Influencer fit post vs celebrity fit post: which one engages Instagram users more?

Abstract

Purpose

The purpose of this paper is to compare the effects of the fit between two types of opinion leaders (influencer vs celebrity) and the products promoted in a fashion marketing campaign on user engagement (willingness to like and share the fit posts). Instagram involvement is examined as an applied moderator in the proposed theoretical model of this research.

Methodology

To test the hypotheses of the research, an experimental study was conducted. The sample was formed by female participants who were randomly assigned to one of the two conditions (fit of influencer–product vs fit of celebrity–product).

Findings

The results showed the superiority of the fit of influencer–product over the fit of celebrity–product on users’ engagement. It also indicates that highly involved Instagram users (vs lowly involved Instagram users) moderate the more prominent effect of the fit of influencer–product post on users‘ engagement.

Practical implications

The research helps brands to increase their knowledge of marketing campaigns formed through the fit of the opinion leaders with the products promoted.

Value

This research in addition to highlighting the role of users involved with Instagram contributes to a better understanding of the importance of promotional post fitting.

Citation

Ahmadi, A. and Ieamsom, S. (2022), "Influencer fit post vs celebrity fit post: which one engages Instagram users more?", Spanish Journal of Marketing - ESIC, Vol. 26 No. 1, pp. 98-116. https://doi.org/10.1108/SJME-12-2020-0217

1. Introduction

In recent years, a rising number of companies are progressively looking for opinion leaders (social media influencers and celebrities) to endorse their products through their social media channels. Because customers progressively use social networks to update themselves about brands and products in which they are interested (Casaló et al., 2008, 2011). Instagram, as a sample of the fastest-growing social networks (WebsitePlanet, 2019), is a platform with a higher level of engagement than other social networks, which are extensively applied by brands to promote their products (RivalIQ, 2019). Instagram is also a social media app that permits users to collect followers and their likes, connect with dissimilar brands and entities and simplify social communications among consumers (Blight et al., 2017). It is the most used platform by opinion leaders in the fashion industry (Casaló et al., 2020). These opinion leaders include influencers (Casaló et al., 2020) and celebrities (Jin and Ryu, 2019; Baila and Nicolau, 2018) who have so many fans on social media.

After observing a product, Instagram users incline to perform positive behaviors (Casaló et al., 2021). This happens chiefly in the fashion industry, whose visual nature matches the essence of Instagram (Casaló et al., 2021). For example, 54% of people surveyed say they made a purchase either in the moment or after seeing a fashion product on Instagram (Facebook, 2019). Other important user behaviors can also include their engagement (known as “user behavioral engagement,” Gummerus et al., 2012; Zheng et al., 2015) with the product’s posts such as liking or sharing the posts. Effect of liking or sharing is so important that posts prefer to get more likes or shares than long comments (Pool and Nissim, 2016).

Nowadays, deploying influencers and celebrities for branding has become a defining element of social media marketing campaigns (De Veirman et al., 2017). Influencers vary from celebrities, in that they were born on social networking sites and established their reputations among their followers through the behaviors they performed on these platforms (Schouten et al., 2019). Whereas, the renown appreciated by celebrities frequently results from their activities outside social networking sites (e.g. TV, music, sports). Some previous studies stated that influencers compared to celebrities can develop closer ties with their audiences, and their behaviors can be more perceived as valuable by their audiences; they encourage them and support them to discover new brands/products that, in the end, the audiences might purchase or recommend (Djafarova and Rushworth, 2017; Rakuten Marketing, 2019).

Considering the importance of marketing campaigns on social media, especially Instagram, comparing the effectiveness of the promotional posts through opinion leaders (influencers and celebrities) is essentially valuable. Although previous studies have focused on the metrics applied to assess the impact of influencers’ behaviors marketing campaigns (Arora et al., 2019; Tafesse and Wood, 2021; Boerman, 2020; Jiménez-Castillo and Sánchez-Fernández, 2019; Casaló et al., 2020), studies on comparing the effects of influencers and celebrities on customers’ online behaviors are limited, and only a few studies have recently been conducted in this area (Schouten et al., 2019; Jin et al., 2019). Regarding the importance of this phenomenon, more research is essential to understand the effectiveness of marketing campaigns in terms of their effect on customers’ online behaviors.

