Hybrid Selling: The Optimal Mix of Face-to-Face and Remote Sales Calls with Existing B2B Customers

Abstract

Growing cost pressures and customer demands toward digital engagement at the organizational frontline have accelerated business-to-business (B2B) salespeople’s adoption of hybrid selling, in which the same frontline salesperson combines face-to-face and remote (video or phone) sales calls to service the same customer. However, many salespeople fear that efficiency and potential interaction quantity gains come at the cost of interaction quality losses, potentially harming their customer relationships. We conceptualize hybrid selling and establish its financial and relational consequences based on two field studies with a leading European B2B manufacturer. Study 1 establishes an inverted U-shaped sales performance effect using customer relationship management (CRM) data from 3,565 customers (15,172 interactions). Study 2 matches customer survey and CRM data for additional 651 customers (5,971 interactions) to extend sales performance findings to relationship quality and show that product complexity, customer channel preferences, and salesperson interaction knowledge shape its effectiveness. We find that, in comparison to pure face-to-face selling, the optimal hybrid mix can increase performance by up to 27 percent in some interaction contexts, while in others, a pure remote approach can harm performance by up to 32 percent. Our results provide guidelines for salespeople and their executives when developing effective hybrid selling strategies to improve B2B performance.

Graphical Abstract

Keywords

hybrid selling sales management customer–salesperson interaction remote selling personal selling

Introduction

For business-to-business (B2B) firms faced with increasing demands to increase sales efficiency and lower selling costs, the simultaneous demand from customers to offer digital engagement at the organizational frontline creates a pressing challenge (Plotkin et al. 2024; Travasoni et al. 2024). In response, many firms adopt hybrid selling, a selling approach in which a salesperson combines face-to-face and remote (via video or phone) sales calls to service the same customer. Hybrid sellers account for nearly half of B2B sales roles, and their share is expected to grow by 60 percent in coming years (Donchak et al. 2022, 2023), reflecting the appeal of a scenario in which salespeople, “freed from spending hours on the road traveling between meetings, [. . .] could pack more into their day and respond to customers in real-time” (English 2022). Some real-world evidence confirms such predictions. Remote interactions, relative to face-to-face versions, enable salespeople to cover up to four times more customer interactions with 50 percent lower selling costs (Cencioni et al. 2024). In turn, firms that have combined face-to-face with remote customer interactions reportedly achieve up to 20 percent to 50 percent revenue growth (Cencioni et al. 2024; Donchak et al. 2022).

For sales executives, such numbers underscore the upside potential of hybrid selling with greater efficiency, lower costs, and potentially more growth. Frontline salespeople appear more cautious though as remote calls may prove insufficient in complex selling situations (Ahearne et al. 2025), misalign with customers’ preferences (Godfrey et al. 2011), or limit the sorts of tacit knowledge sharing or informal discussions that can spark opportunities (Mangus et al. 2024; Moffett et al. 2021). As one key account manager noted, “half my opportunities come from informal customer discussions. And off-the-record conversations never happen over video” (Zoltners et al. 2022). Such perspectives highlight a central managerial tension as executives might prefer and expect hybrid selling to increase interaction quantity and growth, but salespeople fear that efficiency gains come at the cost of interaction quality and long-term relationship strength.

Academic insights into the relational and financial consequences of hybrid selling in B2B contexts remain scarce. That is, some studies provide conceptual arguments for hybrid selling (e.g., Ahearne et al. 2022; Moffet et al. 2021) but miss field-based empirical evidence of its relational and financial performance effects. Research on organizational frontlines shows that technology reshapes frontline interactions (Arnold and Marinova 2023; Singh et al. 2017), and multiformat communication research has indicated the effectiveness of synchronous interaction modes (e.g., face-to-face, video, and phone), depending on media richness and task complexity considerations (Mangus et al. 2024; Moffett et al. 2021). Furthermore, sales channel research details how firms might combine inside and field sales across channels (Ahearne et al. 2025; Shi et al. 2024; Sleep et al. 2020), though not how a single salesperson might alternate between face-to-face and remote sales calls with the same customer.

