When the Irrelevant Becomes Relevant: The Impact of Small Talk on Consumer Engagement with Livestreamers

Abstract

<p>Livestreaming is rapidly gaining popularity worldwide, with users and firms finding diverse applications for this technology, including game livestreaming and livestream shopping. The lucrative livestreaming market has attracted an increasing number of livestreamers, leading to intense competition for viewer engagement. The current research investigates how to improve viewer engagement by exploring a novel aspect of streamers’ activities – namely, small talk.</p><p>During a livestreaming broadcast, livestreamers primarily provide information related to the core theme of the broadcast, such as describing a product being sold, or demonstrating skills related to a particular video game. Yet, livestreamers may also include talk that is not related to the primary agenda. Inspired by literature on the contents of talk in the workplace (Holmes 2000), we suggest that such “off-topic” talk can be grouped into three categories. The first category is work-related talk, which includes content that is loosely related to the overall theme but is not strictly necessary (e.g., gaming streamers discuss gaming equipment). The second category is phatic communication, which refers to formulaic greeting and parting exchanges (e.g., “How are you?”). The third category is small talk, which is completely irrelevant to the theme of the broadcast and can include a wide range of topics such as weather, food, movies, personal stories and anecdotes, and expressions of personal attitudes and feelings (Aoki 2010; Peter and Kashima 2007).</p><p>We propose that including small talk in livestreaming can elicit viewers’ perception that they have a communal relationship with the livestreamer. Small talk is generally more characteristic of communal (friendship-like) relationships than of exchange (business-like) relationships, because when a relationship is more intimate, the conversations become broader, leading to more diverse topics that are irrelevant to business (Altman and Taylor 1973). Indeed, the presence and quantity of irrelevant talk in an interaction can indicate intimacy in a relationship (Goodwin 1996). Accordingly, streamers’ small talk may signal their closeness with viewers, leading to viewers’ perception that they are friends with the streamer.</p><p>If small talk in livestreaming can boost perceptions of a communal, friendship-like relationship, it is expected to increase consumer engagement. Friendship is inherently enjoyable and rewarding (Carrier 1999), and is more intrinsically oriented than a business-like relationship (Grayson 2007). Research found that people in a relationship based on intrinsic (vs. instrumental and extrinsic) motives showed more involvement and expressions of positive emotions (Rempel, Homes, and Zanna 1985). In the context of livestreaming, viewers can manifest such feelings and expressions through various engagement behaviors (e.g., liking content, posting live comments, and following the streamer). Therefore, we hypothesize that small talk can enhance viewer engagement, and this effect is attributable to viewers’ enhanced perceptions of having communal relationships with the livestreamer.</p><p>We tested the hypotheses through three studies. Study 1 (pre-registered) used a one-factor (small talk: presence vs. absence) between-subjects design. Participants (<em>N </em>= 236; <em>M</em>_age = 21.45, <em>SD </em>= 2.49; 39% female) were randomly assigned to one of the two conditions. All participants watched the clip of a livestream shopping video where a livestreamer introduced a product. In the small-talk-present condition, the streamer also discussed a personal anecdote – discovering a swallow nest outside her studio. To control for video length, the streamer in the small-talk-absent condition explained how to receive coupons on her livestreaming channel. After watching the video, participants indicated their levels of engagement with the livestreamer (i.e., willingness to follow, like, or preview) and their perceptions of having a communal relationship with the livestreamer (“communality perception”, a 12-item scale adapted from Aggarwal 2004). Results show that a presence (vs. an absence) of small talk significantly enhanced viewers’ communality perception (<em>M</em>_presence = 3.79, <em>SD</em> = 1.02; <em>M</em>_absence = 3.52, <em>SD</em> = 0.99, <em>F</em>(1, 234) = 4.32,<em> p</em> = 0.039, η_p² = 0.018) and engagement (<em>M</em>_presence = 3.02, <em>SD</em> = 1.43; <em>M</em>_absence = 2.63, <em>SD</em> = 1.35, <em>F</em>(1, 234) = 4.64, <em>p</em> = 0.032, η_p² = 0.019). A mediation analysis (PROCESS model 4, Hayes 2022) with 5,000 bootstrap samples supports that viewers’ communality perception significantly mediated the effect of small talk on their engagement (0.22; 95% confidence interval [0.02, 0.43]).