The significant research issue that must be concentrated on is what explains the comparison of influencers and celebrities in recent years. To answer this question, we ran an experiment associating how users react differently to fit posts containing the same product endorsed by an influencer vs a celebrity on the Instagram platform, and evaluated their effects on different variables. Hence, the current research aims to address this gap. By way of the aim of going into better profundity in this research domain, the present study examines the effect of opinion leaders’ fit posts on individuals’ engagement with the brand posts. More specifically, the research tests how the fit posts of a fashion influencer (vs fit posts of a fashion celebrity) can have a greater effect on users’ engagement (i.e. willingness to like and willingness to share the product posts). Marketers typically use different activities to motivate Instagram users’ willingness to such positive reactions. This motivation in user willingness to like or “share” a promotional post can lead to the success of the post on Instagram.

The research model also includes a moderating variable. We explore how Instagram involvement moderates the effect of opinion leaders’ fit posts on user engagement. The research findings reveal the effect of congruity in the content published by influencers (vs celebrities) cooperating with brands. Brands and fashion companies can use the research findings to develop effective influencer marketing proceedings that lead to positive reactions of users.

2. Literature review and hypothesis development

2.1 Overview of the fit posts

Fit usually means congruency, resemblance, relevance or consistency between the endorser and the product or brand (Kamins and Gupta, 1994). Several studies propose that a good match-up between opinion leader images and product images should result in a more positive perception of the advertisement, the opinion leader and the product than a poor match-up would (Belanche et al., 2021; Choi and Rifon, 2012; Kamins and Gupta, 1994). The particular brand management literature often communicates the personality of a brand (Aaker, 1997) with the image that individuals have of themselves, i.e. their self-image (Achouri and Bouslama, 2010). In particular, individuals search for congruence between the characteristics they connect with their own personality, or self-image, and the characteristics that connect with the brand image (Belk, 1988). In this way, customers state their personalities through the products and brands they consume (Phillips, 2003). On social networks, just as individuals positively assess the fit between their personalities and the personalities connected with the brands and products they use, it appears reasonable to imagine that they will also positively value the congruence between the characteristics connected with others’ personalities and the products/brands they promote. In summary, by generalizing this type of association, followers will positively value the fit between the characteristics connected with influencers/celebrities and those of the brands or products they promote in their posts (Schouten et al., 2019). In this study, the fit of promotional posts is examined and explained as the degree of similarity that exists between the characteristics connected with influencers/celebrities and the products and brands promoted on their accounts. The previous investigation on communication has found out that a commercial post positioned in a congruent (vs incongruent) context upsurges its effectiveness (Gunter et al., 2002). In an online setting, the customer assessment of the post expanded if the post has the congruency between the opinion leader image and the branded product.

Based on the priming theory (Iyengar et al., 1982) and corresponded with the spreading activation model (Collins and Loftus, 1975), individuals’ minds move through a network of unified ideas, which helps the processing of thoughts that fit with their knowledge construction (Janssens et al., 2012). If viewers’ prospects in terms of the content of posts are satisfied, they are willing to welcome and process the information in messages (Belanche et al., 2017). Furthermore, users perceive more positively commercial messages when there is a thematic similarity between the message and the context in which it is set (Moorman et al., 2002). Instead, when individuals perceive that information does not match their preceding perceptions or beliefs, the information is frequently discarded due to selective perception bias (Klapper, 1960; Das and Teng, 1999). In this condition, it is less expensive to discard, and thus not process, inconsistent information than to reinterpret reality to adapt it to the new information (Klapper, 1960). Given these points, it is reasonable to assume that an appropriate post published by an opinion leader will increase the behavioral interaction with the opinion leader (liking and sharing the promoted product post). Besides, when there is congruence between an endorser and the product which are set, customers perceive promotional posts more positively (Xu and Pratt, 2018). Inline, an absence of fit between a post and an opinion leader will decrease the likelihood of a customer performing these engagements. In this case, the perceived non-fit between the content and the opinion leader will undoubtedly inspire the user to discard the information.