To conceptually develop how and when hybrid selling affects B2B relationship performance (i.e., sales performance and relationship quality), we propose a contingent approach based on media richness theory (Daft and Lengel 1986). We posit that the effectiveness of hybrid selling depends on the fit between the interaction mode and the requirements of the interaction (Mangus et al. 2024; Moffet et al. 2021). Remote calls generate efficiency gains that increase interaction quantity (Ahearne et al. 2025), but they exclude the rich proximal cues available during in-person interactions, which can foster tacit knowledge sharing, interpersonal involvement, and trust (Mangus et al. 2024; Marinova et al. 2018; Moffett et al. 2021). These countervailing mechanisms of interaction quantity benefits and interaction quality losses suggest an inverted U-shaped effect of hybrid selling on relationship performance. Its balance should be contingent on interaction requirements, such as product complexity, customer channel preferences, and salesperson interaction knowledge, which make face-to-face or remote interaction modes more or less effective in each case.

To test the conceptual model, we conduct two complementary studies with matched, large-scale field data gathered from a leading European B2B manufacturer. Study 1 uses eighteen months of customer relationship management (CRM) records, covering 3,565 customers (15,172 interactions in total, nested in twenty-seven salespeople), and tests for a main effect of hybrid selling on sales performance. Study 2 then integrates the CRM data with a customer survey, yielding a matched sample of a different set of 651 customers (5,971 interactions, nested in twenty-seven salespeople). With these multisource data that combine customers’ perspectives with the selling firm’s, we also test for a relationship quality effect and potential moderation by product complexity, customer channel preferences, and salesperson interaction knowledge.

These analyses, along with several robustness checks, reveal that hybrid selling exerts an inverted U-shaped effect on sales performance, such that moderate shares of remote calls outperform both purely face-to-face and purely remote approaches. In Study 2, we also identify an inverted U-shaped effect on customers’ perceived relationship quality. The strength and shape of the effects depend on contextual contingencies, such that product complexity accelerates interaction quality losses and shifts the curve to the left, whereas customer channel preference and salesperson interaction knowledge mitigate such losses and shift the curve to the right. Relationship quality partially mediates the link between hybrid selling and sales performance. The results indicate that effective hybrid selling demands a balance between remote and face-to-face calls, contingent on product-, customer-, and salesperson-related interaction requirements.

This study makes four main contributions to service and sales research. First, we establish hybrid selling as a growing organizational frontline phenomenon and recognize that value creation at the frontline depends on the availability of multiple interaction formats, as well as their careful orchestration in dyadic relationships (Arnold and Marinova 2023; Singh et al. 2017). Second, in relation to multiformat communication research, we validate the fit logic of media richness theory in a B2B selling context as the inverted U-shaped effect of hybrid selling (Daft and Lengel 1986; Moffett et al. 2021). Third, we extend sales channel research by addressing within-salesperson channel hybridization with the salesperson as the unit of orchestration (Shi et al. 2024; Sleep et al. 2020). Fourth, we outline the influences of contextual boundary conditions on hybrid selling effectiveness. Product complexity, customer channel preferences, and salesperson interaction knowledge shape the inverted U, offering meaningful indications on when hybrid selling is likely to enhance or undermine relationship performance.

Sales executives can use these findings to design effective hybrid selling strategies and directions for their salespeople to help them enhance their performance across customers with different interaction contexts. In our research context, customers who exhibit average interaction requirements perform best at about 36 percent share of remote calls. If the relationship involves low product complexity, strong remote preferences, and salesperson interaction knowledge, a 62 percent remote share leads to 27 percent better sales performance than pure face-to-face selling. In contrast, for more demanding relationships with high complexity, low remote preferences, limited interaction knowledge a purely remote approach can reduce sales performance by up to 32 percent.

Literature Review

In B2B customer relationships, sales calls involve interactions in which salespeople seek to identify customer needs, communicate value, and maintain or develop the customer relationship (Ahearne et al. 2022; Homburg et al. 2011a, 2011b). Conducting such sales calls requires adaptive relational communication, to align interaction modes with situational requirements in B2B selling, which can be enabled by synchronous formats such as face-to-face or remote formats (Mangus et al. 2024; Moffett et al. 2021). Face-to-face sales calls are in-person visits that often take place at the customer’s site and offer the richest and “most effective communication vehicle” (Spiro and Weitz 1990, p. 61). Such interactions provide proximal, visual, and verbal cues (Moffet et al. 2021), which enable salespeople to gain in-depth knowledge, foster interpersonal trust, and reduce uncertainty during the buying process (Ahearne et al. 2022; Bendapudi and Leone 2002). In contrast, remote sales calls occur at a distance, using synchronous modes such as video or phone (Sleep et al. 2020). They provide efficiency and flexibility, but they typically offer less richness than face-to-face interactions (Moffett et al. 2021). When used in combination within the same customer-salesperson relationship, face-to-face and remote calls imply hybrid selling. Considering the lack of research on hybrid selling effectiveness, we draw on three related streams of literature to theorize its mechanisms and performance outcomes, pertaining to (a) organizational frontlines, (b) multiformat communication, and (c) sales channels. We further position our research in relation to prior contributions in Table 1, and conceptually define and differentiate hybrid selling from related concepts in Appendix A.

