VK, and its historical data follow in the subsequent section.

Data and Methods
This paper bears data from a randomized field experiment in the form of a random coupon allocation at an f2f event. While random allocation is largely used in medical contexts, it achieved its stand in social sciences, also known as A/B testing.
Random allocation is considered "the best method to prove causality in spite of various limitations" [34]. The overall sample is divided into a treatment and control group. This happens by chance, neglecting preferences or the will of both researcher and participants of the study. The easiest method of random allocation is simple randomization, where two entirely random groups within a sample are assigned a treatment or not [34,35]. Usually, the allocation sequence is generated using software to ensure randomization. However, manual methods such as tossing a coin are also possible [35]. Kim and Shin emphasize the importance of a minimum sample size of 100 to avoid unequal assignments and to achieve significant results [34].
Vinokilo (VK) is a second-hand clothing market that operates primarily offline. They acquire their supply from sorting companies all over the world, allowing for attractive prices and original second-hand fashion. VK faces challenges in branding and is to date unaware of what drives customers to participate in its offline model. VK organizes pop-up events in different cities where they sell second-hand clothing by weight. The clothes are leftovers acquired from sorting companies, allowing for attractive prices while maintaining a certain level of quality. Although VK could be considered a flea market, they differ from the common perception of shabbiness due to a modified business concept. Besides second-hand clothing, VK comprises cultural aspects in the form of regional artists and designers who use the events as a launch platform. Accompanied by music, wine and street food, VK sale events bear a resemblance to small parties rather than stereotypical flea markets. As a consequence, not only individuals subscribing to more sustainable fashion-especially the upcycling and recycling mechanisms comprised in circular consumption principles-but also individuals interested in one of the various other aspects of VK's product portfolio and shopping experience are targeted successfully.
In order to be able to benchmark the observed Berlin flea market of VK, historical data was consulted (see Table 1). Event duration (in h) 12 12. 5 16 In 2016, VK hosted 15 events in both Germany and the Netherlands, attracting more than 30,000 visitors. VK achieved six-digit revenue and profitability in its first operational year. Events that differed significantly from the usual concept were omitted in order to achieve a realistic perspective on VK and its performance and a solid data set for further calculations. As especially revenue and average basket size were observed during the experiment, these variables were studied more extensively. A Shapiro-Wilk test [36] was conducted to confirm normal distribution of these variables. The p-values of respective Shapiro-Wilk tests were larger than 0.05 (p = 0.24 for Revenue; p = 0.1 for Average Basket Size). Thus, the null hypothesis that these two data sets are normally distributed could be accepted.
In order to develop a formula to quantify the effect of f2f communication on the overall revenue a multi-linear regression was calculated with R.
The initial formula suggested that five attributes influenced overall revenue. Although this formula accounted for a good fit (Adj. R-sq. = 0.95) only two variables were significant (see Table 2). X4, X5 and X6 were omitted due to lack of significance. The adapted regression formula explained 96% of variation of revenue (see Table 3). For the regression to hold true, data was required to be random and independent and residuals had to be normally distributed, with constant variance. The regression formula was then used to quantify the impact of f2f communication. The underlying field experiment observed basket sizes of coupon users after a private event at fashion week in Berlin. If the average basket size of the coupon users differed from the average basket size, the formula would serve to determine what overall impact such an f2f communication with all customers would have on overall revenue of all historical events. Following the fashion event, a regular flea market sale a few days later was used to test the outcome of the f2f event. The hypothesis of the subsequent experiment was that f2f communication results in incremental sales, i.e., people from the fashion week will become actual new customers of VK and will participate in the following flea market sale.
The main goal was to observe how f2f communication during an event at fashion week in Berlin affects the regular sale and how visitors from the f2f event differ in purchasing behavior. Visitors of the f2f event were "tracked" for the following behavior. Coupons adhering to the common company practice of "one free kg" were distributed in a randomized manner. The random sequence was generated manually by assigning coupons to a goodie bag. At the f2f event, every second goodie bag contained a white coupon of credit card size and the wording, "Gift Card for 1 Kilo". The number of coupons required was fixed to 100, adhering to the suggested significance level. In consequence, it was unambiguous that a person using a white coupon had visited the installation at the f2f event before or at least was connected to someone who had. Randomization in this experiment was correctly implemented, as the probability of getting allocated to the treatment group; thus, receiving a coupon was not correlated with individual characteristics nor non-indicated past transactions. There were no seating orders and all goodie bags looked exactly the same (a) from the outside and (b) through the credit card size of the coupon also at the first look on the inside.
