In this paper, we aim to advance the research on how companies navigate channel integration by examining the internal and external challenges they encounter. Specifically, we investigate how internal obstacles and external industry drivers affect the level of channel integration.
In our quantitative study, we collected the relevant data from 412 firms operating in over 20 diverse industries and offering both online and offline channels. We also explore how organizational omnichannel capabilities moderate the relation between internal and external factors and the level of channel integration.
Our results indicate that channel integration is hindered by internal barriers, including limitations in operational efficiency, strategy and organizational culture. Additionally, external pressures stemming from industry-specific factors contribute to these challenges. Conversely, positive influences may arise from micro-environmental factors, such as an existing customer base already literate with omnichannel solutions or competitors advanced in omnichannel strategies.
To evaluate the effects of channel integration, we examine its influence on performance across multiple dimensions (short-term, long-term and comparative), extending prior research that has predominantly emphasized short-term performance metrics.
Introduction
Omnichannel is considered the post-pandemic new normal, both for retailers and consumer behaviours (Alexander and Varley, 2025; Sharma and Dutta, 2025; Sharma and Fatima, 2024; Hickman et al., 2020; Silva et al., 2024). The technological advances and the proliferation of channels create ongoing opportunities for testing and switches in the channel portfolio (Sharma and Fatima, 2024; Yin et al., 2022). Such variability makes the challenge of channel integration everlasting and scholars examine it from various theoretical and practical perspectives, with the seminal works of Verhoef et al. (2015), Cao and Li (2015), or Neslin et al. (2006). Despite numerous studies on omnichannel in the retailing stream (Yan et al., 2021; Oh et al., 2012; Sharma and Fatima, 2024; Alexander and Varley, 2025; Silva et al., 2020), research on how companies manage channel integration, what obstacles they encounter, and how they overcome them seem to be underdeveloped (Radomska et al., 2025). Scholars have applied various standpoints to investigate operational factors impacting channel integration (de Borba et al., 2021). However, despite previous suggestions that its level may be sensitive to various internal influences (Hajdas et al., 2022), the relations between these constructs have not been fully investigated. Thus, our first research question considers: (RQ1) What is the relationship between the level of channel integration and internal obstacles firms face when going omnichannel?
Scholars argue that the degree of omnichannel implementation may vary depending on the industry’s features (Sharma and Dutta, 2025; Iglesías-Pradas et al., 2021; Hajdas et al., 2022; Silva et al., 2020). However, these suggestions have not been sufficiently explored. Despite scholars’ suggestions about the potential role of industry’s features in the degree of omnichannel implementation (Iglesías-Pradas et al., 2021; Hajdas et al., 2022), a broader perspective verifying these suggestions is lacking, leaving the question of whether all industries are equally prone to omnichannel operations unanswered. Therefore, our second research question is: (RQ2) How do industry characteristics affect channel integration efforts?
According to Akter et al. (2024) and Barbosa and Casais (2022), a particular organizational capability can lower the transformational hurdles of integrating channels. However, the exact role of these capabilities in managing internal struggles and responding to external challenges remains understudied. Our study fills this gap and explores omnichannel management capabilities beyond the single industry context. Therefore, our final research question is: (RQ3) How do organizational capabilities contribute to overcoming internal barriers and dealing with external challenges when integrating channels?
We aim to verify and explain if and how internal barriers and external factors, such as industry characteristics, contribute to the level of channel integration. The scope of our quantitative study covers 412 firms operating in over 20 diverse industries [1]. We investigate how our respondents evaluate the nature of their industries in terms of the competition, demand, internationalization practices, etc. We then compare these findings with how they evaluate their channel integration efforts. Our investigation allowed us to gain a bigger picture of how various industry characteristics affect channel integration possibilities. We contribute to extant omnichannel scholarship with a validated framework showing the scope of internal obstacles, specific organizational capabilities (organizational omnichannel capabilities) to overcome them, as well as the external industry drivers affecting the level of channel integration. We also empirically investigate the relationship between the level of channel integration and firm performance, a complex construct characterised by relatively high dynamics, where short-term successes are not always sustained in the long run (Ricciardi et al., 2022). Given that research typically adopts a static, short-term approach (Tagashira and Minami, 2019), we selected a more comprehensive perspective when investigating the outcomes of channel integration efforts. For this reason, in this study, we adopt three perspectives: short-term, long-term, and comparative performance.
Theoretical background and hypotheses development
Internal perspective: strategic and operational obstacles
Internal obstacles to channel integration exhibit a dual nature: strategic or operational (Hajdas et al., 2022). The strategic obstacles include misaligned corporate motivations (Hübner et al., 2016), resulting from a siloed nature of channel management (Picot-Coupey et al., 2016), which may lead to a gap between marketing and sales (Rouziès et al., 2005) or logistics and marketing (Oh et al., 2012), as well as to structural conflicts between business units regarding channel autonomy (Larke et al., 2018) or between physical and digital channels (Xu and Cao, 2019). Conflicting interests across various channels are mainly caused by an inconsistent strategy (Lewis et al., 2014), including the lack of an effective communication strategy (Webb, 2002) and the lack of coherence between the long-term vision and short-term actions (Chopra, 2016). The lack of vision is considered the most severe strategic challenge (Ye et al., 2018).
The operational issues are the second group of internal obstacles to achieving a high level of channel integration. Among them, data integration seems to be the main challenge due to more data fuelling databases (Brynjolfsson et al., 2013). Integrative technologies (Iftikhar et al., 2019) aligning the data flows require not only vast financial investment (Herhausen et al., 2015) but also skills. Retailers often struggle with selecting and implementing the adequate integrative technology, which hampers the investment in technological omnichannel solutions (Zimmermann et al., 2023). Another significant type of operational barriers firms face when integrating the channels are the reverse channel problem (Tanriverdic and Aydın, 2023; Risberg, 2022), structuring and allocating resources across channels (Lewis et al., 2014) as well as measuring the channel’s efficiency (Cai and Choi, 2023). Such measurement is challenging as a single channel may support the performance of the overall omnichannel system but it may not be profitable in isolation (Hajdas et al., 2022).
Furthermore, the internal obstacles faced by firms transitioning to omnichannel are latent constructs – complex and not directly measurable – covering two major dimensions: (1) operational efficiency and (2) strategy and organizational culture (Radomska et al., 2025). Further studies are needed to confirm the exact relation between the internal obstacles, the level of channel integration, and overall firm performance. Therefore, we hypothesise that:
Internal obstacles negatively impact the level of channel integration.