Table 1.

Literature Review on Hybrid Selling in B2B Markets.

Authors	Research Setting	Research Strategy	Sales Performance Outcomes		Data	Field Salesperson Interaction Modes
Authors	Research Setting	Research Strategy	Relational	Financial	Data	Face-to-face	Remote	Optimal Mix	Channel hybridization	Contextual Variables
Venkatesan and Kumar (2004)	B2B, manufacturer of computer hardware and software	Conceptual + empirical	—	Customer lifetime value	CRM data: 48 months of 1,316 customers, 873 observations	✓	—	—	—	None
Vickery et al. (2004)	B2B, logistic services companies	Conceptual + empirical	• Trust • Satisfaction		Survey data: 69 logistic firms, 129 matched customers	✓	—	—	—	None
Reinartz et al. (2005)	B2B, high-technology manufacturer	Conceptual + empirical	—	• Profit• Acquisition likelihood	CRM data: 4 years of 12,024 prospects	✓	—	—	—	None
Bharadwaj and Shipley (2020)	B2B, conceptual study of communication effectiveness in digital sales interactions	Conceptual	—	—	None	✓	✓	—	Within same field salesperson	None
Moffett et al. (2021)	B2B, conceptual study of a theory of multiformat communication	Conceptual	Seller’s relational service	Seller’s financial performance	None	✓	✓	—	Within same field salesperson	• Message (e.g., involvement)• Temporal (e.g., relationship stage)• Dyadic (e.g., AI agent)
Ahearne et al. (2022)	B2B, conceptual study of shifts in buyer-seller interactions	Conceptual	—	—	None	✓	✓	—	Within same field salesperson	None
Shi et al., (2024)	B2B, distributor of lighting, sensors, switches, and relays	Conceptual +empirical	—	Customer sales	CRM data: 6 years, customer-level transaction data with 1,000+ inside outside sales rep dyads	✓	—	—	Across inside and field salesperson	• Product knowledge diversity• Customer relationship tenure• Customer product need complexity
Ahearne et al. (2025)	B2B and B2C, cross-industry	Conceptual + empirical	—	Firm performance (ROA)	Firm-level panel data: 196 firms with 85,000+ salespeople	✓	—	—	Across inside and field salesperson	• Product complexity• Span of control• Competitive intensity
Our study	B2B, manufacturer of sanitary products	Conceptual + empirical	Relationship quality	Share of wallet	• Study 1: CRM data, 18 months, 3,565 customers, • Study 2: Dyadic matched CRM and customer survey data, 18 months, 651 customers	✓	✓	✓	Within same field salesperson	Interaction requirements: • Product complexity• Customer channel preference• Salesperson interaction knowledge

Note. CRM = customer relationship management. Bold = Key attributes of our study.

Organizational Frontlines Research

Relevant Findings

Research on organizational frontlines considers interfaces where frontline employees, customers, and technologies meet to (co-) create value (Singh et al. 2017). Frontline employees’ behaviors in managing customer relationships, such as their exhibitions of customer orientation and empathy, drive their performance (Schepers and Van der Borgh 2020). Technologies like AI and remote interaction modes have rapidly transformed organizational frontlines and especially how frontline employees interact with customers (Huang and Rust 2018; van Doorn et al. 2017).

Key Takeaways

In line with existing findings that technologies reshape dynamics at the organizational frontline, hybrid selling that mixes face-to-face with remote interactions could affect value creation within the interface. Specifically, frontline salespeople likely need to carefully balance both modes, so as to create value and avoid harm to their relationships (Arnold and Marinova 2023; Singh et al. 2017; Singh and Bridge 2023).

Research Gaps

Organizational frontlines research mostly focuses on consumer and service contexts. To the best of our knowledge, research on frontline salespeople in B2B settings that combine interaction formats with the same customer over time, as well as studies to investigate potential sales performance trade-offs, remains scarce. To address this gap, we identify hybrid selling as a growing organizational frontline phenomenon and test contextual boundary conditions.

