In-app advertising: a systematic literature review and implications for future research

Purpose – The purpose of this study is to review and synthesize the literature on in-app advertising, identifygaps andpropose future researchdirections. Design/methodology/approach – The authors use a systematic literature review (SLR) approach, following the PRISMA guidelines, to investigate the current state of research in in-app advertising. The study uses 44 shortlisted articles from the Scopus and Web of Science databases. Using the Theory-Context-Characteristics-Methodology (TCCM) framework, the authors analyze the gaps in theory, context, characteristics andmethods. Findings – Using thematic analysis, the authors identify ﬁ ve main themes in the in-app advertising literature, namely, ad platform optimization; mobile app user psychology and behavior; ad effectiveness; ad fraud; and security, privacy and other user concerns. The ﬁ ndings show the need for empirical research, with a strong theoreticalfoundation inemerging adformats ofin-app advertising, user behaviorandbuy-side of in-appadvertising. Originality/value – This is a maiden study to conduct a domain-based SLR in the emerging ﬁ eld of in-app advertising using the TCCM framework. The authors highlight the key differences between in-app advertising and mobile web advertising. The authors propose theories in the advertising ﬁ eld that could be used in futureempirical studies of in-app advertising.


Introduction
The exponential growth and penetration of mobile (6.5 billion smartphone users globally as per Statista, 2022) have bestowed unique opportunities to use it as an advertising medium.Lately, the focus of mobile marketing has moved from the mobile web to mobile apps (Keem and Lee, 2018), and in-app advertising has emerged as a new conduit for mobile advertising.The Interactive Advertising Bureau (2022) defines in-app advertising as "ads and ad campaigns that are delivered within mobile applications, including smartphones, tablets, or wearable devices".Sensor Tower has estimated that global in-app advertising spending will reach US$233bn by 2026 (Chan, 2022).Mobile apps are popular, as 97% of the apps are free to download (Statista, 2023), easily accessible round the clock and gratify a range of specific In-app advertising user needs (Sigurdsson et al., 2018).In-app advertisements have become an important source of revenue for app developers (Gao et al., 2022;Mattke et al., 2021;Truong et al., 2019).
However, researchers posit that there is little understanding of the differences between the two mobile channels: mobile web and mobile apps (Park and Park, 2020).Mobile apps are considered catalysts for "new" customer experiences and engagement, delivering a unique source of customer value.The "always on" points of interaction in mobile apps shape positive and interactive customer journeys (Stocchi et al., 2022).Mobile apps are developed for a particular operating system, whereas mobile websites are delivered through smartphone browsers.According to e-marketer, the in-app advertisement viewability rate exceeds mobile web and desktop viewability worldwide (Handel, 2022).The key differences between mobile web and in-app advertisements are summarized in Appendix 1.Compared to mobile web advertisements, in-app advertisements have a captive and more engaged audience.Targeted and personalized advertisements, with higher accuracy, can be placed due to the availability of location, demographics and app content.
Mobile advertising exists in many forms, such as web-based, mobile search, in-app, crossapp, location-based and social media (Jebarajakirthy et al., 2021).Maseeh et al. (2021), in their recent meta-analysis of mobile advertising research, avow the underrepresentation of empirical studies on in-app advertising in the literature.Although in-app advertising has been celebrated for revolutionizing the advertising landscape, scholarly attention on in-app advertising is limited when compared to other domains of mobile advertising (Truong et al., 2019).
Although there are a few comprehensive systematic literature reviews (SLRs) on mobile marketing, in general (Leppäniemi et al., 2006;Narang and Shankar, 2019;Varnali and Toker, 2010) and specifically on mobile advertising (Billore and Sadh, 2015;Grewal et al., 2016;Jebarajakirthy et al., 2021;Maseeh et al., 2021), there is limited discussion on issues in in-app advertising.Previous reviews in the domain have focused on the broader perspective of mobile advertising and a summary of literature reviews in the mobile advertising domain is presented in Appendix 2. Truong et al.'s (2019) review of in-app advertising specifically examines the relationships between factors controlled by publishers and the effectiveness of in-app advertising.However, there are no SLRs that examine the broader research issues in in-app advertising research.
Hence, this study reviews the extant literature on in-app advertising, focusing on three research questions: RQ1.What are the key overarching themes latent in the literature on in-app advertising?
RQ2.What are the most prevalent theories, contexts, characteristics and methods used in the in-app advertising literature?
RQ3.What are the future areas of research on in-app advertising?
This review makes several contributions to the emerging research area of in-app advertising.First, the review is the maiden attempt to examine the extant literature on inapp advertising and identifies five key themes.Second, the study identifies research gaps in the theories, characteristics, context and methods used in the in-app advertising literature.
Third, the study proposes future areas of research and identifies potential theories that could shed light on new dimensions of in-app advertising.