2.2 Effect of the fit posts on user engagement

Likes are a way to give positive feedback or to connect with things people care about on social networks such as Instagram. For brand marketers, consumers’ likes not only enhance brand connections and customer engagements but also benefit them through online brand endorsement (Bernritter et al., 2016; Hoffman and Fodor, 2010). Sharing promotional posts by users are also so important that provide companies a competitive advantage, as it upsurges their odds of selecting their brand over other competitors, whether they are entering a well-developed business or a start-up, they erect higher odds of increasing customer retention with promotional products (Tafesse and Wien, 2018). Sharing a post means that Instagram permits users to send their favorite posts to their page or story, or share them with their followers or other social networks (Later, 2019).

Based on the previous research, influencers do not have a wealth of cultural meanings celebrities can draw from what they have shaped in mainstream media (Jin et al., 2019). Even, whatever symbolic value influencers have is based on the personal brands they have shaped through Instagram posts and stories, the events they have been capable to attend and the brands they have connected themselves with. Users who follow an influencer (Casaló et al., 2017a, 2017b) are usually sensitive to the fit of products promoted by an influencer in his/her posts. As such, that a fit post affects their intention to look for more information about the product promoted (Schouten et al., 2019).

Previous research has revealed that the effectiveness of an endorsed advertisement is inseparably tied to the degree to which the image, personality or expertise of the endorser fits the advertised product (Kamins, 1990; Kamins and Gupta, 1994). Multiple studies have verified that when an endorser’s perceived expertise matches the product endorsed, this enhances product evaluation and purchase intention (Fink et al., 2004; Till and Busler, 2000). If influencers vs celebrities are more using their skills, knowledge and experience to continually generate new content to become opinion leaders and attract a broad follower network (De Veirman et al., 2017), and even, messages attraction can be much greater if celebrities and professional models are not used (Schouten et al., 2019), then a good fit between the endorser and the product could, therefore, be of higher importance for influencers than for celebrities.

For influencers, endorser relevance to the product is mentioned by consumers as important to endorsement success as well (Djafarova and Rushworth, 2017). Indeed, the effect of product–opinion leader fit on user engagement may be even more pronounced for influencers than for celebrity endorsers. As influencers have successfully branded themselves as representative of a specific domain of interest, such as “beauty vlogger,” and regularly share product information with their followers (Balog et al., 2008), an associative connection between a product and opinion leader may be more easily created. In addition, as compared to celebrities, influencers may be more likely to be frowned upon when they endorse products that do not fit their specific “niche” specializations and may only be perceived as credible information sources in endorsements that fit well with their domain of interest (Schouten et al., 2019). The positive nature toward a more recognizable influencer fit post would result in more enthusiasm to like or share the post of promoted products (Schouten, et al., 2019). With these tips in mind, it is a reasonable assumption that a fit post published by an influencer will more increase users’ willingness to like and share the promoted products (i.e. fit of influencer and product); thus, we propose the following hypotheses:

H1.

Compared to a celebrity fit post, an influencer fit post has a greater effect on the user’s willingness to like the fit post.

H2.

Compared to a celebrity fit post, an influencer fit post has a greater effect on the user’s willingness to share the fit post.

2.3 Moderating effect of Instagram involvement

Instagram involvement contains Instagram usage duration, several followed profiles and number of profiles followed (Wnent, 2016) and other activities such as liking, commenting or posting (Pixlee.com, 2018). Previous investigations on social influencer advertising have given little consideration to individual differences that may explain the impacts of social influencer advertising (Tiggemann et al., 2018; Amornpashara et al., 2015). A prior study has distinguished several types of social media users, which are categorized in terms of their behavior on social media (Krasnova et al., 2013). Highly involved social media users are generally people who engage in social media actively (e.g. posting, liking and commenting). They are users who have been using Instagram for a long period of time in an orderly and energetic mode (Nijs, 2019). Whereas, lowly involved users are individuals who only view the content and are not actively taking part. Because of further activities of highly involved Instagram users, it can assume that they have more impact on sharing the brand/product posts endorsed by the Instagram influencers. On the other hand, a post published of a brand/product by an Instagram influencer is expected to be less important for lowly involved Instagram users.

Highly involved Instagram users exposed to Instagram influencers’ brand/product posts perceive the post source to be more valuable, show a more positive attitude toward the endorsed brand and are more attracted to like or share the fit posts. They have a stronger connectedness toward Instagram influencers than traditional celebrities (Tran and Strutton, 2014) and perceive them as more authentic and attractive (Stefanone et al., 2010). The perceived reality of those non-traditional stars and the deep connectedness leads to greater purchase intention of the products they endorse because consumers personally identify with them and try to emulate them (Tran and Strutton, 2014). Furthermore, if highly involved Instagram are users who are more familiar with Instagram influencers and they engage more in social media actively (e.g. more liking, commenting or sharing), we expect that they will be more enthusiastic to spread the published posts of an Instagram influencer than a traditional celebrity (Schouten et al., 2019).