Multiformat Communication Research

Relevant Findings

Media richness theory (Daft and Lengel 1986) stresses that communication effectiveness (i.e., whether information can enhance mutual understanding within a certain time interval) depends on the level of fit between the richness of the communication format and the complexity or ambiguity of the focal task. Synchronous remote interactions can provide rich cues that previously might have been limited to face-to-face interactions (Bharadwaj and Shipley 2020; Moffett et al. 2021). Mangus et al. (2024) offer a comparison of communication format effectiveness (i.e., face-to-face, video, phone, email) after B2B salesperson transgressions. They show that synchronous formats (face-to-face, video, phone) tend to outperform asynchronous ones (e.g., email) in contexts of relational transgressions. For customers, face-to-face and video interactions may be equally effective in some situations (e.g., improving word of mouth after relational transgressions), but in others, a face-to-face approach might be more effective (e.g., regaining relationship satisfaction after relational transgressions).

Key Takeaways

A key insight provided by this research stream is the recognition that the relative effectiveness of face-to-face versus remote depends on the context as there is no universally better format (Mangus et al. 2024). Communication effectiveness depends on the fit between communication format richness and task complexity (Moffet et al. 2021). In turn, we anticipate that the effectiveness of hybrid selling also reflects a fit logic. Salespeople might alternate between face-to-face and remote calls with the same customer to achieve efficiency in routine tasks (Sleep et al. 2020) but richness in complex tasks (Mangus et al. 2024), contingent on context-specific interaction requirements.

Research Gaps

Among these indications of the potential benefits of remote sales calls, research lacks field-based evidence on how the combined use of multiple synchronous formats might create value in B2B exchange relationships. In particular, we know of no study that addresses the possibility that hybrid selling effects are nonlinear (e.g., inverted U-shaped) and depend on the interaction context, as might be defined by product complexity, customer channel preferences, or salesperson interaction knowledge.

Sales Channel Research

Relevant Findings

B2B firms combine field sales, inside sales, and online sales to serve customers (Shi et al. 2024; Sleep et al. 2020; Thaichon et al. 2018). While field sales enable rich and trust-building interactions at higher selling costs, inside sales provide efficiency and lower selling costs but offer less interaction depth (Ahearne et al. 2025; Ramos et al. 2023; Sleep et al. 2020).

Key Takeaways

Inside–outside sales rep dyads can increase value creation with the same customer when combined within hybrid sales structures (Shi et al. 2024). Inside salespeople can handle remote tasks (e.g., qualification, follow-ups; Sleep et al. 2020), while outside sales reps conduct rich face-to-face interactions (Shi et al. 2024). Hybrid selling differs from such multichannel approaches, though, because the same salesperson manages both face-to-face and remote interactions with the same customer. Therefore, it might support cost and efficiency gains while also preserving relational trust and knowledge.

Research Gaps

Prior research into combinations of different channels (e.g., field and inside sales) mostly overlooks how an individual salesperson might adopt multiple interaction modes. A hybrid sales structure features channel integration at the firm level, which is distinct from hybrid selling by a single salesperson who mixes interaction modes with the same customer. We propose analyzing hybrid selling as channel hybridization within a customer-salesperson relationship to extend sales channel research. In summary, prior research offers relevant insights but little information about the performance effects when a salesperson combines remote and face-to-face interaction modes within a customer relationship, and the contingencies that shape hybrid selling effectiveness.

Conceptual Framework and Hypotheses

Building on media richness theory (Daft and Lengel 1986), we argue that the performance effects of hybrid selling depend on the fit between the interaction mode and the interaction requirements (Figure 1). Thus, salespeople may balance the mix of remote and face-to-face calls they use by accounting for contextual requirements. When balance is lacking, due to an overreliance on either face-to-face (low efficiency) or remote (low effectiveness) calls, relationship outcomes likely decline (Moffet et al. 2021). To capture and examine both relational and financial outcomes of hybrid selling in B2B relationships, we focus on two distinct dimensions: sales performance, which reflects financial outcomes (e.g., share of wallet), and relationship quality, which reflects the customer’s overall evaluation of the exchange relationship (Palmatier et al. 2007; Sheth et al. 2000). Proximal cues delivered by face-to-face interactions foster effective and experiential communication goals and convey interpersonal and sensory involvement (Moffett et al. 2021; Short et al. 1976; Walther 1996). They might enhance overall exchange evaluations (Hornik 1992). In particular, such rich interaction modes should be most effective in interactions with high ambiguity and task complexity, whereas leaner modes can be superior for simple or unambiguous interaction tasks (Moffett et al. 2021).

Figure 1.

Research framework.