Review methodology
The study uses a SLR, as suggested by Transfield et al. (2003).Systematic studies seek to discover, examine and summarize the contents in an integrative approach, thereby making SJME 28,3 the evidence more conducive for scholarly audit (Bergman and Holden, 2010;Transfield et al., 2003).SLRs are suitable for niche research areas with a narrow scope (Donthu et al., 2021) and have a smaller number of papers for review (Snyder, 2019).Donthu et al. (2021) posit that SLR methodology is often used when the review goal is to summarize and synthesize findings of a research topic and the volume of the data set for review is small for manual review.Paul and Criado (2020) broadly categorize SLRs into domain-based, theory-based and method-based reviews.The authors posit that a domain-based review can be termed a framework-based review if the study adopts an existing framework or develops a framework and uses it to organize the review.In this study, we perform a framework-based SLR with 44 shortlisted articles on in-app advertising.We adopt the Theory-Context-Characteristics-Methodology (TCCM) framework developed by Paul and Rosado-Serrano (2019) for examining and analyzing the in-app advertising literature and presenting the findings, gaps and future areas of research.The TCCM framework is extensively used in marketing to examine the literature and present the findings of the review (Bhattacharjee et al., 2022;Celik et al., 2022;Jebarajakirthy et al., 2021;Singh and Dhir, 2019).In this paper, we use this framework to present theories on in-app advertising (used in the literature and potential for future research), context (country, industry, advertising format), characteristics (antecedents and outcomes, mediators and moderators) and methods (research approaches and analysis techniques).
We perform thematic analysis on the selected articles to identify themes in the in-app advertising literature.Thematic analysis is a method for identifying, analyzing and recording patterns within the data (Braun and Clarke, 2006) and synthesizing them into themes (Thomas and Harden, 2008).We followed the process outlined by Braun and Clarke (2006) and Thomas and Harden (2008) to conduct a thematic analysis as presented in Section 3.

Search strategy, information retrieval and information processing
The PRISMA guidelines have assisted the article search and selection process, as followed by Page et al. (2021).To address the aforementioned research questions, we follow a threestep process to identify the relevant studies.The first step is to determine an appropriate database to retrieve documents for the study.Articles were sourced from the Scopus and Web of Science databases, and articles until the end of May 2023 were included in this review.The two journal repositories are considered the largest and most widely used online databases and search engines by researchers (Paul et al., 2021).Peer-reviewed journals in the English language constitute the prime source of articles included in this study.
The second step consists of selecting appropriate keywords to extract relevant documents.The search keywords "in-app advertisements", "in-app ads", "in-app advertising", "mobile app advertisements", "mobile app advertising", "mobile app ads" and "mobile application advertising" were selected to exemplify the line of inquiry effectively.In the third step, the data set was gathered and refined for the study.
The search method of title, abstract and keywords was used with Boolean operators of relevant keywords.One additional round of search iteration with the cited references was carried out to ensure the inclusion of relevant research studies in the area.The PRISMA protocol for reporting literature in systematic reviews was used to report the search strategy (Page et al., 2021).Figure 1 outlines the methodology of the SLR.The initial search yielded 209 studies published in the two databases.
The data retrieved were exported to Microsoft Excel for scrutiny and 105 studies were shortlisted for SLR after removing duplicates and non-English studies from the two databases.Book chapters, conference papers, editorials and lecture notes were excluded In-app advertising because they were not necessarily peer-reviewed and the exclusion resulted in 56 records.To winnow down the out-of-scope studies, abstract screening was performed and 12 records were excluded because they did not encapsulate the essence and conceptual boundary of inapp advertising.Thus, 44 studies were included in the final analysis of this review.