In the advertising context, it has been noted that greater involvement motivates the adoption of a more attentive state of mind, which allows the consumer to better process information (Gunter et al., 2002; Yoo et al., 2004). Thus, users who are more involved with the promoted product by an Instagram influencer pay more attention to commercial messages and process them more intensively. Accordingly, they will be more enthusiastic to like, or spread, the message fit of an influencer than a celebrity (Schouten et al., 2019). In addition, when users are involved, they feel less irritated when presented with commercial information (Edwards et al., 2002). On the other hand, individuals with low involvement with Instagram often pay less attention to commercial messages and are unwilling to make great efforts to process any information received. Thus, when there is a fit post published by an influencer, greater involvement among her/his followers increases their interest and the attention they pay and, consequently, their intention to like and share the published fit post. Hence:

H3.

Instagram involvement moderates the superior effect of influencer fit compared to celebrity fit on willingness to like the fit posts, such that this effect will be greater for highly involved compared to lowly involved users.

H4.

Instagram involvement moderates the superior effect of influencer fit compared to celebrity fit on willingness to share the fit posts, such that this effect will be greater for highly involved compared to lowly involved users.

The proposed research model and summarized constructs of this study are presented in Figure 1.

3. Method

3.1 Overview of study

We focused on the fit posts of the popular fashion opinion leaders (the fit post of influencer vs the fit post of celebrity) on Instagram, as the most prototypical example of influencer marketing (Chun et al., 2018). The sample was shaped only by women, which is appropriate for our investigation context and frequent practice in fashion marketing studies (Michon et al., 2008). We manipulated the type of post by creating the post promoted with either a celebrity or influencer. The fit was manipulated by pairing the opinion leader with the same product that fit or did not fit with the opinion leader’s profession. The product was selected to be experience goods as its product characteristics (such as quality) are difficult to determine pre-purchase, so people tend to rely more on the evaluations of others (Park and Lee, 2009).

3.2 Participants and design

The female undergraduate students from a large university in Thailand participated in a two-cell, one-way (type of the fit: the fit of influencer–product vs the fit of celebrity–product) between-subjects design for a small gift (candy bar). They were randomly assigned to one of the two conditions. Following Casaló et al.’s (2020) research, we selected a specific influencer account with a large and increasing number of followers, concentrated on the fashion industry, who had not previously enjoyed celebrity status (e.g. in contrast to singers and sportspeople). We also ran the same procedure to choose a celebrity with growing popularity in the media. Both accounts have more than one million followers. A high number of Instagram users helps not only ensures ecological isomorphism and a reasonable level of external validity but also controls the level of quantitative Instagram popularity across conditions (Jin et al., 2019). The influencer account belongs to a beautiful young Thai woman (Archita Siri) who is well known in Thailand. The celebrity account belongs to another young Thai woman (Kannarun Wongkajornklai (also known as Prang Kannarun) who was a famous Thai actress and model. Both opinion leaders were selected based on our own knowledge of a celebrity and an influencer in the fashion industry, and a Google search showed the popularity of the influencer and the celebrity in the fashion industry. Of our participants, 258 (92.5%) of the respondents (Mage = 28.44 years, SD = 1.06) were completely familiar with both influencer and celebrity. Thus, we continued the survey with this amount of respondents.

Respondents were shown two fit posts promoted for a hair care product (shampoo), two promotional fit posts for a skincare product (sunscreen) and also two fit posts promoted for a cosmetic product (face powder). The six manipulated fit posts used in the experiment consisted of existing advertisements of the hair care product, skincare product and cosmetic product, combined with the picture of the opinion leaders. The first fit post consisted of a picture of the celebrity who had advertised a shampoo (hair care product), the second fit post included the picture of the celebrity who advertised a sunscreen (skincare product) and the third fit post comprised the picture of the celebrity who advertised a cosmetic product (face powder). In the influencer fit conditions, we considered the same procedure as such in the first fit post, the picture of the influencer was combined with a hair care product, in the second fit post, the picture of the influencer was amalgamated with a skincare product and the third fit post the picture of the influencer was joined with a cosmetic product.