The extent to which interaction quantity benefits outweigh interaction quality losses is unlikely to be uniform across all customer relationships. In B2B exchanges, the interaction requirements vary with the complexity of the product being sold, the customer’s channel preferences, and the salesperson’s interaction knowledge. In turn, they alter how much interaction quality depends on rich versus leaner modes of communication (Mangus et al. 2024; Moffett et al. 2021). For example, selling complex products requires effective, trust-building communication (Ahearne et al. 2022), so increasing the share of remote calls too much might induce greater interaction quality losses (Ahearne et al. 2025). But if customers prefer remote interactions or salespeople possess extensive interaction knowledge, they might be able to mitigate such interaction quality losses, and in turn, a greater share of remote calls might be effective. Figure 2 illustrates mechanisms of this predicted inverted U-shaped effect.

Figure 2.

Mechanisms of the inverted U-shaped effect.

Main Effects of Hybrid Selling on Relationship Performance

We expect that hybrid selling introduces two countervailing mechanisms. On the one hand, increasing the share of remote calls produces interaction quantity benefits, because remote calls save travel time and allow salespeople to increase the number of touchpoints with customers, potentially improving need identification, value communication, and opportunities for cross- and up-selling (Reinartz et al. 2005; Schmitz et al. 2014). These benefits may be subject to diminishing returns though, such that additional remote calls might add less value past a certain level (Reinartz et al. 2005). On the other hand, greater reliance on remote calls induces interaction quality losses, which refer to the extent to which a customer-salesperson interaction lacks efficient and effective communication and does not provide sufficient cues (e.g., proximal, visual, verbal) to address the complexity or ambiguity of the communication task (Moffet et al. 2021).

When the share of remote calls is high, it implies salespeople are relying more on remote interaction modes, even for complex customer interactions that require stronger trust-building and uncertainty reduction, neither of which is effectively enabled by remote communication formats (Ahearne et al. 2025). The resulting limitations might erode the salesperson’s ability to identify latent needs, convincingly communicate value, reduce customer uncertainty, and gain customers’ trust, implying negative effects for their sales performance (Ahearne et al. 2022; Zoltners et al. 2022). An inverted U-shape emerges from the net effect of these mechanisms, marked by diminishing positive returns from interaction quantity and amplifying negative effects from declining interaction quality (Brown et al. 2011; Haans et al. 2016; Xu et al. 2022).

Sales performance depends on buying decisions, and relationship quality instead reflects customers’ broader evaluation of whether their relationship with the selling firm is perceived as successful or exceeds their expectations (Mullins et al. 2014; Palmatier et al. 2007). We expect that customers’ evaluations of relationship quality depend on whether their interactions with salespeople have been both efficient (i.e., avoiding waste of effort and time) and effective (i.e., solutions to their needs and problems; Moffet et al. 2021).

At moderate levels of remote share, hybrid selling might improve relationship quality by enabling more efficient and flexible interactions that signal the salesperson’s responsiveness and effort (Godfrey et al. 2011). However, we predict that these benefits are subject to diminishing returns too. Beyond some threshold, greater reliance on remote calls might amplify the losses in interaction quality (Ahearne et al. 2025). For customers engaged in demanding B2B buying tasks, such losses can hinder trust-building and effective problem solving, which are crucial to their evaluations of collaboration success (Ahearne et al. 2022; Marinova, Singh and Singh 2018). An overuse of remote calls thus may prevent salespeople from meeting customers’ expectations of effective collaboration and reduce their satisfaction with the relationship. Formally,

H₁: The share of remote calls has an inverted U-shaped effect on sales performance.

H₂: The share of remote calls has an inverted U-shaped effect on relationship quality.

Moderating Effects

We predict that the relationship outcomes of hybrid selling depend on the requirements of different customer-salesperson exchanges, because these requirements shape the extent to which effective, efficient communication depends on richer versus leaner interaction modes (Moffett et al. 2021). We consider three determinants of interaction requirements: (a) product complexity, (b) customer channel preference, and (c) salesperson interaction knowledge.

Product Complexity

When customers buy complex products, customer-salesperson interactions involve high ambiguity and more demanding problem solving (Macdonald et al. 2016; Webster and Wind 1972). From customers’ perspective, complex purchases entail greater risks and uncertainty, so they rely more on the salesperson for guidance (Dotzel and Shankar 2019; Patterson et al. 1997). From the salesperson’s perspective, selling complex products demands sophisticated need identification and effective value communication to foster trust and realize cross- and up-selling potential (Ahearne et al. 2022). These product-related requirements increase the need for rich communication formats that provide multiple, dynamic cues and enable rapid development of mutual understanding and trust (Daft and Lengel 1986; Moffett et al. 2021).