External perspective: Industry drivers
The degree of omnichannel implementation may vary significantly depending on the distinctive features of each industry (Hajdas et al., 2022; Iglesías-Pradas et al., 2021) or product type (Sharma and Dutta, 2025). Such a suggestion aligns with studies that highlight that strategic choices, in general, are highly context-dependent (Elliott et al., 2018). Previous qualitative studies indicate that product-related, competitive, market-related, and legal factors are key omnichannel industry drivers influencing channel integration (Hajdas et al., 2022). However, further research—particularly quantitative—is needed to validate these findings. Consequently, we conceptualise the industry driver as a factor determining the feature of the industry, neutral in its nature. Only its specification exposes whether such industry features work in favour or against a particular strategy (omnichannel in our case). Industry drivers specification can be done based on binary oppositions. For example, an industry driver can be the nature of demand for a given good (local or global demand), and it cannot be hypothesised that the nature of demand has a positive effect on omnichannel. However, it can be hypothesised that, for example, global demand (binary opposition of a local demand) is conducive to omnichannel, and local demand (binary opposition of a global demand) is hindering omnichannel. That is why our hypothesis for this variable is neutral, and we hypothesise the following:
Industry drivers impact the level of channel integration.
The moderating effect of organizational omnichannel capabilities
Dynamic capabilities are key in omnichannel management (Akter et al., 2024; Li et al., 2023). However, previous examinations of the relationship between dynamic capabilities and channel integration yield ambiguous results. For example Hossain et al. (2020a, b) and Bahar et al. (2021) perceive channel integration as the dynamic capability allowing firms to coordinate resources and processes across channels or to create new resources configuration in omnichannel marketing. Akter et al. (2024) describes the omnichannel management capability as a dynamic capability, whereas others scholars (Cao and Li, 2015; Tagashira and Minami, 2019) consider channel integration as a digital strategy to support omnichannel marketing.
Based on previous literature (Akter et al., 2024; Li et al., 2023; Hossain et al., 2020a, b; Barbosa and Casais, 2022; Solem et al., 2022), we argue that some specific omnichannel organizational capabilities may influence the relationships between the level of channel integration and both managing internal obstacles and dealing with external challenges. Barbosa and Casais (2022) show, for example, how retailers overcome various barriers by integrating information technology, accomplishing organizational changes, and optimising customers’ feedback. It suggests that a particular set of capabilities may lower firms’ transformational barriers when integrating channels. We argue that firms that develop their organizational omnichannel capabilities related to omnichannel management are more resourceful in terms of dealing with both internal obstacles and industry drivers hindering channel integration. Those organizational capabilities are considered organizational omnichannel capabilities, defined as “the ability of a retailer’s marketing channels to provide the same level of service, assortment, notification (informing), return and delivery options from the consumers’ perspective” (Yumurtacı Hüseyinoğlu et al., 2018). Therefore, we hypothesise that:
Organizational omnichannel capabilities weaken the negative impact of internal obstacles on the level of channel integration
Organizational omnichannel capabilities moderate the relationship between industry drivers and the level of channel integration
Level of channel integration and performance
Channel integration can be defined as a firm’s ability to provide customers with a seamless purchasing experience across channels (Sousa and Voss, 2006). Yan et al. (2021) and Stojković et al. (2023) suggest that an integrated channel approach leads to better overall performance, attracting consumer demand. Oh et al. (2012) showed the positive impact of retail channel integration on firm competencies and performance. Thus, increasing channel integration levels is mainly aimed at maximising firm performance (Kolbe et al., 2022). It is important to note that firm performance is a multidimensional construct, encompassing various aspects such as financial results, firm power, and profitability (Ricciardi et al., 2022). In addition, firm performance should be viewed as a dynamic concept that evolves over time in response to market and internal changes (Klimas et al., 2024). Therefore, we assume that:
The level of channel integration positively impacts comparative performance.
The level of channel integration positively impacts short-term performance.
The level of channel integration positively impacts long-term performance.
Research design
Sample and data collection
We have collected data from companies with two established channels (offline and online for the same products), deploying a random sampling method. The sampling frame covered 71,000 companies as the population of Polish online shops. The planned sample size was calculated as 383 with a confidence level of 95%, and an assumption of 50% response distribution, with the acceptable margin of error being 5% (Roasoft, 2023). Our final sample was higher than expected and included 412 valid responses. While calculating the research sample, we followed the study claiming that 19% of Polish retailers allow customers to shop as omnichannel (Mierwinski, 2022). At the end of 2021, the number of retail stores in Poland was estimated at 376,000 (Mazurkiewicz, 2022). Three reasons justify choosing Polish retail as the research context: (1) the overall retail market in Poland continues to grow (Retail Market in Poland, 2024), (2) the number of e-commerce users in Poland is expected to grow to 20.1 million in 2029 (Statista, 2024a) and (3) due to the COVID-19 pandemic, bidirectional transitions were observed in Polish e-commerce sector – many traditional retail chains shifted sales online, while many online retailers developed the offline channels (Statista, 2024b).
The data collection was outsourced to a professional agency, and the data-gathering phase covered informants holding leading positions dealing with channel integration (IT, sales, marketing) and having expert knowledge (Bagozzi et al., 1991). We used the mixed-mode survey (CATI, CAWI, PAPI, and CAPI) (De Leeuw, 2005) and a 7-point Likert symmetric scale (ranging from 1 – I strongly disagree to 7- I strongly agree) as a recommended measurement approach in social sciences (Taherdoost, 2019), including research applying structural equation modelling (Tarka, 2017).
Conceptual model and construct operationalization
We have developed a conceptual model tracing hypothetical links between the considered theoretical constructs – Figure 1.
The diagram starts on the left with two vertically arranged boxes labeled from top to bottom as “Internal obstacles” and “Industry drivers.” Two individual right arrows from these boxes point to a central box labeled “Level of channel integration.” Individual right arrows from the central box leads to three vertically arranged boxes on the right, labeled from top to bottom as “Short-term performance,” “Long-term performance,” and “Comparative performance.” A box on the top is labeled “Organizational omnichannel capabilities.” The arrows are labeled as follows: The arrow from “Internal obstacles” to “Level of channel integration” is labeled “H subscript.” The arrow from “Industry drivers” to “Level of channel integration” is labeled “H subscript 2.” The arrow from “Organizational omnichannel capabilities” to the arrow between “Internal obstacles” and “Level of channel integration” is labeled “H subscript 3.” The arrow from “Organizational omnichannel capabilities” to the arrow between “Industry drivers” and “Level of channel integration” is labeled “H subscript 4.” The arrow from “Level of channel integration” to “Short-term performance” is labeled “H subscript 5 a.” The arrow from “Level of channel integration” to “Long-term performance” is labeled “H subscript 5 b.” The arrow from “Level of channel integration” to “Comparative performance” is labeled “H subscript 5 c.”Conceptual model. Source: Authors’ own work
The diagram starts on the left with two vertically arranged boxes labeled from top to bottom as “Internal obstacles” and “Industry drivers.” Two individual right arrows from these boxes point to a central box labeled “Level of channel integration.” Individual right arrows from the central box leads to three vertically arranged boxes on the right, labeled from top to bottom as “Short-term performance,” “Long-term performance,” and “Comparative performance.” A box on the top is labeled “Organizational omnichannel capabilities.” The arrows are labeled as follows: The arrow from “Internal obstacles” to “Level of channel integration” is labeled “H subscript.” The arrow from “Industry drivers” to “Level of channel integration” is labeled “H subscript 2.” The arrow from “Organizational omnichannel capabilities” to the arrow between “Internal obstacles” and “Level of channel integration” is labeled “H subscript 3.” The arrow from “Organizational omnichannel capabilities” to the arrow between “Industry drivers” and “Level of channel integration” is labeled “H subscript 4.” The arrow from “Level of channel integration” to “Short-term performance” is labeled “H subscript 5 a.” The arrow from “Level of channel integration” to “Long-term performance” is labeled “H subscript 5 b.” The arrow from “Level of channel integration” to “Comparative performance” is labeled “H subscript 5 c.”Conceptual model. Source: Authors’ own work
We have adopted the scales validated by Radomska et al. (2025) to measure the internal obstacles. To measure the level of channel integration, we have used the scales validated by Cao and Li (2015) and Shi et al. (2020). We have used the catalogue of items proposed by Luo et al. (2016), Oh et al. (2012), and Von Briel (2018) to measure the organizational omnichannel capabilities. To measure the performance, we have used the three scales derived from Ben-Oz and Greve (2015), and Czakon et al. (2023) for long and short-term performance, supplemented by items used to measure the comparative performance from Czakon et al. (2020), as each of the scales captures a distinct yet interrelated aspect of firm performance, thus offering a more comprehensive perspective. On the one hand, short- and long-term focused scales refer to self-assessment of performance with respect to aspects evaluated over a shorter and longer time horizon, respectively. On the other hand, comparative scale refers to the assessment of performance in relation to – that is, in comparison with – the firm’s main direct competitors. The inclusion of the perception of relative standing not only complements the assessment of short-term and long-term performance but also aligns with performance measurement approaches adopted in prior seminal academic research (e.g. Morgan et al., 2009). Finally, a scale to measure industry drivers was based on the literature (Hajdas et al., 2022), and it consisted of 12 items. To discover which sub-constructs can be included in the industry drivers, we conducted an exploratory factor analysis (EFA).