Thematic analysis of in-app advertising literature
Research on in-app advertising spans several areas, such as information technology, consumer behavior and psychology, marketing, advertising and communication.We performed a thematic analysis of the shortlisted articles to identify the broad areas of research in the in-app advertising literature.The coding of text and generating and reviewing themes were performed based on the thematic analysis process suggested by Braun and Clarke (2006).Each shortlisted article was first coded manually by examining the title and abstract by the first author.Seventeen descriptive themes emerged by grouping similar codes.Figure 2 illustrates the thematic map emerging from the literature.The descriptive themes were labeled based on their significance and context in relation to the findings of the primary studies.The descriptive themes were then verified for relevance by the second author and finalized after discussion.
In the next phase, the analytical themes were inferred from the descriptive themes (Thomas and Harden, 2008) and five themes emerged in the discussions between the authors: ad platform optimization; mobile app user psychology and behavior; ad effectiveness; ad fraud; and security, privacy and other user concerns.
The thematic analysis indicates the relationship between the two themes in the literature, namely, ad platform optimizationad effectiveness (Truong, 2023) and ad effectiveness -   (Mattke et al., 2021).The click-through rate was one of the outcome variables that was investigated by several researchers for the optimal combination of in-app advertisement factors and the optimal sequencing of in-app advertisements (Adikari and Dutta, 2020;Sun et al., 2017;Truong et al., 2019).The relationship between the themes has explored the factors controlled by advertisers, consumers and ad networks on click behavior.However, the relationship between other themes was not explored in past literature.

Theme 1: ad platform optimization
The in-app advertising ecosystem is divided into players on the buy-side or demand-side and the sell-side or supply-side.Optimizing the in-app advertising platform is of utmost importance, keeping in view the interest of both buy-side and sell-side players.Each player in the in-app advertising ecosystem has varied outcomes of interest (Truong et al., 2019).
Publishers can either sell their inventory directly to ad networks or buy-side or engage in real-time bidding (Kaplan, 2021).Truong (2023) found that the optimal combination of publisher-controlled factors such as duration, timing, size and position of the ad space significantly improves the click-through rate of in-app advertising.
Several recent studies have examined the techniques and guidelines for app developers to assist them balance the trade-offs in-app advertising costs and benefits.Ghose and Han (2014) find that demand for mobile apps decreases with in-app advertisements and suggest several app design choices for app developers.App developers often integrate ad libraries with apps and there are more than 63 ad libraries (Ahasanuzzaman et al., 2020) in the Google Play Store.Ahasanuzzaman et al. (2022) develop a reference architecture for ad library integration for app developers.
The ad networks connect publishers and advertisers, allowing advertisers to bid on the ad space provided by the publisher.Although publishers may have many ad networks ready to buy ad inventories, the performance of networks differs.Adikari and Dutta (2020) have suggested an optimal mechanism for publishers to select the best ad network.Recent research In-app advertising (Lee et al., 2021;Rafieian and Yoganarasimhan, 2020;Sun et al., 2017;Yang and Cao, 2018) examines efficient techniques for ad networks using machine learning techniques.
Ad exchange orchestrates the buying and selling of ads and can connect publishers with multiple ad networks.Revenue optimization is a vital task for an ad exchange.Scholars have examined techniques to improve ad auction mechanisms (Liu and Liu, 2019) and optimal revenue-sharing methods (Hao et al., 2016;Hu et al., 2023).Mukherjee et al. (2017) develop a machine learning-based in-app ad selection algorithm and propose an optimization model to maximize supply-side platform revenue.
The current research on this theme is primarily exploratory and uses data-driven approaches.There is no theoretical framework applied to the studies and machine learning techniques are prominent.Researchers provide either a novel optimal solution or suggest an extension of an existing technique.Although the sell-side platform has gained traction, academic literature on the buy-side is sparse.An exception is the field research by Rosenkrans and Myers (2018) that proposes how predictive analytics could be used by advertisers to optimize mobile location-based ads.The optimization of advertisements based on ad characteristics such as design, context, size and media types and their influence on information processing and consumer response remain unexplored.