Accordingly, high product complexity should accelerate the interaction quality losses that occur with increasing reliance on remote calls (Ahearne et al. 2025). When customers are purchasing complex products, the limited information richness induced by remote calls undermines collaboration, causing relationship quality to deteriorate. For salespeople, similar constraints undermine their trust-building and hinder both their identification of latent needs and their communication of value propositions, which likely reduces sales performance (Ahearne et al. 2022). Thus, we predict that high product complexity shifts the inverted U-shaped effects of the share of remote calls to the left, for both relationship quality and sales performance.

H₃: Product complexity moderates the effect of the share of remote calls, such that it shifts this inverted U-shaped effect on (a) sales performance and (b) perceived relationship quality to the left.

Customer Channel Preferences

Communication channel preferences vary across customers, and those preferences influence their responses to different communication modes (Godfrey et al. 2011). Some customers perceive costly modes (e.g., face-to-face calls) as a proxy for relationship investments, but others regard such resource-intensive modes as inefficient expenses that signal the risk of higher prices (Palmatier 2008). We propose that customers might appreciate it when the share of remote calls matches their own preferences for remote interactions. Such matching can signal a type of personalization and a meaningful relationship investment, both of which can evoke reciprocity (De Wulf et al. 2001). As Fitzsimons and Lehmann (2004) show, matching channel preferences also enhances customers’ overall satisfaction with the buying process. Customers with a strong preference for remote interactions also could have more experience and skills with remote collaboration, enabling them to gain more utility from these calls (Mullins and Agnihotri 2022), due to their ability to participate actively in developing effective and efficient solutions. Thus, matching their preferences should enhance perceived interaction quality, as well as the salesperson’s need identification and value communication efforts, as customers might positively respond to relational investments aligned with their preferences (Homburg et al. 2011a, 2011b). Formally,

H₄: Customer channel preference moderates the effect of the share of remote calls, such that it shifts the inverted U-shaped effect on (a) sales performance and (b) perceived relationship quality to the right.

Salesperson Interaction Knowledge

Salespeople develop interaction knowledge by tracking details about customers’ interaction goals, product interests, and the outcomes of prior interactions (Joshi 2009). In our research context, we can gauge it according to the amount of customer-specific information recorded by the salesperson in the firm’s CRM system and gathered across multiple interactions. When salespeople gain a deeper understanding of customers’ needs or preferences, and prior exchanges, they can leverage those insights in subsequent interactions to tailor their communication and align solutions more effectively (Marinova et al. 2018). In B2B exchange relationships, salespeople also help close knowledge gaps between the selling firm and customers (Zoltners et al. 2022). Their interaction knowledge can benefit all their interactions with customers, but it may be particularly critical in remote calls. Compared with face-to-face interactions, remote calls provide fewer proximal and dynamic cues, which can constrain salespeople’s ability to identify latent needs or build mutual understanding and ultimately trust (Ahearne et al. 2022; Moffett et al. 2021). In such settings, access to tracked knowledge enables salespeople to compensate for reduced cue richness by referencing prior discussions accurately and communicating value propositions with greater precision (Bharadwaj and Shipley 2020; Boyer and Jap 2022). Thus, we predict that interaction knowledge mitigates losses in interaction quality that typically arise at higher shares of remote calls. For salespeople in possession of such knowledge, remote calls might remain effective, even at higher shares. In contrast, when salespeople’s interaction knowledge is low, remote calls might be more likely to reduce interaction quality. Formally,

H₅: Salesperson interaction knowledge moderates the effect of the share of remote calls, such that it shifts the inverted U-shaped effect on (a) sales performance and (b) perceived relationship quality to the right.

Study 1: Hybrid Selling and Sales Performance

Research Context and Sample Structure

To test our main effect hypotheses, we needed evidence related to how B2B salespeople use hybrid selling and combine face-to-face and remote sales calls during interactions with their existing customers. We received support and collected large-scale data from a European B2B manufacturer of sanitary products that generates revenues of approximately $3.8 billion annually with 11,000 employees. It sells its products and systems to B2B customers (e.g., plumbers, planners, architects, industrial customers, general contractors), primarily through a field sales force and a “one face to the customer” philosophy. That is, each salesperson is responsible for her or his customers and promotes all products offered by the firm, with the support of specialists like product managers, engineers, or the marketing department. To manage their customer relationships and achieve professional sales interactions, salespeople have the autonomy to choose between face-to-face and remote sales calls (video or phone) for each interaction with their customers. They are required to track each interaction in a visit report in the CRM system that details information about the customer and the specific interaction, including its type (remote or face-to-face), the specific selling situation (e.g., first contact, product presentation, negotiation), and interaction-specific knowledge (e.g., interaction goals, product interests, outcomes). The firm puts strong emphasis on CRM data quality and employs CRM specialists to ensure that interactions are logged.