We employed non-orthogonal rotation because the dimensions of industry drivers showed high inter-item correlations (Watkins, 2021). Promax was used as the rotation method. We did not impose a number of factors, as the EFA aims to identify the structure of the latent theoretical construct. We assessed the adequacy of our data for sampling through the Kaiser-Meyer-Olkin (KMO) and Bartlett’s test. The results confirmed that our data was appropriate for factor analysis, with a KMO value of 0.819 (greater than 0.5) and a statistically significant Bartlett’s test result (p = 0.00, less than 0.5) (Hair et al., 2019a). Three sub-constructs of industry drivers were identified: internationalization-related, micro-environment, and product-related industry drivers, covering respectively 4, 5, and 3 items (Table 1). We have evaluated these sub-constructs’ potential risk of common method bias (CMB) using the unrotated factor solution and Harman’s single-factor test (Sharma et al., 2009). The analysis indicated that no single factor dominated the solution, and no general factor accounted for most of the covariance. There is no evidence of CMB, as the unrotated solution revealed 12 factors with eigenvalues greater than 1, collectively explaining 61.570% of the total variance. Additionally, the factor with the highest eigenvalue explained only 28.414% of the variance, well below half. These findings confirm the absence of CMB risk.
Research results
| Convergent and discriminant validity | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| CR | AVE | Operational efficiency | Strategy and organizational culture | Micro-environment | Product-related | Internationalization-related | Level of channel integration | Short-term | Comparative | Long-term | |
| Operational efficiency | 0.945 | 0.635 | 0.797 | ||||||||
| Strategy and organizational culture | 0.936 | 0.596 | 0.612 | 0.772 | |||||||
| Micro-environment | 0.722 | 0.468 | −0.094 | −0.283 | 0.684 | ||||||
| Product-related | 0.561 | 0.395 | 0.333 | 0.128 | 0.382 | 0.628 | |||||
| Internationalization-related | 0.883 | 0.656 | 0.273 | 0.052 | 0.502 | 0.715 | 0.81 | ||||
| Level of channel integration | 0.908 | 0.473 | −0.332 | −0.482 | 0.398 | 0.17 | 0.136 | 0.688 | |||
| Short-term | 0.855 | 0.598 | −0.362 | −0.421 | 0.53 | 0.118 | 0.175 | 0.648 | 0.773 | ||
| Comparative | 0.854 | 0.597 | −0.25 | −0.403 | 0.514 | 0.218 | 0.211 | 0.532 | 0.716 | 0.772 | |
| Long-term | 0.853 | 0.592 | −0.304 | −0.395 | 0.587 | 0.159 | 0.206 | 0.672 | 0.767 | 0.754 | 0.769 |
| Relationships between constructs in the structural model | ||||
|---|---|---|---|---|
| Path (hypothesis) | Standardized parameter | Result of hypothesis testing | ||
| Internal obstacles | ||||
| Level of channel integration | ← | Operational efficiency | −0.114** | H1: supported |
| Level of channel integration | ← | Strategy and organizational culture | −0.162** | |
| Industry drivers | ||||
| Level of channel integration | ← | Micro-environment drivers | 0.241* | H2: partially supported |
| Level of channel integration | ← | Internationalization-related industry drivers | −0.193* | |
| Level of channel integration | ← | Product-related industry drivers | 0.198* | |
| Organizational omnichannel capabilities | ||||
| Level of channel integration | ← | Moderator capabilities-obstacles | 0.117** | H3: supported |
| Level of channel integration | ← | Moderator capabilities-drivers | −0.017 | H4: rejected |
| Performance | ||||
| Short-term performance | ← | Level of channel integration | 0.882*** | H5a: supported |
| Long-term performance | ← | Level of channel integration | 0.713*** | H5b: supported |
| Comparative performance | ← | Level of channel integration | 0.776*** | H5c: supported |
| Items, constructs, sub-constructs, and Cronbach’s alpha | |||
|---|---|---|---|
| Items | Code | Construct/sub-constructs with references | Cronbach’s α |
| My company has inefficiency in logistics operations | OE8 | Operational efficiency (Radomska et al., 2025) | 0.949 |
| My company has difficulty measuring the returned volume | OE3 | ||
| My company has misplaced products | OE6 | ||
| My company has product restocking problems | OE1 | ||
| My company has a store forecast imbalance | OE4 | ||
| My company has an increase in operating costs | OE2 | ||
| My company has difficulty with receiving orders | OE9 | ||
| My company has difficulty in managing the processing of order | OE10 | ||
| My company has high return costs | OE11 | ||
| My company has the risk of stockout | OE5 | ||
| My company has an inconsistent organizational strategy | SC8 | Strategy and organizational culture (Radomska et al., 2025) | 0.94 |
| My company has an ineffective communication strategy | SC9 | ||
| My company has a low level of knowledge and information sharing between the departments | SC3 | ||
| My company has no measurement process for channel efficiency | SC11 | ||
| My company has misaligned corporate motivations | SC5 | ||
| My company has difficulties in providing a consistent consumer experience | SC1 | ||
| My company has no personnel skilled in channel integration capabilities | SC2 | ||
| My company has a problem with a willingness to share information or knowledge between employees or across different departments within a company | SC7 | ||
| My company has a low level of organizational learning competencies | SC4 | ||
| My company has different mindsets between departments concerning how to integrate the different channels | SC6 | ||
| In our industry, there is a global demand for our products | DI4 | Internationalization-related industry drivers (Hajdas et al., 2022) | 0.874 |
| In our industry, marketing regulations are diversified across countries | DI10 | ||
| In our industry, tariffs and taxation systems are diversified across countries | DI11 | ||
| In our industry, currency is common across geographical markets | DI12 | ||
| In our industry, the customer needs for omnichannel are already developed | DC1 | Micro-environment drivers (Hajdas et al., 2022) | 0.751 |
| In our industry, the customers are literate about various omnichannel solutions | DC2 | ||
| In our industry, our competitors are already using an omnichannel strategy | DC5 | ||
| In our industry, there are practices of cooperation with competitors (coopetition) related to building omnichannel solutions | DC6 | ||
| In our industry, logistics efficiency is high | DC7 | ||
| In our industry, there is a local demand for our products | DP3 | Product-related industry drivers (Hajdas et al., 2022) | 0.707 |