3.3 Pretest

For the posts of the products promoted, the influencer’s and celebrity’s good-fit condition were selected based on a pretest (n = 116) to examine influencers and celebrities could serve as good fits with the selected products. For each opinion leader, we considered three products. If the manipulated images of opinion leaders with the same product acquired the highest score, we entered into the next step. But, if the manipulated pictures of opinion leaders with the same product could not gain the highest score, we had to test the other products.

The participants indicated the perceived fit of the presented celebrity or influencer with a hair care product, the perceived fit with a skincare product and the perceived fit with a cosmetic product on seven-point scales (1 = totally disagree, 7 = totally agree). The respondents were more agreed with the fit of the influencer with a hair care product (M = 5.93, SD = 0.997) than the fit of the celebrity with a hair care product (M = 5.12, SD = 1.24). They also were more agreed with the fit of the influencer with a skincare product (M = 3.74, SD = 1.58) than the fit of the celebrity with a skincare product (M = 3.32, SD = 1.95). However, the respondents were less agreed with the fit of the influencer with a cosmetic product (M = 3.97, SD = 1. 34) than the fit of the celebrity with a cosmetic product (M = 4.84, SD = 1.16).

For the fit of opinion leaders with the hair care product (M = 5.47, SD = 1.08), the skincare product (M = 3.51, SD = 1.39) and the cosmetic product (M = 4.92, SD = 1.39), the data confirmed that the promotional posts with hair care–opinion leader fit were indeed considered a better fit than the promotional posts with skincare–opinion leader fit, t_haircare (114) = 12.33, p < 0.001; t_skincare (114) = 5.72, p < 0.001 and the promotional posts with cosmetic–opinion leader fit, t_haircare (114) = 10.56, p < 0.001; t_cosmetic (114) = 6.93, p < 0.001. As the fit of influencer and celebrity with the hair care product reached better and the highest fit score, we considered the fit of opinion leaders with the hair care product for the main experience (Figure 2).

3.4 Procedure and measure

All respondents were presented with two stimuli. They were given the promoted posts in an Instagram frame (i.e. manipulated post promoted by the celebrity and the influencer). First, respondents were presented with the celebrity’s post, and afterward, they were shown with the influencer’s post. After each of which they filled in a questionnaire. Respondents then assessed the fit between the influencer and the endorsed product (α = 0.93) and also the fit between the celebrity and the endorsed product (α = 0.95) through three items in the basis of “matching,” “compatibility” and “alignment” (1 = “strongly disagree” to 7 = “strongly agree”; adapted from (Xu and Pratt, 2018). To evaluate Instagram involvement, we used several items altered from previous studies (Tiggemann et al., 2018; Ting, 2014; Ellison and Steinfield, 2007) to create an understanding of participant’s activities on the platform (i.e. “I spend a lot of time in a week using Instagram,” “The number of my followers is noticeable,” “I follow many people during one week,” “I give many likes and comments for someone’ posts on Instagram, weekly,” “I feel out of touch when I have not logged onto Instagram for a while” and “I feel I am part of the Instagram-community”; α = 0.89). The scale consisted of seven points, ranging from 1 = totally disagree to 7 = totally agree. To make a comparison between participants based on their involvement with Instagram, we split the total sample (n = 258) into high-and low-involvement groups by splitting the sample at the mean value of Instagram involvement (5.38). After splitting the total sample, the sample size of the high-involvement group was 137 and also that of the low-involvement group was 121. Respondents evaluated willingness to like through three items modified from Bernritter et al.’s (2016) and Halaszovich and Nel’s (2017) study (e.g. “I would like to give a like for this post promoted,” “I would like to click the like button below the product post to show my positive view of the product,” “I would like to click the like button below this post to recommend the fashion product to my Instagram friends”; 1 = “totally disagree,” 7 = “totally agree”; α = 0.91). Willingness to share was also measured by three modified items, which are extracted from Stewart et al.’s (2018) and Yang’s (2011) study, for example, “I predict I will share the post about this product shortly” (α = 0.93). Respondents finally registered their demographic profiles.