Other field experiments of scholars supported the assumption of a low redemption quota of both on-and offline coupons (e.g., 3% online coupon redemption in [37]); the effect of the coupon was assumed to be neglectable. In their study, Jung and Lee [38] compared redemption rates of printed and electronic coupons. They found that e-coupons had higher redemption rates as compared to offline coupons as well as that absolute discounts led to increased redemption as compared to relative discounts. Thus, in order to decrease the effect of the coupon itself, VK issued the lower performing offline coupon with a relative face-value ("one kilo for free"). These findings are supported by reports for the United States, stating an average redemption rate of coupons delivered in a free standing insert (which constitutes the most common distribution form and the one we employed in the experiment with VK) of only 0.6% [39]. Furthermore, the redemption rate in the non-food segment, the corresponding sector to fashion, was even lower, yielding an average of 0.5% redemption.
For this reason, we can causally link the f2f communication with observed differences in consumer behavior between treated and control groups.
Consult Figure 2 for an overview of the field experiment. Two days after the f2f event, the regular sales took place in Berlin. This proximity to the fashion week both in terms of time and location aimed to avoid the coupons falling into oblivion and to test for immediate effects of f2f communication.
After the regular VK sale, the number of white coupons and respective basket sizes were analyzed in relation to basket sizes of prior events.

Results
The general outcome of the Berlin event was above average as benchmarked to historical data of prior events yet failed to be a significantly positive outlier. Referring to the research question whether f2f communication results in incremental sales and how it affects average basket size, Table  4 can be consulted. The coupon redemption quota was calculated by dividing the number of coupons used at the regular sales event with the number of distributed coupons at the private event. As stated in Figure 3a, 11 coupons were redeemed at the sales event. This equals a redemption rate of 11%. Basket sizes of coupon users ranged from 3.5 to 37.8 Euros.
Basket size (in €) of the overall event was 17.68 as compared to 19.75 Euros for coupon users. This result represents an average basket size increase of 11.8% as compared to the average of all visitors of the Berlin event.  Employing the regression formula deducted above, the difference in revenue as associated with increased basket sizes through f2f activity is displayed in Figure 3b. It was applied both for the Berlin event and the historical event data provided for the business year 2016. While for the Berlin event 1% revenue increases could have been achieved if all visitors had an f2f experience with the brand, the impact of f2f communication considering all events in 2016 was significantly higher. The respective analysis reveals a possible revenue increase by 60.8% through f2f communication (Table 5). Consequently, incremental sales as well as increased basket sizes were observed at the regular sales event of VK. This allowed a confirmation of short-term effects of f2f communication on participation and basket sizes in second-hand fashion sales.
The main hypothesis of this study was that f2f communication with customers in form of an f2f event at fashion week, representing the brand and its core values, will result in incremental secondhand sales.
This hypothesis can be tentatively confirmed, given that the average coupon redemption rate of 0.5% for non-food products [40] was surpassed by 22 times. Furthermore, the guests at the f2f event were not connected to VK before the event. Although the experiment was conducted at a small scale due to the complex and special setting, these results can be confidentially considered potent tendencies for the success of f2f activities for second-hand businesses.

Discussion
Our study confirms premises of f2f communication in regular "linear" sales processes to hold true for second-hand fashion sales. While second-hand purchases are mostly made based on monetary considerations, f2f communication appears to strengthen eco-conscious attitudes. This finding establishes f2f communication as a powerful tool to potentially bridge attitude-behavior gaps in second-hand fashion sales.
As the epitome of f2f communication-one-on-one interactions-is frequently not possible in commercial settings for reasons of time and cost, best practices for addressing larger target groups have developed over time. First, roadshows are considered popular as they can achieve representative insights by enabling a company to cluster f2f communications (e.g., via city) and develop best practices.
Second, and often considered the most efficient measure in terms of effort/value ratio, is participation in fairs. Major uncertainties are outsourced to a third party and costs are clear and predictable. Different target groups can therefore be accessed and their compatibility with the business tested without high financial risk.
Third, f2f activities with customers can take place in the form of organized public or private events. An own event is connected with the most planning effort and uncertainty. The company will be responsible for ensuring enough visitors as well as planning and execution of a program. This can be too resource intense for a young company in terms of money, time and people.
Still, an own event offers great opportunity to advance trust in f2f communication and can aid brand development significantly. It allows a company to shape its brand perception and simulate a social interaction with the brand via employees and atmosphere without requiring one-on-one f2f communication. This form of f2f communication was chosen by VK due to the industry standard and the high control over brand and customer experience [41].