| In our industry, products are perishable | DP8 | ||
| In our industry, products have a luxury appeal | DP9 | ||
| My company has a well-developed IT infrastructure | OC1 | Organizational omnichannel capabilities (Luo et al., 2016; Oh et al., 2012; Von Briel, 2018) | 0.947 |
| My company has a well-developed enterprise resource planning | OC2 | ||
| My company has a well-developed supply-chain management systems (SCM) | OC3 | ||
| My company has a well-developed order-management systems | OC4 | ||
| My company has a well-developed data mining | OC5 | ||
| My company has a well-developed business intelligence | OC6 | ||
| My company has a well-developed customer relationship management (CRM) | OC7 | ||
| In my company, staff at the physical stores knows about the products/services provided at the Website | OC8 | ||
| In my company, the staff understands our cross-channel integration strategies | OC9 | ||
| In my company, the staff has the ability to implement our cross-channel integration strategies | OC10 | ||
| In my company, the staff is competent in the use of information technology to support our cross-channel integration strategies | OC11 | ||
| In my company, we have the ability to reduce distribution costs | OC12 | ||
| In my company, we have the ability to reduce customer service costs | OC13 | ||
| In my company, we have the ability to involve customers in personalizing their shopping experience | OC14 | ||
| In my company, we have the ability to differentiate our products/services from those of our competitors | OC15 | ||
| In my company, we have the ability to launch new marketing strategies | OC16 | ||
| In my company, we have the ability to provide new ways of performing transactions | OC17 | ||
| In my company, we have the ability to offer new ways of order fulfilment | OC18 | ||
| In my company, we have the ability to reallocate resources quickly in response to changes in market conditions | OC19 | ||
| In my company, we have the ability to adjust the organizational mindset to the major challenges in channel integration | OC20 | ||
| In my company, we have the ability to adjust C-level skills to the major challenges in channel integration | OC21 | ||
| In my company, we have the ability to adjust store associate skills to the major challenges in channel integration | OC22 | ||
| In my company, we have the ability to increase operational productivity in all channels | OC23 | ||
| In my company, we have the ability to enable integrated (multi) brand management in all channels | OC24 | ||
| In my company, we have the ability to enable real-time inventory management in all channels | OC25 | ||
| In my company, we have the ability to provide customer profiling in all channels | OC26 | ||
| In my company, we have the ability to optimize the conversion rate in all channels | OC27 | ||
| In my company, we have the ability to provide real-time analytics in all channels | OC28 | ||
| In my company, we have the ability to provide real-time information dissemination in all channels | OC29 | ||
| My company has well-developed aligned services across channels | CI1 | Level of channel integration (Cao and Li, 2015; Shi et al., 2020) | 0.915 |
| My company has a well-developed aligned price across channels | CI2 | ||
| My company has a well-developed aligned loyalty program across channels | CI3 | ||
| My company has a well-developed aligned assortment across channels | CI4 | ||
| My company has a well-developed integration of information systems across channels | CI5 | ||
| My company has a well-developed integration of a database of clients across channels | CI6 | ||
| In my company, the customer’s interactions across different channels are integrated | CI7 | ||
| In my company, the descriptions of products are integrated across different channels | CI8 | ||
| In my company, new product launches are synchronous across different channels | CI9 | ||
| In my company, the product attributes can be equally allocated across different channels | CI10 | ||
| In my company, the promotion activities are aligned across different channels | CI11 | ||
| Meeting sales objectives | SP1 | Short-term performance (Ben-Oz and Greve, 2015; Czakon et al., 2023) | 0.759 |
| Achieving sales growth | SP2 | ||
| Meeting profitability targets | SP3 | ||
| Increasing profitability | SP4 | ||
| Meeting the company’s strategic goals | LP5 | Long-term performance (Ben-Oz and Greve, 2015; Czakon et al., 2023) | 0.863 |
| Introducing new products/services | LP6 | ||
| Introducing more new service products than competitors | LP7 | ||
| New products/services achieve market success | LP8 | ||
| Sales | CP1 | Comparative performance (Czakon et al., 2020) | 0.856 |
| Profit | CP2 | ||
| Market share | CP3 | ||
| Return on investment | CP4 | ||
Note(s): The order of items is presented based on the level of factor loading
*p < 0.05; **p < 0.01; ***p < 0.001
The testing of hypotheses involving latent variables, as well as the intention to jointly capture the relationships between all of the variables, determined the choice of structural equation modelling (SEM) as the method for hypothesis testing. Given that the primary goal of our investigation was to test theory-driven assumptions rather than prediction or theory development, our hypotheses were analysed using covariance-based structural equation modelling (CB-SEM) to evaluate the significance and quality of the developed and tested model (Hair et al., 2019b). In line with the key assumptions of CB-SEM (Hair et al., 2017), we first specified reflective measurement models and ensured that the measurement scales underwent multidimensional validation. We then applied 7-point measurement scales, which can be considered continuous and are required for parametric methods (Ibrahim, 2025). Finally, we ensured that the sample size was adequate for covariance-based SEM (n = 412 exceeds the recommended minimum of n > 100). In order to evaluate the model fit, we used typical complementary indicators including standardized χ2, RMSEA, GFI, AGFI, IFI, TLI, CFI, PGFI, and PNFI. To enhance the model’s fit, we conducted a modification index (MI) analysis (MacCallum et al., 1992). Typically, if MI > 4, error covariances can be introduced within the factor. However, in our model, we applied a stricter criterion, using MI > 10.
All analyses have been done using IBM SPSS Statistics software (ver. 29) and IBM SPSS Amos (ver. 29). We present the final set of items in Table 1.