</p><p>Study 2 used the livestream shopping data (300 streamers, 14,037 broadcasts) from <em>TikTok</em>. We used panel regressions to test the relationship between small talk and livestreamers’ performances in terms of viewer engagement (e.g., the numbers of live viewers, new fans, live comments, and likes) and product sales (e.g., sales value, sales volume). We included both linear and quadratic terms of small talk to test a possible inverted-U shape relationship between the amount of small talk and performances: small talk has a positive effect on viewer engagement and sales, but an excessive amount of small talk may have negative impacts. Individual streamer fixed effects were included in the panel regressions to control for unobserved heterogeneity. Results showed that on all engagement and sales measures, the linear term of small talks was positive and significant, and the quadratic term was negative and significant, supporting an inverted-U shape relationship between small talk and performances. <img alt="锚点" title="锚点" align=""></p><p>Study 3 used the game livestreaming data (498 streamers, 498 videos) from <em>Twitch.tv</em>. Research assistants coded whether the talk in each minute contained phatic talk, work-related talk, or small talk (as defined in previous sections). We conducted regressions to test the relationship between the proportion of small talk in a video (both linear and square terms) and viewer engagement (i.e., number of live viewers and number of live comments), while controlling various features (e.g., visual variation, voice) of broadcasts. Regression results support an inverted-U shape relationship between small talk and viewer engagement. However, other types of off-topic talk (phatic or work-related) did not significantly affected engagement. Results of a panel regression model with minute-by-minute video data to control for streamers’ unobserved heterogeneity again support a positive effect of small talk on viewer engagement.</p><p>Overall, the present research enriches the marketing literature regarding livestreaming and offers a new mechanism regarding the positive effect of irrelevant information in marketing communications. It also provides practical insights to livestreamers, brand managers, and other stakeholders seeking to engage viewers.</p><p>Livestreaming is rapidly gaining popularity worldwide, with users and firms finding diverse applications for this technology, including game livestreaming and livestream shopping. The lucrative livestreaming market has attracted an increasing number of livestreamers, leading to intense competition for viewer engagement. The current research investigates how to improve viewer engagement by exploring a novel aspect of streamers’ activities – namely, small talk.</p><p>During a livestreaming broadcast, livestreamers primarily provide information related to the core theme of the broadcast, such as describing a product being sold, or demonstrating skills related to a particular video game. Yet, livestreamers may also include talk that is not related to the primary agenda. Inspired by literature on the contents of talk in the workplace (Holmes 2000), we suggest that such “off-topic” talk can be grouped into three categories. The first category is work-related talk, which includes content that is loosely related to the overall theme but is not strictly necessary (e.g., gaming streamers discuss gaming equipment). The second category is phatic communication, which refers to formulaic greeting and parting exchanges (e.g., “How are you?”). The third category is small talk, which is completely irrelevant to the theme of the broadcast and can include a wide range of topics such as weather, food, movies, personal stories and anecdotes, and expressions of personal attitudes and feelings (Aoki 2010; Peter and Kashima 2007).</p><p>We propose that including small talk in livestreaming can elicit viewers’ perception that they have a communal relationship with the livestreamer. Small talk is generally more characteristic of communal (friendship-like) relationships than of exchange (business-like) relationships, because when a relationship is more intimate, the conversations become broader, leading to more diverse topics that are irrelevant to business (Altman and Taylor 1973). Indeed, the presence and quantity of irrelevant talk in an interaction can indicate intimacy in a relationship (Goodwin 1996). Accordingly, streamers’ small talk may signal their closeness with viewers, leading to viewers’ perception that they are friends with the streamer.