Theme 2: mobile app user psychology and behavior
This theme primarily focuses on studies on attitude, behavioral intention and user behavior toward in-app advertising.Attitudes toward in-app advertising have been examined by several researchers and found to impact people's intention to use the app (Sigurdsson et al., 2018).Logan (2017) posits that attitude varies according to the gratification sought from the mobile app category.Scholars posit that the perceived value of in-app advertising, trust toward ads, ad source and perceived irritation toward ads are significant factors affecting attitudes toward advertising (Aydin and Karamehmet, 2017).Users' propensity to trust affects their trust in in-app advertising and trust impacts attitudes toward in-app ads and their intention to view in-app ads (Cheung and To, 2017).Attitude also mediates the relationship between trust and purchase intentions (Sung, 2020).Ad factors such as credibility, information and perceived entertainment considerably improve attitudes toward in-app ads (Sigurdsson et al., 2018).
The behavioral response toward ads is better when there is congruity between the ad and the mobile environment (Wang and Chou, 2019).Mattke et al. (2021) conduct a qualitative comparative analysis and empirically show the sequential processing of structural and semantic factors of in-app advertisements.Keem and Lee (2018) examine the types of advertisements consumers prefer or adversely react to when shopping through mobile apps.
Consumers control their in-app advertising performance through their behavior and personal characteristics (Truong et al., 2019).However, studies on personal characteristics and their impact on in-app advertising are scarce.User psychology and behavior can vary based on the type of app used, user loyalty and ad formats.In addition, there is ample opportunity to understand aspects of user psychology and behavior through theoretical lenses.
3.3 Theme 3: ad effectiveness Rodgers and Thorson (2000) categorize all factors affecting advertising effectiveness into factors controlled by consumers and advertisers.The researchers posit that attention, memory, attitude and outcomes are controlled by consumers.Therefore, these are important for advertisers to measure ad effectiveness.Ad recall is one of the popular metrics used to SJME 28,3 measure ad effectiveness.Cicek et al. (2018) examine the effects of banner location, app type and orientation on ad recall and find that banner ads positioned at the top can trigger better recall.
The literature addresses only a few consumer/advertiser factors, as noted above.The impact of creativity on ad effectiveness has received less attention.Moreover, publishercontrolled factors (ad timing, ad position) and network-controlled factors (location, device type and size, which impact ad effectiveness) remain unexplored.

Theme 4: ad fraud
The rise in popularity of free and open-source Android platforms for mobile has resulted in fraudulent and repackaged apps.Repackaged apps are modified versions of existing popular apps, which are designed to capture advertising revenue and obtain user information.The lack of in-app ad-blocking solutions among mobile users has led to abusive in-app advertising practices such as click fraud and malvertising.App developers can generate fraudulent ad clicks in an automated manner, resulting in losses for advertisers.Recent studies have examined the vulnerability of ad networks to click fraud attacks and proposed techniques for ad networks to detect click fraud (Cho et al., 2016;Mouawi et al., 2019).Cho et al. (2016) note that there is a dearth of research on the vulnerability of ad networks to different varieties of ad click threats and methods to prevent them.New forms of ad fraud, such as ad stacking, click flooding, device ID reset fraud, bot fraud, app fraud, bundle ID spoofing and software development kit (SDK) spoofing (TrendMicro, 2019), have received little attention from researchers.There are ample research opportunities to leverage developments in machine learning algorithms and artificial intelligence to combat these threats.