Results
Scales validation
We have validated the adopted measurement in several ways. First, we assessed the internal consistency of the measures by conducting a reliability test. All Cronbach’s alpha values fall within the acceptable range, between 0.7 and 0.95, meeting the required standards (Hair et al., 2019a). Next, we verified convergent validity by calculating standardised factor loadings, composite reliability (CR), and average variance extracted (AVE). All standardised factor loadings (except for DC1, DC2, and DP3 – removed from the final model) are above 0.5, as expected in the literature (Hair et al., 2019a). The CR values (except for product-related industry drivers) are greater than 0.7 (Bagozzi et al., 1991). All AVEs, except for product-related and micro-environment drivers and level of channel integration, are above 0.5 (Fornell and Larcker, 1981) (see Table 1). The convergent validity is still acceptable if the AVE is at least 0.4 and the CR exceeds 0.6 (Fornell and Larcker, 1981). This condition is satisfied for micro-environment drivers and level of channel integration, but not for product-related industry drivers. The square root of the AVE for each variable (except for product-related industry drivers) is higher than the absolute value of its correlations with any other variable, further confirming discriminant validity. The results of our analyses provide reliability support, as well as convergent and discriminant validity (except for product-related industry drivers).
Model testing
The path diagram starts on the left with five circles arranged in a vertical series. From top to bottom, they are labeled as follows: “Operational efficiency,” “Strategy and organizational culture,” “Internationalization-related industry drivers,” “Micro-environment drivers,” and “Product-related industry drivers.” From “Operational efficiency,” eleven individual leftward arrows connect to ten vertically arranged rectangles labeled from top to bottom as follows: The first arrow, with a path coefficient of 0.78, points to “O E 1.” The second arrow, with a path coefficient of 0.74, points to “O E 2.” The third arrow, with a path coefficient of 0.77, points to “O E 3.” The fourth arrow, with a path coefficient of 0.77, points to “O E 4.” The fifth arrow, with a path coefficient of 0.75, points to “O E 5.” The sixth arrow, with a path coefficient of 0.84, points to “O E 6.” The seventh arrow, with a path coefficient of 0.85, points to “O E 8.” The eighth arrow, with a path coefficient of 0.81, points to “O E 9.” The ninth arrow, with a path coefficient of 0.83, points to “O E 10.” The tenth arrow, with a path coefficient of 0.82, points to “O E 11.” From “Strategy and organizational culture,” ten individual leftward arrows connect to eleven vertically arranged rectangles labeled from top to bottom as follows: The first arrow, with a path coefficient of 0.79, points to “S C 1.” The second arrow, with a path coefficient of 0.77, points to “S C 2.” The third arrow, with a path coefficient of 0.84, points to “S C 3.” The fourth arrow, with a path coefficient of 0.80, points to “S C 4.” The fifth arrow, with a path coefficient of 0.82, points to “S C 5.” The sixth arrow, with a path coefficient of 0.65, points to “S C 6.” The seventh arrow, with a path coefficient of 0.66, points to “S C 7.” The eighth arrow, with a path coefficient of 0.79, points to “S C 8.” The ninth arrow, with a path coefficient of 0.79, points to “S C 9.” The tenth arrow, with a path coefficient of 0.80, points to “S C 11.” From “Internationalization-related industry drivers,” four individual leftward arrows connect to four vertically arranged rectangles labeled from top to bottom as follows: The first arrow, with a path coefficient of 0.66, points to “D I 4.” The second arrow, with a path coefficient of 0.83, points to “D I 10.” The third arrow, with a path coefficient of 0.89, points to “D I 11.” The fourth arrow, with a path coefficient of 0.85, points to “D I 12.” From “Micro-environment drivers,” three individual leftward arrows connect to three vertically arranged rectangles labeled from top to bottom as follows: The first arrow, with a path coefficient of 0.54, points to “D C 5.” The second arrow, with a path coefficient of 0.62, points to “D C 6.” The third arrow, with a path coefficient of 0.60, points to “D C 7.” From “Product-related industry drivers,” two individual leftward arrows connect to two vertically arranged rectangles labeled from top to bottom as follows: The first arrow, with a path coefficient of 0.73, points to “D P 8.” The second arrow, with a path coefficient of 0.54, points to “D P 9.” Each oval is also connected with curved arrows to the other ovals as follows: The arrow from “Operational efficiency” to “Strategy and organizational culture” has a path coefficient of 0.61. The arrow from “Operational efficiency” to “Internationalization-related industry drivers” has a path coefficient of 0.28. The arrow from “Operational efficiency” to “Micro-environment drivers” has a path coefficient of negative 0.14. The arrow from “Operational efficiency” to “Product-related industry drivers” has a path coefficient of 0.33. The arrow from “Strategy and organizational culture” to “Internationalization-related industry drivers” has a path coefficient of 0.05. The arrow from “Strategy and organizational culture” to “Micro-environment drivers” has a path coefficient of 0.71. The arrow from “Strategy and organizational culture” to “Product-related industry drivers” has a path coefficient of 0.13. The arrow from “Internationalization-related industry drivers” to “Micro-environment drivers” has a path coefficient of 0.58. The arrow from “Internationalization-related industry drivers” to “Product-related industry drivers” has a path coefficient of negative 0.36. The arrow from “Micro-environment drivers” to “Product-related industry drivers” has a path coefficient of 0.36. Each of the five circles has a rightward arrow connecting to an oval in the center labeled “Level of channel integration.” The path coefficients, from top to bottom, are: The arrow from “Operational efficiency” to “Level of channel integration” has a path coefficient of negative 0.11. The arrow from “Strategy and organizational culture” to “Level of channel integration” has a path coefficient of negative 0.16. The arrow from “Internationalization-related industry drivers” to “Level of channel integration” has a path coefficient of negative 0.19. The arrow from “Micro-environment drivers” to “Level of channel integration” has a path coefficient of 0.24. The arrow from “Product-related industry drivers” to “Level of channel integration” has a path coefficient of 0.20. From “Level of channel integration,” eleven individual upward arrows connect to eleven horizontally arranged boxes labeled from left to right as follows: The first arrow, with a path coefficient of 0.63, points to “C I 1.” The second arrow, with a path coefficient of 0.66, points to “C I 2.” The third arrow, with a path coefficient of 0.63, points to “C I 3.” The fourth arrow, with a path coefficient of 0.62, points to “C I 4.” The fifth arrow, with a path coefficient of 0.61, points to “C I 5.” The sixth arrow, with a path coefficient of 0.61, points to “C I 6.” The seventh arrow, with a path coefficient of 0.64, points to “C I 7.” The eighth arrow, with a path coefficient of 0.61, points to “C I 8.” The ninth arrow, with a path coefficient of 0.68, points to “C I 9.” The tenth arrow, with a path coefficient of 0.66, points to “C I 10.” The eleventh arrow, with a path coefficient of 0.68, points to “C I 11.” Above the center, two rectangles are labeled “Moderator capabilities - obstacles” and “Moderator capabilities - drivers,” connected to “Level of channel integration,” with path coefficients of 0.12 and negative 0.02, respectively. From “Level of channel integration,” three rightward arrows point to three vertically arranged ovals labeled from top to bottom as “Short-term performance,” “Long-term performance,” and “Comparative performance.” The arrow from “Level of channel integration” to “Short-term performance” has a path coefficient of 0.88. The arrow from “Level of channel integration” to “Long-term performance” has a path coefficient of 0.71. The arrow from “Level of channel integration” to “Comparative performance” has a path coefficient of 0.78. On the right of “Short-term performance,” four boxes are arranged vertically and point back with a left arrow to “Short-term performance” with path coefficients as follows: The arrow from “S P 1” to “Short-term performance” has a path coefficient of 0.67. The arrow from “S P 2” to “Short-term performance” has a path coefficient of 0.79. The arrow from “S P 3” to “Short-term performance” has a path coefficient of 0.78. The arrow from “S P 4” to “Short-term performance” has a path coefficient of 0.77. On the right of “Long-term performance,” four boxes are arranged vertically and point back with a left arrow to “Long-term performance” with path coefficients as follows: The arrow from “L P 5” to “Long-term performance” has a path coefficient of 0.81. The arrow from “L P 6” to “Long-term performance” has a path coefficient of 0.76. The arrow from “L P 7” to “Long-term performance” has a path coefficient of 0.74. The arrow from “L P 8” to “Long-term performance” has a path coefficient of 0.76. On the right of “Comparative performance,” four boxes are arranged vertically and point back with a left arrow to “Comparative performance” with path coefficients as follows: The arrow from “C P 1” to “Comparative performance” has a path coefficient of 0.70. The arrow from “C P 2” to “Comparative performance” has a path coefficient of 0.89. The arrow from “C P 3” to “Comparative performance” has a path coefficient of 0.80. The arrow from “C P 4” to “Comparative performance” has a path coefficient of 0.67.Structural model. Source: Authors’ own work
The path diagram starts on the left with five circles arranged in a vertical series. From top to bottom, they are labeled as follows: “Operational efficiency,” “Strategy and organizational culture,” “Internationalization-related industry drivers,” “Micro-environment drivers,” and “Product-related industry drivers.” From “Operational efficiency,” eleven individual leftward arrows connect to ten vertically arranged rectangles labeled from top to bottom as follows: The first arrow, with a path coefficient of 0.78, points to “O E 1.” The second arrow, with a path coefficient of 0.74, points to “O E 2.” The third arrow, with a path coefficient of 0.77, points to “O E 3.” The fourth arrow, with a path coefficient of 0.77, points to “O E 4.” The fifth arrow, with a path coefficient of 0.75, points to “O E 5.” The sixth arrow, with a path coefficient of 0.84, points to “O E 6.” The seventh arrow, with a path coefficient of 0.85, points to “O E 8.” The eighth arrow, with a path coefficient of 0.81, points to “O E 9.” The ninth arrow, with a path coefficient of 0.83, points to “O E 10.” The tenth arrow, with a path coefficient of 0.82, points to “O E 11.” From “Strategy and organizational culture,” ten individual leftward arrows connect to eleven vertically arranged rectangles labeled from top to bottom as follows: The first arrow, with a path coefficient of 0.79, points to “S C 1.” The second arrow, with a path coefficient of 0.77, points to “S C 2.” The third arrow, with a path coefficient of 0.84, points to “S C 3.” The fourth arrow, with a path coefficient of 0.80, points to “S C 4.” The fifth arrow, with a path coefficient of 0.82, points to “S C 5.” The sixth arrow, with a path coefficient of 0.65, points to “S C 6.” The seventh arrow, with a path coefficient of 0.66, points to “S C 7.” The eighth arrow, with a path coefficient of 0.79, points to “S C 8.” The ninth arrow, with a path coefficient of 0.79, points to “S C 9.” The tenth arrow, with a path coefficient of 0.80, points to “S C 11.” From “Internationalization-related industry drivers,” four individual leftward arrows connect to four vertically arranged rectangles labeled from top to bottom as follows: The first arrow, with a path coefficient of 0.66, points to “D I 4.” The second arrow, with a path coefficient of 0.83, points to “D I 10.” The third arrow, with a path coefficient of 0.89, points to “D I 11.” The fourth arrow, with a path coefficient of 0.85, points to “D I 12.” From “Micro-environment drivers,” three individual leftward arrows connect to three vertically arranged rectangles labeled from top to bottom as follows: The first arrow, with a path coefficient of 0.54, points to “D C 5.” The second arrow, with a path coefficient of 0.62, points to “D C 6.” The third arrow, with a path coefficient of 0.60, points to “D C 7.” From “Product-related industry drivers,” two individual leftward arrows connect to two vertically arranged rectangles labeled from top to bottom as follows: The first arrow, with a path coefficient of 0.73, points to “D P 8.” The second arrow, with a path coefficient of 0.54, points to “D P 9.” Each oval is also connected with curved arrows to the other ovals as follows: The arrow from “Operational efficiency” to “Strategy and organizational culture” has a path coefficient of 0.61. The arrow from “Operational efficiency” to “Internationalization-related industry drivers” has a path coefficient of 0.28. The arrow from “Operational efficiency” to “Micro-environment drivers” has a path coefficient of negative 0.14. The arrow from “Operational efficiency” to “Product-related industry drivers” has a path coefficient of 0.33. The arrow from “Strategy and organizational culture” to “Internationalization-related industry drivers” has a path coefficient of 0.05. The arrow from “Strategy and organizational culture” to “Micro-environment drivers” has a path coefficient of 0.71. The arrow from “Strategy and organizational culture” to “Product-related industry drivers” has a path coefficient of 0.13. The arrow from “Internationalization-related industry drivers” to “Micro-environment drivers” has a path coefficient of 0.58. The arrow from “Internationalization-related industry drivers” to “Product-related industry drivers” has a path coefficient of negative 0.36. The arrow from “Micro-environment drivers” to “Product-related industry drivers” has a path coefficient of 0.36. Each of the five circles has a rightward arrow connecting to an oval in the center labeled “Level of channel integration.” The path coefficients, from top to bottom, are: The arrow from “Operational efficiency” to “Level of channel integration” has a path coefficient of negative 0.11. The arrow from “Strategy and organizational culture” to “Level of channel integration” has a path coefficient of negative 0.16. The arrow from “Internationalization-related industry drivers” to “Level of channel integration” has a path coefficient of negative 0.19. The arrow from “Micro-environment drivers” to “Level of channel integration” has a path coefficient of 0.24. The arrow from “Product-related industry drivers” to “Level of channel integration” has a path coefficient of 0.20. From “Level of channel integration,” eleven individual upward arrows connect to eleven horizontally arranged boxes labeled from left to right as follows: The first arrow, with a path coefficient of 0.63, points to “C I 1.” The second arrow, with a path coefficient of 0.66, points to “C I 2.” The third arrow, with a path coefficient of 0.63, points to “C I 3.” The fourth arrow, with a path coefficient of 0.62, points to “C I 4.” The fifth arrow, with a path coefficient of 0.61, points to “C I 5.” The sixth arrow, with a path coefficient of 0.61, points to “C I 6.” The seventh arrow, with a path coefficient of 0.64, points to “C I 7.” The eighth arrow, with a path coefficient of 0.61, points to “C I 8.” The ninth arrow, with a path coefficient of 0.68, points to “C I 9.” The tenth arrow, with a path coefficient of 0.66, points to “C I 10.” The eleventh arrow, with a path coefficient of 0.68, points to “C I 11.” Above the center, two rectangles are labeled “Moderator capabilities - obstacles” and “Moderator capabilities - drivers,” connected to “Level of channel integration,” with path coefficients of 0.12 and negative 0.02, respectively. From “Level of channel integration,” three rightward arrows point to three vertically arranged ovals labeled from top to bottom as “Short-term performance,” “Long-term performance,” and “Comparative performance.” The arrow from “Level of channel integration” to “Short-term performance” has a path coefficient of 