We gathered CRM records pertaining to 3,565 existing B2B customers, managed by twenty-seven field sales representatives across two sales regions, over an eighteen-month period. This data set comprises 15,172 sales interactions (remote and face-to-face). All the salespeople in this sample were men. On average, each salesperson managed 132.04 customers (SD = 37.56), and 33.63 percent of their sales calls were remote (SD = 32.86% at the customer level; SD = 7.93% at the salesperson level). Across the observation period, salespeople interacted with a customer on average 4.26 times. Customer firms in the sample employed an average of 9.13 employees (SD = 25.71). Regarding industry distribution, most customers were plumbers (78%), followed by planners (11.4%), industrial customers (4%), architects (2%), general contractors (2%), and other segments (3%).

Measures

To measure the share of remote calls (SRC), we take the ratio of remote to total sales calls (face-to-face + remote) for each customer during the eighteen-month period. In addition, we measure the share of wallet, which represents a key relationship performance metric for B2B customer retention (Sheth et al. 2000; Zeithaml 2000). It indicates the share of business a customer conducts with a particular selling firm (Keiningham et al. 2003). We received a share of wallet (SOW) measure that the selling firm deploys internally. It captures the share of wallet for each individual customer across all nine product lines of the selling firm. We also account for some control variables. First, we noted the total calls, or the sum of face-to-face and remote sales calls, to each customer in the eighteen-month period. Second, to control for the individual salesperson’s potential impact, we added salesperson fixed effects. Third, in recognition of the nested data structure, such that customers in our data are nested within the twenty-seven salespeople, we obtained cluster-correlated robust estimates of variance (Rogers 1994). Fourth, to control industry-specific impacts, we added industry fixed effects. Table 2 contains the descriptive statistics and correlations.

Table 2.

Descriptive Statistics and Correlations, Study 1.

Variables	1	2	3	4
1: SOW
2: SRC	.10**
3: SRC²	.09**	.55**
4: Total calls	.19**	.14**	.31**
Mean	.35	.34	.11	4.26
SD	.18	.33	.12	4.62
Min	.01	0	0	1
Max	1	1	.44	72

Note. Dummy variables are excluded. SOW = share of wallet, SRC = share of remote calls.

p < .05. **p < .01. Two-tailed tests of significance.

Model Estimation and Results

A regression analysis, conducted in STATA SE/17, indicates support for H₁, because the effect of the share of remote calls on sales performance exhibits an inverted U-shape (Model 1: b_SRC→SOW = .045, p < .01; b_SRC²_→SOW = −.169, p < .01; see Table 3). We specified both linear and quadratic terms for the share of remote calls and integrated both into our equation. As suggested by Haans et al. (2016), we tested the robustness of the quadratic effect by adding the effect of the cubic term. The quadratic effect remained significant.

Table 3.

Study 1 Results.

Models		Model 1:	Model 2:
Dependent Variable		SOW	SOW
Independent Variables	Hypothesis	Estimates (SE)	Estimates (SE)
Main effect
SRC		.045*** (.009)	−.148^n.s. (.108)
SRC²	H₁	−.169***(.026)	−.169***(.026)
Control variables
Total calls		.006*** (.001)	.006*** (.001)
Salesperson ID and industry (dummy)		Included	Included
Endogeneity correction
Gaussian copula for SRC			.039* (.022)
R ²		.532	.532

Note. We report unstandardized coefficients (robust standard errors in brackets are clustered on individual salespeople). SOW = share of wallet, SRC = share of remote calls.

p < .10.**p < .05. ***p < .01. n.s. = not significant (p > .10). Two-tailed tests of significance.

Noting the potential for endogeneity, we also applied Park and Gupta’s (2012) recommendation and a Gaussian copula approach. The two key assumptions required for a copula method hold as the distribution of the endogenous regressor (a) contains rich information and (b) differs from the distribution of the structural error, which is assumed to be normal. A Kolmogorov–Smirnov test (p < .001), visual inspection of the distribution of the share of remote calls, and Shapiro-Wilk tests (share of remote calls: .989, p < .001) all indicate a non-normal distribution. Therefore, we included the copula in our model and reran the analysis in Model 2. The results indicate the robustness of the inverted U-shaped effect (b_SRC→SOW = .148, n.s.; b_SRC²_→SOW = −.169, p < .01). We discuss some further potential endogeneity issues and remedies in Study 2.