All scale items correlated positively and loaded on a single factor. Composite reliability and the average variance extracted (AVE) were calculated. All the scales indicated values above 0.7 on theAVE, which are consistent with the recommendations (Hair et al., 1995). Discriminant validity is evidenced by the fact that all correlations between the constructs are significantly smaller than 1 and the squared correlations computed for each pair of constructs is always smaller than the variance extracted for corresponding constructs (Fornell and Larcker, 1981), thereby confirming the discriminant validity (Table 1).

3.5 Manipulation check

To a manipulation check, we first ran an independent samples test to evaluate that whether the person appearing in the post is perceived as an influencer or a celebrity. The analysis disclosed the superiority of influencer to celebrity (M_inf = 6.02, M_cel = 5.14; t [256] = 16.48; p < 0.01), representing the achievement of the manipulation. To check manipulation of the fit of influencer–hair care product vs fit of celebrity–hair care product manipulation, an independent samples test once more was used. The examination showed the superiority of fit of influencer–product to fit of celebrity–product (M_finf = 6.33, M_fcel = 5.26; t [256] = 18.42; p < 0.01), demonstrating the accomplishment of the manipulation.

4. Results

4.1 Effect of main variables

To examine the direct effects of types of the fit posts (fit between the influencer and the product vs fit between the celebrity and the product), we conducted an independent samples test. The analysis’s results showed that the fit of influencer–product has a greater effect on user willingness to like than the celebrity–product (t[256] = 8.53, p < 0.01, M_finf = 5.44, M_fcel = 4.62); therefore, H1 was supported. Besides, the findings revealed that the fit of influencer–product has a better effect on user willingness to share than the celebrity–product (t[256] = 9.18, p < 0.01, M_finf = 5.69, M_fcel = 4.37), indicating H2 is supported.

4.2 Effect of Instagram involvement

To test the moderating effect of Instagram involvement proposed in H3, we conducted a 2 (type of the fit post: influencer fit vs celebrity fit) × 2 (high vs low Instagram involvement) ANOVA on willingness to like the fit posts. The examinations indicated a significant interaction effect between the fits and involvement on the willingness to like the fit posts F [2,256] = 8.44, p < 0.01, ηP2 = 0.02). The results also showed that respondents who have high involvement with Instagram are more willing to like a fit post of the influencer than individuals who have low involvement with Instagram (M_{highinvolvement} = 4.88 vs M_{lowinvolvement} = 3.75; F (1, 256) = 7.20, p < 0.01, ηP2 = 0.01). On the contrary, respondents who have low involvement with Instagram are more willing to like a fit post of the celebrity than individuals who have high involvement with Instagram (M_{highinvolvement} = 3.19, vs M_{lowinvolvement} = 4.34; F (1, 256) = 6.58, p = 0.12, ηP2 = 0.02). These results support H3.

To examine the moderating effect of the variable (Instagram involvement) proposed in H4, we once more ran a 2 (type of the fit: influencer fit vs celebrity fit) × 2 (high vs low Instagram involvement) ANOVAs on willingness to share. The results revealed a significant interaction between the fits and Instagram involvement on the customer’s willingness to share the post (F [2,256] = 9.38, p < 0.01, ηP2 = 0.01). The findings also indicated that individuals who have high involvement with Instagram are more willing to share a fit influencer–product post than individuals who have low involvement with Instagram (M_{highinvolvement} = 4.73 vs M_{lowinvolvement} = 3.59; F (1, 256) = 7.84, p < 0.01, ηP2 = 0.01). By contrast, individuals who have high involvement with Instagram are less encouraged to share a fit celebrity–product post than individuals who have low involvement with Instagram (M_{highinvolvement} = 3.38, vs M_{lowinvolvement} = 4.62; F (1, 256) = 6.34, p < 0.01, ηP2 = 0.02). On the basis of these findings, H4 is supported. The interaction effect between opinion leaders–product fit and Instagram involvement on users’ engagement with the brand posts is presented in Figure 3.

5. Discussion

The current study aims to compare the fit of two types of social media opinion leaders (influencer vs celebrity) with the products promoted in their posts on the Instagram environment and tests their effects on users’ engagement. In the other words, the study examines a comparison of the effects of the fit of the influencer with the product promoted vs the fit of the celebrity with the product promoted on users’ engagement (willingness to like the fit post and willingness to share the fit posts) on Instagram. The study’s hypotheses have been tested in one experience, using a celebrity and an influencer who each advertises a beauty product.