0.88. The arrow from “Level of channel integration” to “Long-term performance” has a path coefficient of 0.71. The arrow from “Level of channel integration” to “Comparative performance” has a path coefficient of 0.78. On the right of “Short-term performance,” four boxes are arranged vertically and point back with a left arrow to “Short-term performance” with path coefficients as follows: The arrow from “S P 1” to “Short-term performance” has a path coefficient of 0.67. The arrow from “S P 2” to “Short-term performance” has a path coefficient of 0.79. The arrow from “S P 3” to “Short-term performance” has a path coefficient of 0.78. The arrow from “S P 4” to “Short-term performance” has a path coefficient of 0.77. On the right of “Long-term performance,” four boxes are arranged vertically and point back with a left arrow to “Long-term performance” with path coefficients as follows: The arrow from “L P 5” to “Long-term performance” has a path coefficient of 0.81. The arrow from “L P 6” to “Long-term performance” has a path coefficient of 0.76. The arrow from “L P 7” to “Long-term performance” has a path coefficient of 0.74. The arrow from “L P 8” to “Long-term performance” has a path coefficient of 0.76. On the right of “Comparative performance,” four boxes are arranged vertically and point back with a left arrow to “Comparative performance” with path coefficients as follows: The arrow from “C P 1” to “Comparative performance” has a path coefficient of 0.70. The arrow from “C P 2” to “Comparative performance” has a path coefficient of 0.89. The arrow from “C P 3” to “Comparative performance” has a path coefficient of 0.80. The arrow from “C P 4” to “Comparative performance” has a path coefficient of 0.67.Structural model. Source: Authors’ own work
The model’s goodness of fit was assessed using various indices. The standardised χ2 (χ2/df) is 1.785, well below the threshold of 5. The root mean square error of approximation (RMSEA) is 0.044, indicating a good fit. The lower bound of the 90% confidence interval (LO 90) is near zero (0.041), and the upper bound (HI 90) is significantly lower than the maximum acceptable value of 0.080 (0.047). The goodness of fit index (GFI) is 0.828, and the adjusted goodness of fit index (AGFI) is 0.804. Although these values are below the ideal value of 0.9, they are still considered acceptable according to Baumgartner and Homburg (1996), who suggest a minimum acceptable value of 0.8. Other fit indices also demonstrate a strong model fit, with the incremental fit index (IFI), Tucker-Lewis index (TLI), and comparative fit index (CFI) all reaching the required minimum of 0.9 (0.930, 0.923, and 0.930, respectively). Parsimony fit indices also meet the criteria for a well-fitting model, with the parsimony goodness of fit index (PGFI) and parsimony normed fit index (PNFI) both exceeding 0.6, at 0.726 and 0.776, respectively.
The analysed model achieved satisfactory fit measures based on the empirical data, providing a foundation for further analysis and the evaluation of research hypotheses. Table 1 presents the parameters for the paths defined in the model. Internal obstacles, i.e. operational efficiency, strategy, and organizational culture, negatively affect the level of channel integration (H1). Thus, hypothesis H1: Internal obstacles negatively impact the level of channel integration has been supported. Regarding industry drivers (H2), micro-environment drivers have a significant positive effect on the level of channel integration. In contrast, internationalization-related industry drivers have a statistically significant negative impact on channel integration. Product-related industry drivers, which have a positive effect, did not meet the reliability criteria. Therefore, hypothesis H2: Industry drivers impact the level of channel integration has only been partially supported. Next, we tested whether organizational omnichannel capabilities moderate the impact of internal obstacles on channel integration (H3) and industry drivers on channel integration (H4). Hypothesis H3: Organizational omnichannel capabilities weaken the negative impact of internal obstacles on the level of channel integration has been positively tested, but H4: Organizational omnichannel capabilities moderate the relationship between industry drivers and the level of channel integration has not. It means that organizational capabilities mitigate the negative relationship between internal obstacles and the level of channel integration – the greater the level of organizational capabilities, the less impact internal obstacles have on the level of channel integration. However, organizational capabilities do not moderate the relationship between industry drivers and the level of channel integration. Finally, the impact of the level of channel integration on performance has been examined. It has been found to have a strong and statistically significant impact on comparative (H5a), short-term (H5b) and long-term (H5c) performance, highlighting its crucial role in driving firm performance and underlying competitiveness. Therefore, hypotheses H5a: The level of channel integration positively impacts comparative performance, H5b: The level of channel integration positively impacts short-term performance, and H5c: The level of channel integration positively impacts long-term performance have been supported. The results of all hypothesis testing are presented in Table 1.
Discussion and conclusions
Referring to our RQ1 and RQ2 our findings indicate that the level of integration is sensitive to the negative influence of internal obstacles (H1), and internationalization-related industry factors while remaining positively influenced by micro-environment drivers (H2). It confirms our conceptualization of an industry driver as a neutral factor reflecting a particular industry feature, which needs to be conceptualised on a scale of binary oppositions. Referring to our RQ3 we show that the negative impact of internal factors can be mitigated through the development of organizational omnichannel capabilities (H3), which largely include managerial capabilities, skills, and solutions related to channel management, coordination, as well as logistics and resource allocation. However, these capabilities do not moderate the effect of industry drivers on the level of channel integration (H4). Finally, we examine the factors directly and indirectly (using the lenses of internal and external factors) impacting multidimensionally perceived performance and demonstrate that channel integration is indeed a reasonable strategy for enhancing firm performance (H5a–c).
We expand a common, short-term approach to channel integration performance (Song et al., 2019) by showing that the positive effects of channel integration are significant also in the long-term run. Achieving a certain level of retail complexity with integrated channels (e.g. physical store retailing, multichannel retailing, or omnichannel retailing) should be viewed as a long-term endeavour. It is typically the result of long-term efforts focused on allocating new resources and reallocating existing ones, where the benefits are realised over an extended period.
This research, on the one hand, contributes to the ongoing debate on the importance of internal factors, reinforcing earlier studies highlighting the significance of operational efficiency (de Borba et al., 2021), strategic approach (Radomska et al., 2025), and the role of organizational culture (Radomska et al., 2025) for channel integration. On the other hand, in light of the conducted research, the level of integration, consistent with the previous suggestions by Hajdas et al. (2022), is also shaped by external factors, i.e. industry drivers. However, contrary to conclusions drawn in previous studies (Hajdas et al., 2022), this influence is not exclusively stimulating, as the significance of external factors proves to be more complex and not always positive. In the empirical context selected for this study, in line with logical expectations, factors related to internationalization (e.g. market and regulatory heterogeneity across countries) reduce the level of integration. Meanwhile, customer-related factors (e.g. customer knowledge and needs regarding omnichannel solutions) enhance it.