All of these formats are fast to implement and easy to follow-up if connected to a trackable campaign, e.g., via coupon codes or direct on-site registration. Chosen employees of the relevant company will have to be trained in order to portray information adequately and represent the company in a credible manner. This can be done informally or via a small internal task force.
Referring to the introductory comment, f2f activities can have high relevance, particularly for startups, in assessing customer feedback directly, sometimes indicating the need for product or service improvements at the same time as fostering customer acquisition in complex business models.
As noted above, sustainable fashion consumption may be fostered by informed customers where information may relate to core values of the company and about benefits of participation. By including participants in the art installation and making them part of a social interaction with the brand and its employees, visitors of the f2f event were able to assess VK also in terms of associated sustainability dimensions and, consequently, become active customers.
Another opportunity while considering f2f communication is to have events partially sponsored by other brand-supporting companies. The higher the resource efficiency the lower the risk and, thus, the higher the viability of an event that is entirely under the company's control. For small to medium sized companies, this aspect is especially important and has to be taken into consideration adequately. Large fashion corporations are also changing their business operations. One intriguing example is the sports brand Nike. Nike is refocusing its activities from product towards collaborative communities, e.g., reducing their traditional celebrity-starred advertising for about 55% and investing in (virtual) communities like Nike+. An estimated 40% of community members were motivated for their first-time Nike purchase through the shared brand experience at Nike+ [42]. Hence, it can be stated that by interacting, users of Nike+ were more likely to buy Nike products, suggesting future research comparing f2f and virtual interaction in the context of CE and sustainable fashion to devise more scalable solutions.
In summary, f2f communication was a successful measure for second-hand sales at fashion week in Berlin. Eleven percent of coupons were used, resulting in approximately 20 new customers, taking all 200 attendees into account. The increased average basket sizes show the value enhancing in purchase behavior. Particularly in fashion, where f2f communication is indispensable, private events to get in touch with the respective brand and values appear to be very effective in second-hand contexts as well, surpassing expectations and industry standards, i.e., 0.5% coupon redemption.
For practitioners, this experiment shows the potential of offline operations in the context of second-hand fashion and reveals a powerful tool to achieve the relevant shift from purchase intention to action. Considering the often-short-term planning horizon of SMEs, these findings show attractive short-term conversion of customers and respectively incremental sales. Moreover, this study emphasizes the importance of adequate knowledge transfer to the consumer and to support target customers in overcoming entry barriers. While the literature suggests that f2f communication is currently one of the most effective measures to increase trust and facilitate knowledge transfer, the cost factor of f2f activities is an important aspect for its large-scale applicability. Alternatives to own events vary from less expensive tradeshows to roadshows and allow specific regional clustering of customers. Digital solutions such as celebrities on social networks [23] or sustainable fashion bloggers [43] could offer similar benefits at lower expenses. Practitioners as well as scholars are advised to keep an eye on technological development in order to test how digital opportunities can (a) leverage participation in combination with f2f operations and (b) become a viable alternative to f2f communication overall due to ameliorated capacities in knowledge transfer, especially with respect to sustainability-related product characteristics.

Avenues for Future Research
Face-to-face communication can be beneficial for businesses with a circular business model, as demonstrated in this paper. VK, the study object, was not only characterized by its circular character, but also by its unique shopping experience. Further study could indicate the extent to which the character of the shopping experience determines the potential impact of f2f interventions. Similarly, future research could seek to demonstrate how much other types of fashion sustainability (e.g., based on fair labor or a toxics-free production chain) could benefit from f2f communication. Given the characteristics of the purchasers of sustainable fashion, not all forms of f2f may be equally appropriate. Big-budget, lavish events, for example, may not resonate with the lifestyle and buying psychology of sustainable fashion buyers. Further research could aim to find out what type of f2f interventions are most appropriate for second-hand fashion audiences.
Another intriguing aspect is the overall interest in second-hand clothing. While being introduced to Berlin in the contrasting context of fashion week, the Berlin sales event of VK was profitable. This indicates market interest in used and left-over clothing items. Second-hand fashion, if properly communicated, can be a viable and more sustainable alternative to newly produced fashion. However, and this is another avenue for future research, it should be studied if an increase of secondhand fashion sales does in fact have a reducing effect on linear fashion sales or if they are complementary and thus, might even amount to an increase of overall fashion consumption, satirizing the idea of increased sustainability by consuming second-hand fashion.