The results of the research showed that users’ willingness to like and share the promoted fit post is affected more through influencer–product fit than the celebrity–product fit. The reason behind this is that users more interact with influencers. People feel more similar to influencers than celebrities and as a result like their endorsements more (Schouten et al., 2019) and are more inclined to share their endorsed products. Additionally, the former investigations revealed that celebrities are perceived as less trustworthy than influencers, because of people’s skepticism (De Veirman et al., 2017; Obermiller and Spangenberg, 1998). Thus, another valid reason could be in line with earlier studies that stress the role of more authenticity and trust in influencer endorsements (Uzunoĝlu and Kip, 2014) that can be another confirmation of the superiority of influencers compared to celebrities in encouraging Instagram users to like or share the fit posts. Another case that can be pointed is that each influencer is recognized as a promoter of a particular class of product (e.g. fashion products). Therefore, as they are considered experts in a particular area of knowledge (Rahman et al., 2014), users follow them more because they want to keep up to date with developments there. Therefore, they share more such promotional fits (a fit between an influencer and product), as a way to inform others (and show that they are up to date), define their identity through sharing and maintain (and build) relationships.

The findings of the study testing the moderating effect of Instagram involvement showed that Instagram involvement plays an important role in determining the customer’s reaction toward the posts. Users who are more highly involved (vs lowly involved) with Instagram tend to develop more positive behavioral intentions toward the products promoted by the influencer. These results are logical, given that users highly involved with Instagram tend to have more positive attitudes toward commercial campaigns developed through influencers (Nijs, 2019). This degree of involvement of an internet user even affects the attitude to the advertisement, brand and purchase intention (Yang, 2011). Therefore, highly involved Instagram users have a greater positive impact on their subsequent behaviors toward the fit post advertised by an influencer (vs a celebrity), and further reinforce the influences of campaigns shaped through a fit between an influencer and product.

5.1 Theoretical implications

The research provides several theoretical implications. First, this study adds to the literature about social media marketing and, in particular, about customers’ reactions toward influencers’ posts on social networks (Casaló et al., 2020). Previous research has focused mainly on the role of social media influencers and celebrity endorsements in promotions (Schouten et al., 2019), or out of the subject of fit posts have been investigated (Jin et al., 2019). Thus, given that influencer-based marketing activities on social networks are increasingly commonplace, this research tries to clarify this phenomenon to extent a deeper understanding of how brands might collaborate with social media influencers to promote their products. In this respect, this investigation underscores that the fit between the influencer (vs the celebrity) and the products promoted include more crucial differences in developing positive behavioral intentions in marketing campaigns. The findings of the study have shown that influencers are deemed more effective than celebrities, and users are more engaged with influencers’ fit posts. They are more willing to like and share a fit post of the influencer with the products promoted in his/her posts. Therefore, the second implication of the study supports the previous study that explains influencers can be more effective product endorsers than celebrities in marketing campaigns (Schouten et al., 2019).

The third implication is investigating the role of Instagram involvement. It is an important explanation for why users who are highly involved with Instagram are more engaged with the influencers’ advertising posts than celebrities’ advertising posts. Due to frequent Instagram use, highly involved Instagram users are probably more confronted with influencers and their advertisements (Tran and Strutton, 2014). Consequently, they are more familiar and involved with the influencers than lowly involved Instagram users (Nijs, 2019). Lowly involved Instagram users, in contrast, are possibly less familiar with influencers and their efficiency in promoting the products and care less about the posts advertised by influencers. In particular, presenting products that match the influencer’s style is a marketing action more effective for engaging users highly involved with Instagram and for prompting them to like and share the posts.

5.2 Managerial implications

Some management implications are provided for this article. Given that, mainly in the case of a product post advertised on Instagram, users’ behavioral engagement is established through their active participation in the functionalities Instagram presents such as liking and sharing the post, the first implication is that companies can apply the fit posts to motivate users’ willingness to such positive reactions. Though, in shaping such posts using influencers are more effective than celebrities. Publishing fit posts of influencers with products engages more users to like and share such posts, which results in obtaining valuable feedback from the community (Algesheimer et al., 2005).