</p><p>If small talk in livestreaming can boost perceptions of a communal, friendship-like relationship, it is expected to increase consumer engagement. Friendship is inherently enjoyable and rewarding (Carrier 1999), and is more intrinsically oriented than a business-like relationship (Grayson 2007). Research found that people in a relationship based on intrinsic (vs. instrumental and extrinsic) motives showed more involvement and expressions of positive emotions (Rempel, Homes, and Zanna 1985). In the context of livestreaming, viewers can manifest such feelings and expressions through various engagement behaviors (e.g., liking content, posting live comments, and following the streamer). Therefore, we hypothesize that small talk can enhance viewer engagement, and this effect is attributable to viewers’ enhanced perceptions of having communal relationships with the livestreamer.</p><p>We tested the hypotheses through three studies. Study 1 (pre-registered) used a one-factor (small talk: presence vs. absence) between-subjects design. Participants (<em>N </em>= 236; <em>M</em>_age = 21.45, <em>SD </em>= 2.49; 39% female) were randomly assigned to one of the two conditions. All participants watched the clip of a livestream shopping video where a livestreamer introduced a product. In the small-talk-present condition, the streamer also discussed a personal anecdote – discovering a swallow nest outside her studio. To control for video length, the streamer in the small-talk-absent condition explained how to receive coupons on her livestreaming channel. After watching the video, participants indicated their levels of engagement with the livestreamer (i.e., willingness to follow, like, or preview) and their perceptions of having a communal relationship with the livestreamer (“communality perception”, a 12-item scale adapted from Aggarwal 2004). Results show that a presence (vs. an absence) of small talk significantly enhanced viewers’ communality perception (<em>M</em>_presence = 3.79, <em>SD</em> = 1.02; <em>M</em>_absence = 3.52, <em>SD</em> = 0.99, <em>F</em>(1, 234) = 4.32,<em> p</em> = 0.039, η_p² = 0.018) and engagement (<em>M</em>_presence = 3.02, <em>SD</em> = 1.43; <em>M</em>_absence = 2.63, <em>SD</em> = 1.35, <em>F</em>(1, 234) = 4.64, <em>p</em> = 0.032, η_p² = 0.019). A mediation analysis (PROCESS model 4, Hayes 2022) with 5,000 bootstrap samples supports that viewers’ communality perception significantly mediated the effect of small talk on their engagement (0.22; 95% confidence interval [0.02, 0.43]).</p><p>Study 2 used the livestream shopping data (300 streamers, 14,037 broadcasts) from <em>TikTok</em>. We used panel regressions to test the relationship between small talk and livestreamers’ performances in terms of viewer engagement (e.g., the numbers of live viewers, new fans, live comments, and likes) and product sales (e.g., sales value, sales volume). We included both linear and quadratic terms of small talk to test a possible inverted-U shape relationship between the amount of small talk and performances: small talk has a positive effect on viewer engagement and sales, but an excessive amount of small talk may have negative impacts. Individual streamer fixed effects were included in the panel regressions to control for unobserved heterogeneity. Results showed that on all engagement and sales measures, the linear term of small talks was positive and significant, and the quadratic term was negative and significant, supporting an inverted-U shape relationship between small talk and performances. <img alt="锚点" title="锚点" align=""></p><p>Study 3 used the game livestreaming data (498 streamers, 498 videos) from <em>Twitch.tv</em>. Research assistants coded whether the talk in each minute contained phatic talk, work-related talk, or small talk (as defined in previous sections). We conducted regressions to test the relationship between the proportion of small talk in a video (both linear and square terms) and viewer engagement (i.e., number of live viewers and number of live comments), while controlling various features (e.g., visual variation, voice) of broadcasts. Regression results support an inverted-U shape relationship between small talk and viewer engagement. However, other types of off-topic talk (phatic or work-related) did not significantly affected engagement. Results of a panel regression model with minute-by-minute video data to control for streamers’ unobserved heterogeneity again support a positive effect of small talk on viewer engagement.</p><p>Overall, the present research enriches the marketing literature regarding livestreaming and offers a new mechanism regarding the positive effect of irrelevant information in marketing communications. It also provides practical insights to livestreamers, brand managers, and other stakeholders seeking to engage viewers.</p>