Theory contribution
First, we contribute to extant omnichannel scholarship by offering a validated framework showing that channel integration efforts are affected not only by the scope of internal obstacles, but also by external industry drivers. Moreover, our framework explains the role of organisational omnichannel capabilities in dealing with internal challenges and allows for meaningful capture of interaction dynamics between these variables in channel integration studies.
Second, our study introduces a new construct to the discussion on the challenges of channel integration. By conceptualising industry drivers and investigating this construct in a broad scope of industries, we claim that not all industries are equally welcoming in terms of omnichannel operations, confirming previous suggestions (Hajdas et al., 2022; Iglesías-Pradas et al., 2021). Our findings show that the micro-environment, including the omnichannel-literate consumer base and competitors, are the factors increasing the industry’s omnichannel potential. On the other hand, the industries where international strategies dominate pose more challenges regarding channel integration. We hope our results will facilitate the debate on the role of contextual factors in omnichannel studies, as omnichannel theories dealing with multiple contexts are lacking, limiting our understanding of which theories are universal and which might be framed by the particular context(s). We, therefore, encourage to enhance contextual appreciation (McLaren and Durepos, 2019) in future omnichannel studies.
Third, we suggest refining the omnichannel organizational capabilities concept offered by Yumurtacı Hüseyinoğlu et al. (2018), which focused on external service and delivery issues. We consider the managerial ability to identify, understand, and navigate internal complexities crucial for omnichannel strategies and operations. We suggest that the refined concept covering both external customer services and internal managerial excellence, can offer greater explanatory power in terms of firms’ advancements in channel integration efforts.
Finally, contrary to previous studies that captured performance as a static and a short-term state (Tagashira and Minami, 2019), our study adopted a more comprehensive approach (Ricciardi et al., 2022), including three perspectives: short-term, long-term, and comparative performance to empirically investigate the effects of the level of channel integration, expanding existing findings (Yan et al., 2021; Kolbe et al., 2022) by confirming that the positive effects of channel integration fuel performance not only in the short term but also in the long-term run.
Together, our findings provide a robust conceptual and empirical basis for further developing omnichannel theory by making it more sensitive to context, offering multidimensional perspectives on performance, and improving our understanding of the internal capabilities that shape integration outcomes.
Managerial and social implications
We believe our findings suggest several courses of action for managers responsible for channel integration. First, to improve channel integration, omnichannel strategies should be tailored to industry-specific insights. Understanding the industry characteristics and aligning them with the operational systems of omnichannel implementation requires diversifying the KPIs used to measure the channel integration progress, especially in the case of diversified firms. Second, our study shows that some industries are more advanced in omnichannel solutions, therefore they may serve as benchmarks for laggards in the field. Omnichannel managers should not only conduct competitive audits within their industry but also identify learning opportunities outside their domains to engage in cross-industrial transfers of knowledge and eventually advance their omnichannel practices.
Our study also offers several societal implications. First, although our results show that an existing customer base already literate with omnichannel solutions positively influences the level of channel integration, firms could increase their customer base by promoting digital accessibility and creating inclusive environments for other customer segments, including those with limited digital literacy (Klaus et al., 2024). Second, our results show that a higher level of organizational capabilities dampens the adverse impact of internal obstacles on the level of channel integration. It may have implications for workforce transformation, managerial teaching, and training in terms of managerial roles, requiring reskilling into more cross-functional ones. Third, our results also have an environmental significance, as poor channel integration often leads to inefficient logistics, shipment, or increased product returns, which contributes to environmental harm. Our study may inspire greener organizational practices, which will eventually lead to reducing the carbon footprint of omnichannel solutions.
Limitations and further research
Some limitations of our study should be noted. First and foremost, the empirical context is limited to a single country. While this choice is justified by the high level of omnichannel adoption in the selected country and restricting retailing research to one country is relatively common (e.g. Stanca et al., 2025), it nevertheless prevents drawing general conclusions. Therefore, conducting studies in other geographical contexts is recommended, for example, in a country or countries with significantly lower levels of omnichannel adoption. To achieve broader generalizations, since the present study has removed the limitation of a single industrial context, the next step should involve research that exceeds the boundaries of a single country. Alternatively, our results – particularly the industry drivers’ role in channel integration – create research avenues for context-sensitive omnichannel studies that would result in theories that are contextually oriented.
Second, during the analysis, the risk of common method bias was assessed using Harman’s single-factor test (following, for instance, Sharma et al., 2009). While this approach is dominant in social science research, it is worth noting the recent methodological discussion indicating that this analytical approach alone is insufficient (Podsakoff et al., 2024). Therefore, we recommend to adopt a broader perspective on CMB verification in future investigations (see, e.g. Sharma and Fatima, 2024).
Third, we have only focused our attention on profitability as the effect of channel integration. Although we referred to various profitability perspectives (regarding the horizon and the comparison to the competitors), it may remain limited. Future studies could be devoted to investigating other effects, such as the intangible firm value (Kang et al., 2018).
Fourth, our sample covered retailers that have developed two channels (online and offline). However, although such focus is quite common (e.g. Zelke and Komor, 2025), a more nuanced and contextual study where the channel-specific goals and rewards (Berman and Theler, 2018) are investigated could bring more narrow but valuable idiosyncrasies.
Finally, this study explores the role of internal obstacles and industry divers in the context of omnichannel retailing. It may be valuable to replicate research in the novel context of experiential retail territories (ERT) (Alexander and Varley, 2025). Also, our study can serve as a preliminary step in the process of omnichannel industry drivers’ scale development.
Ethical considerations
The research underpinning the article was positively evaluated by the Ethics Committee of the Wroclaw University of Economics and Business – decision no. 38/2022.
Declaration of not using generative AI in scientific writing
We did not use generative artificial intelligence (AI) and AI-assisted technologies in the writing or subsequent stages of the research process. We used Chat GPT and Grammarly as tools for some proofreading (checking grammar and improving transparency) and some translations. After using Chat GPT/Grammarly for proofreading/translation purposes, we reviewed and edited the content if needed, and take full responsibility for the content of the article.
Submission declaration and verification
The manuscript has not been published previously, it is not under consideration for publication elsewhere, and its publication is approved by all authors and tacitly or explicitly by the responsible authorities where the work was carried out. If accepted, it will not be published elsewhere in the same form, in English, or any other language, including electronically, without the written consent of the copyright holder.
Notes
The investigated firms operate in industries such as computing, telecommunication equipment, electronics, musical instruments, photography and video, sports equipment, jewellery, beauty and cosmetics, wines, apparel, collectibles, pet items, automotive accessories, craft supplies, home and garden, office supplies, toys, arts, home furnishing, gifts and holidays, CDs and DVDs, books.