Nowadays, influencer marketing is progressively becoming a requirement for marketers. These influencer marketing actions have been extensively implemented on Instagram, whose visual nature creates the platform of choice for marketers (Relatabe, 2019). However, in influencer marketing campaigns, the fit of the influencer with the products promoted has a great effect on users’ reactions. Thus, the second recommendation for companies is that collaborating with influencers can better help create online buzz about companies’ products. A fit post of influencer can strengthen their brand’s reputation, improve audience engagement and increase conversions. Companies or brand owners should undertake prior market research studies to select the most appropriate influencers, individuals whose style, themes and orderly content correspond to the products they want to promote in their influencer-based campaigns (Casaló et al., 2020).

The next managerial implication is that brands need to gain insight into and invest in their users on Instagram to see if they are lowly or highly involved with Instagram. Nowadays, organizations invest heavily in optimizing their customer database, knowing who their customer is and how they behave online (e.g. traffic analysis on websites). Remarkably, organizations do nearly nothing to retrieve customer information on Instagram. This undiscovered field of crucial information needs to be exploited. When a brand wants a good advertising strategy, it is detrimental that they learn more about their followers on Instagram and respond to it accordingly.

5.3 Conclusions

The conclusions section includes a brief explanation of the research findings along with the theoretical and managerial implications, displaying in Table 2.

6. Limitations and the direction of future research

Some limitations are existing in this study that can be the way to the expansion of future research directions. First, the paper only focused on fashion products in line with previous studies (Michon et al., 2008; Jin et al., 2019) into Instagram (Phua et al., 2018). For it is becoming a pervasive phenomenon in social media marketing, particularly in visual platforms like Instagram (Jin et al., 2019). Although fashion is regarded as the main industry for performing influencer marketing actions (Klear, 2018), examining other industries would enrich the analysis of the effect of congruity on influencer marketing campaigns. In addition, as various social media apps have affordances and spirits different from what Instagram offers, future studies can also consider other platforms (Barta et al., 2021). Second, the study was conducted by comparing just two opinion leaders (an influencer and a celebrity) who despite being well known in the country of origin of the participants may have some particularities and special features in comparison to other similar opinion leaders. Therefore, to generalize the results, future research should replicate the study with other opinion leaders. Third, this research used pictures on Instagram to manipulate the levels of fit between the opinion leaders and the fashion products. However, Instagram now includes several formats (e.g. video) that opinion leaders use to promote products, so future research might compare the effectiveness of these different formats in terms of fit in influencer campaigns. The last limitation and future research direction for this study are that our sample was made up of women. Nevertheless, future research should include a gender-balanced sample to generalize the results and analyze if gender might play a moderating role in this research context.

Figures

Figure 1.

Proposed research model

Figure 2.

The six manipulated fit posts

Figure 3.

Interaction effect between opinion leaders–product fit and Instagram involvement on willingness to like and willingness to share the fit posts

Table 1.

Composite reliability, the AVE and item-total correlations of per construct

			Correlations
	CR	AVE	1	2	3	4	5
1. Fit of influencer	0.925	0.864	0.912	0.734	0.558	0.437	0.381
2. Fit of celebrity	0.947	0.862	0.725	0.904	0.702	0.424	0.348
3. Instagram involvement	0.958	0.857	0.543	0.698	0.892	0.532	0.264
4. Willingness to like	0.926	0.814	0.419	0.407	0.516	0.833	0.517
5. Willingness to share	0.943	0.872	0.308	0.295	0.258	0.495	0.889

Notes:

Diagonal entries (in italic) are the square root of the AVE; values above the diagonal indicate the HTMT ratio; values below the diagonal entries are the inter-construct correlations

Table 2.

Conclusions, theoretical and managerial implications

Conclusions	Theoretical contributions and managerial implications
Influencers’ fit posts are deemed more effective than celebrities’ fit posts in engaging Instagram users with the fit posts	The research extended a deeper understanding of how brands might collaborate with Instagram influencers to promote their products. Influencers can be more effective product endorsers than celebrities in marketing campaigns on Instagram A fit of influencer–product can reinforce companies’ brand reputation, expand audience engagement and upsurge conversions
Highly involved Instagram users have more efficiency in promoting the fit posts advertised by influencers	Presenting products that match the influencer’s style is a marketing action more effective for engaging users highly involved with Instagram Companies should develop their database, knowing who their customer is and how they behave. When a brand wants a good advertising strategy, it is detrimental that they learn more about their followers on Instagram and respond to it accordingly