Theme 5: security, privacy and other user concerns
Studies have examined the security, privacy and smartphone performance concerns of mobile users resulting from in-app advertising.A mobile app embeds a custom SDK given by an ad network to display ads.This gives greater control over the ad network on the mobile application.Scholars (Su et al., 2018) observe that in-app advertising poses a higher security risk to users by collecting their personal information, downloading malware in the background, disclosing images and personal data and exposing them to inappropriate content.Researchers have voiced their concerns about the noncompliance of app developers with existing relevant international guidelines or technology standards (Haenschen and Wolf, 2019;Scott et al., 2018) and the need for regulation of ad libraries.Several studies have proposed techniques to detect malvertising in ad networks (Shao et al., 2018) and prevent third-party libraries from misusing the permissions of host apps (Hsu et al., 2020).Recent research (Gao et al., 2021(Gao et al., , 2022) ) has identified user concerns with in-app advertising using app reviews.
Our review signals a lack of research on the brand safety of advertised brands and brands powering the app.Further research is needed on improper content, dark patterns in apps and their impact on the brand.Behavioral ad avoidance, such as closing or uninstalling In-app advertising the mobile app, due to content inappropriateness of in-app advertising and dark patterns, is a major concern to publishers as well as advertisers.Future research can explore the impact of such issues on attitudes toward the ad and brand powering the app.Threats posed by inapp advertising among vulnerable groups such as children are another unexplored area.
4. Theory-Context-Characteristics-Methodology in in-app advertising: findings and discussion In accordance with the TCCM framework developed and applied by Paul and Rosado-Serrano (2019) and following the examples of other TCCM reviews (Billore and Anisimova, 2021;Celik et al., 2022;Ghorbani et al., 2022;Jebarajakirthy et al., 2021), we present a comprehensive outline of the findings and gaps existing in the literature.
In the theoretical perspective section, we present the most commonly used theoretical frameworks in the in-app advertising literature.In the context section, we present the research settings (country, industry) used in the in-app advertising literature and the diverse types of ad formats used.In the characteristics section, we examine the antecedents, outcomes, mediators and moderators used in in-app advertising studies.We summarize the TCCM framework for in-app advertising in Figure 3.
4.1 Theoretical perspectives in the in-app advertising literature (T) According to our review, previous studies in in-app advertising have not used theoretical frameworks to understand the unique complexities and features of in-app advertising.Only 11 studies (approximately 25% of the studies) in the in-app advertising literature have used theoretical underpinnings.The studies are primarily on the themes of "mobile app user psychology and behavior" and "ad effectiveness".Table 1 presents the theories applied in the in-app advertising literature.
Theoretical integration of multiple paradigms could serve to better understand and interpret the complex nature of the advertising phenomenon in digital media (Rodgers and Thorson, 2000).For effective advertising research, both exploratory research and theorybased research are important (Huh, 2017); hence, future researchers should establish connections between existing advertising theories and computational advertising research.
The review highlights that in-app advertising has relied on nine theories, as described in Table 2. Uses and gratification theory was the most commonly used theory in the in-app advertising literature, with four studies applying it in their research.There is potential to use several advertising theories to examine new dimensions of in-app advertising research.We propose a few of them in Section 5.

Context of the studies in in-app advertising (C)
The TCCM framework regard "context" as circumstances forming the research setting (Paul and Rosado-Serrano, 2019).We analyzed the countries, industries and ad formats that have appeared in the in-app advertising literature.In-app advertising has been researched in 16 different countries.Data were collected from US participants in 30% of studies, followed by China, (11%), Canada, India, Taiwan (approximately 7% each, i.e. 20%), Hong Kong, Italy, Korea, Turkey, UK, Vietnam (approximately 5% each, i.e. 28%) and Australia, Germany, Singapore, Lebanon, Malaysia, Uganda (approximately 2% each, i.e. 11%).Most of the studies used data from participants from a single country, except for studies by Sigurdsson et al. (2018), which used participants from two countries.

SJME 28,3
However, industries such as financial services, retailing, hospitality, media and entertainment have yet to be explored.The financial services industry extensively uses inapp advertising to upsell and cross-sell financial products and services (Upshot.ai, 2022).Along similar lines, hospitality, retailing, media and entertainment use in-app advertising for both marketing and revenue generation.Hence, there is a vast scope for research in these industries.
In the context of advertisement formats, studies have investigated contextual advertising (Graham et al., 2021), banner advertising (Cicek et al., 2018), interstitial advertising (Wang and Chou, 2019), short messaging services (Aydin and Karamehmet, 2017) and augmented reality advertising (Sung, 2021;Sung et al., 2022).However, the Interactive Advertising Bureau (2022)  (Miller, 1956) Mattke et al. (2021) The objective was to observe the processing of structural and semantic that leads to clicking behavior

Limited Capacity Theory:
The theory posits that attention is limited in overall capacity (Lang, 2000) Uses and Gratification Theory: The theory helps understand why people use particular type of media and what gratifications do they get from using them (Katz et al., 1973) Sigurdsson et al. (2018) The objective was to investigate consumers' attitudes toward in-app ads in terms of ad value Logan (2017) The objective was to investigate attitudes toward in-app ads in different app categories Aydin and Karamehmet (2017) The objective was to determine the major factors influencing consumers' attitudes toward SMS and mobile application advertising Theory of Narrative Transportation: The theory proposes that people are often influenced and changed by a narrative, and it can be used to persuade people to change their attitudes and beliefs (Green and Brock, 2000) Sung et al. (2022) The objective was to examine narrative transportation (in response to ad storytelling) and spatial immersion in AR advertising Theory of Planned Behavior (TPB): An individual's decision to engage in a specific behavior can be predicated by their intention to engage in that behavior (Ajzen, 1991) Cheung and To (2017) The objective was to extend TPB to include the "propensity to trust" and "trust" as antecedents of users' attitudes toward in-app ads Hsu et al. (2020) The objective was to investigate parental perspectives on the adoption of the m-Learning app with in-app ads Schema Congruity Theory: Assimilation of new information is dependent on the levels of fit between new information and the existing schema (Mandler, 1982) Wang and Chou (2019) The objective was to examine how the degree of congruity between promoted products in interstitial ads and the mobile gaming app environment affects consumer response Elaboration Likelihood Model: The model explains how humans process stimuli differently and how these processes change attitudes and, consequently, behavior (Petty andCacioppo, 1986) Cicek et al. (2018) The objective was to investigate the effects of ad banner location, app type and orientation on ad recall

Prospect Theory:
The theory assumes that losses and gains are valued differently, and decisions are made on perceived gains instead of perceived losses (Kahneman and Tversky, 1979) Sung (2020) The objective was to examine factors affecting the relationship between brand trust and purchase intentions.The moderating role of the anticipated gain, i.e. access to discounts was examined SJME 28,3 lists many more advertisement formats, such as reward ads, native ads, video ads, emoji ads, virtual reality ads and audio ads.Future research can explore these emerging advertisement formats.

Characteristics of in-app advertising studies (C)
We examine the empirical research on in-app advertising that used theoretical frameworks and synthesized the antecedents, outcomes, mediators and moderators of inapp advertising in this section.Studies reveal that advertisement-related antecedents such as advertising content, animation and interactivity influence advertising value and have significant effects on emotional engagement (Cassioli, 2019) and visual behavior (Cassioli and Balconi, 2022).Antecedents such as informativeness, credibility, entertainment (Aydin and Karamehmet, 2017), personalization (Sigurdsson et al., 2018), trust (Cheung and To, 2017), attitudinal beliefs (Hsu et al., 2020) and gratification sought (Logan, 2017) influence attitudes toward in-app advertisements.Attitude influences the intention to watch in-app advertisements (Cheung and To, 2017), the intention to adopt m-learning platforms with in-app advertisements (Hsu et al., 2020) and the behavioral intention to click on in-app advertisements (Sigurdsson et al., 2018), which in turn influences user behavioral responses.Self-efficacy, the perceived value of collaboration and incentives influence the intention to interact with in-app ads (Qureshi et al., 2022).Cicek et al. (2018) provides causal evidence that app type, app orientation and banner ad location influence ad recall.Mattke et al. (2021) studied structural factors of in-app advertising, including animation, color brightness, location prominence, size and similarity to background structure, and semantic factors, including perceived Information processing is related to the function of the consumercontrolled aspects of the internet and the advertiser-controlled aspects like ad structures (Rodgers and Thorson, 2000) The theory can help examine the array of responses that result from the encounter of consumer-controlled and advertiser-controlled aspects of in-app advertising and their influence on information processing and consumer responses Perceptual Load Theory The success or failure of selective attention is dependent on the processing demands of the current task (Lavie et al., 2004) The theory can help investigate the degree to which distractors (in-app advertisements) are processed during high and low perceptual loads Media Engagement Framework Engagement is conceptualized not as a behavior but as a motivational experience (Calder and Malthouse, 2009) The theory can help explain the impact of in-app advertising on mobile apps based on the different app types and their engagement experience Theory of

Interactive Media Effects
The level of interactivity in the medium determines the type and strength of the media's effects (Sundar et al., 2015) The theory can aid researchers describe the positive, as well as negative effects, of various forms of in-app advertisements Psychological Reactance Theory When an individual's freedom is threatened, they experience psychological reactance, which is a motivational state that drives freedom restoration (Brehm, 1966) The psychological reactance due to in-app advertisements and their impact on mobile app usage can be studied using the theory In-app advertising entertainment value, informativeness, irritation, privacy concern, credibility, congruity and personalization, to investigate the consequences of clicking behavior.The influence of antecedents such as advertisement types affects the click-through rate (Rosenkrans and Myers, 2018), advertising avoidance behaviors (Keem and Lee, 2018), purchase intention (Sung, 2021) and social media sharing (Sung et al.,2022).The interactive effects of publisher-controlled factors, namely, ad space size, ad space position ad space timing (Truong et al., 2019), and ad space duration (Truong, 2023), with advertiser-controlled factors, namely, ad type, ad medium, location (Truong et al., 2019) and time (Truong, 2023), on the click-through rate were investigated.Wang and Chou (2019) examine how the degree of congruity between promoted products in interstitial ads and the app environment affects consumer response toward the ad.
Mediator constructs such as trust, intention to watch in-app advertisements (Cheung and To, 2017), attitude (Sigurdsson et al., 2018), ad response (Sung, 2020), perceived value of collaboration and incentives (Qureshi et al., 2022) were used to study the impact on behavioral intention.Wang and Chou (2019) investigated the moderating effects of game types and game immersion on advertising effects.Aydin and Karamehmet (2017) examined the mediating effect of advertising value on attitude.Qureshi et al. (2022) assessed the moderating effect of self-efficacy on the intention to interact with in-app advertisements.Based on the synthesis of the literature, we found that with reference to outcomes, clicking behavior, behavioral intention and attitude are the most frequently studied consequences.
In the current studies, antecedents related to consumer-related factors such as personality traits, memory, feelings of intrusiveness, psychological reactance and their impact on behavioral outcomes are scarce.Understanding the unintended consequences of in-app advertising is important, and future studies can focus on reducing negative consumer reactance using an optimal combination of publisher-controlled, consumer-controlled and advertiser-controlled (Mattke et al., 2021) antecedents.In future research, there is also a need to investigate the moderating effect of demographic variables such as age, income and gender on clicking behavior.Empirical studies using theoretical frameworks to examine user behavior in in-app advertising are scarce.

Methods in the in-app advertising literature (M)
The methodology-wise analysis suggests that quantitative research methods were prominent (93 % of the total).Only two studies (Haenschen and Wolf, 2019;Scott et al., 2018) used a qualitative approach, and one study (Logan, 2017) used a mixed method.
As in-app advertising is an interdisciplinary topic, 50% of the studies were from the field of information technology.The methods in these studies were simulation experiments and the development of effective algorithms, optimization models, econometric models or machine learning models.The analysis techniques used in these studies examined the effectiveness of these models, and in two studies (Gao et al., 2021(Gao et al., , 2022)), text mining was used.
The remaining 50% of the total studies were from marketing, advertising and consumer behavior.The methods in these studies include surveys (9), laboratory and field experiments (7), neuroscientific experiments (2), qualitative (2) and mixed methods (1).The analysis techniques used in these studies include SEM, ANOVA and descriptive analysis.
As in-app advertising is still emerging, a mixed-method approach can provide a comprehensive understanding of an emerging phenomenon.SJME 28,3

Future research agenda
There are several gaps in the extant literature that provide avenues for future research.From a theoretical perspective, we observe that research in in-app advertising has not been examined through several theoretical lenses.Future empirical research with theoretical foundations would advance the understanding of consumer behavior in this emerging area.Rodgers and Thorson (2019) offer a comprehensive synthesis of the vast literature on advertising and provide theoretical approaches to advertising.Considering the unique characteristics of mobile apps and the high level of user engagement compared to other mediums, there is scope for applying several advertising theories.We propose five advertising theories in Table 2 for future research on in-app advertising.
There is also ample scope for examining underexplored contexts (country, industry and ad format) in in-app advertising.As research from the USA dominates the literature, future research may examine app user behavior from Asia-Pacific and European regions.Countryspecific issues examining the regulation and compliance of app developers could be undertaken.As there are only a few studies that examine the buy-side or publishers of inapp ads, future research may examine publisher-controlled elements and their impact on ad effectiveness and user behavior.As there are novel formats of in-app ads such as virtual reality ads and many multimedia ad formats, future research may examine issues such as ad avoidance, ad fraud and ad effectiveness in emerging ad types in in-app advertising.
Only a few users' behavior constructs in the in-app advertising context have been studied in the literature.Future studies may examine user demographics, personality traits, motivation, attention and their impact on attitude and ad avoidance.Age and gender can be examined as moderating variables, as they can also influence attitudes toward in-app advertising.
The mobile app is a pull medium and users are highly goal-oriented compared to other mediums.Hence, consumer-controlled aspects (Rodgers and Thorson, 2000) of mobile apps, such as motivation, attention, involvement and their impact on cognitive and emotional processing of in-app advertising, may be explored in future studies.
From a methodological perspective, analytical and quantitative research have dominated the literature.Future empirical research with a theoretical foundation and mixed methods research approaches (both quantitative and qualitative) could be used.Case studies in in-app advertising could also provide more insights into the in-app advertising phenomenon.New machine learning and artificial intelligence techniques need to be researched to combat ad fraud.Table 3 highlights the key research gaps in theories, context, characteristics and methodology in the in-app advertising literature and proposes future research directions.

Theoretical and practical implications and conclusions
This study makes three main theoretical contributions to the growing literature on mobile advertising and digital marketing.First, the study is a maiden attempt to systematically review the literature on in-app advertising and synthesize it into five broad themes.The thematic analysis reveals that there are several research gaps in the in-app advertising literature and there is a need for empirical research on emerging ad formats of in-app advertising, user behavior and issues on the buy-side of in-app advertising.As in-app advertising is a fast-growing mobile advertising format, understanding consumer behavior and its effectiveness in different contexts is essential.We highlight some of the key differences between in-app advertising and mobile web advertising to spark interest in future researchers and undertake further studies.
Second, the study identifies the gaps in the literature and proposes future research areas (refer to Table 3) in in-app advertising based on the TCCM framework.The framework (refer to Figure 3) developed for in-app advertising gives a holistic view of the body of knowledge.It presents the antecedents, outcomes, moderators and mediators in the current literature.Third, we propose five advertising theories (refer to Table 2) that could guide future researchers to investigate issues in in-app advertising.We also highlight future research directions using the proposed theories.

In-app advertising
The study has practical implications for advertisers and other stakeholders in the in-app advertising ecosystem.The review provides practitioners with an overview of the in-app advertising research that could be used to develop more effective strategies for its deployment.The review presents practitioners with an unbiased summary of in-app advertising research that can guide advertisers in designing and delivering in-app advertisements.By providing insights into the privacy and security concerns of users, we encourage practitioners to be mindful of using ethical practices in in-app advertising.One of the limitations of the current review is the paucity of empirical studies.Quantitative analyses such as bibliometric analysis and meta-analysis could not be performed due to the heterogeneity of constructs and the paucity of empirical studies (Donthu et al., 2021).Future research may examine a single phenomenon/construct in in-app advertising.
In-app advertising is set to be a key driver of mobile advertising over the coming years.Advances in mobile technology coupled with industry challenges such as ad fraud, ad avoidance and ad effectiveness measurement provide many opportunities for future research in the in-app advertising domain.

Figure
Figure 1.Flow diagram of systematic literature review on in-app advertising