The results of Study 1 thus provide first empirical evidence of an inverted U-shaped effect of hybrid selling on sales performance. However, this analysis is limited to CRM data, such that it captures the selling firm’s perspective on performance outcomes. To complement these findings and establish additional insights into relational outcomes, we account for customers’ perceptions of relationship quality and potential moderation by product-, customer-, and salesperson-related interaction requirements in Study 2. Therefore, we combined the CRM records with customer survey data. Customer inputs can offer insights into hybrid selling effectiveness and reflect customers’ perceptions of the efficiency and effectiveness of different interaction modes (Mangus et al. 2024), beyond the salesperson’s efforts and performance effects captured by the CRM data. As we have noted, some customers might prefer remote sales calls for flexibility and efficiency reasons while others might value face-to-face interactions as relational assets. These differences are reflected in our moderation hypotheses and in our conceptual model, which explicitly incorporates customers’ evaluations of the exchange relationship (i.e., perceived relationship quality).

Study 2: Contingent Effects of Hybrid Selling

Research Context and Sample Structure

Study 1 findings provide initial evidence of the sales performance effects of hybrid selling. To identify its effects on both relationship quality and sales performance as well as the contingencies of these links, we complement the objective variables that informed Study 1, with input shared by another set of B2B customers. That is, to develop a deeper understanding of hybrid selling from both the selling firm’s and the customer’s perspective, we combined responses to a customer survey with matched CRM data. We initially contacted 5,066 representatives of 2,712 customer firms, from whom we received 1,358 responses. To obtain representative responses, we selected one key informant per customer firm, specifically, the one with the longest general buying experience. For the resulting 993 key informants, we obtained eighteen months of objective CRM records, using the variables that we included in Study 1. Then we matched the objective data with the customer survey responses. We identified 801 respondents by limiting the sample to customers for which at least one valid interaction had been tracked in the eighteen-month window. We excluded forty-eight customer firms with no share of wallet recorded in the CRM system, as well as 102 customers who did not provide some relevant survey responses (e.g., relationship quality, product complexity, channel preference). Thus, the final sample for Study 2 involves 651 customers matched with 5,971 observations of CRM data, again nested within the twenty-seven male field salespeople working in two sales regions. In this sample, each salesperson manages 24.11 customers on average (SD = 14.15), with an average remote call share of 41.47 percent (SD = 27.00% at the customer level; SD = 7.72% at the sales-rep level) and average total interactions per customer of 9.17 within the eighteen-month observation period. Customers were 87 percent men and 13 percent women, with an average age of 47.33 years (SD = 11.41) and average buying experience of 19.51 years (SD = 11.08). Customer firms employ 15.12 people on average (SD = 26.64). The industry composition of these 651 customers is 47 percent plumbers, 38 percent planners, 5 percent architects, 4 percent industrial customers, 2 percent general contractors, and 5 percent other segments.

Measures

For the customer survey, we used well-established scales and adapted as necessary. Relationship quality was measured with a three-item, seven-point rating scale from Jap (2001) (e.g., “Our relationship was successful so far”). To measure product complexity, we used a single item, seven-point rating scale referring to the degree to which a customer perceives the selling firm’s products as complex (“I felt this kind of product was complex in nature”; Heitmann et al. 2007). We measured customer channel preference as the degree to which a customer prefers remote interactions when collaborating with the selling firm (“In general, to what extent do you prefer to use the following communication channels in your collaboration with the selling company?”), gauged on a seven-point scale ranging from “no preference at all” to “very high preference,” as also used by Godfrey et al. (2011). The interaction knowledge measure reflects the average amount of information (i.e., detailed interaction goals, product interests, and outcomes) that the salesperson adds to the CRM system after each interaction. The scales indicated strong psychometric properties, in support of the reliability and validity of the measurements. Furthermore, we added controls for total calls, customer firm size, relationship length, customer buying experience, and customer age, as well as dummy variables for the buying department, product complexity variance, and salesperson and industry fixed effects. Table 4 lists the variable descriptions, representative studies, and their operationalizations for both Studies 1 and 2. Table 5 contains the descriptive statistics and variable correlations.

Table 4.

Variable Descriptions and Operationalizations.

Variables	Description and Conceptual Meaning	Representative Studies	Operationalization in this Study
Relationship performance (dependent variables)
Sales performance (share of wallet)	Salesperson’s sales performance, measured as share of wallet on customer level.	Sheth et al. (2000); Zeithaml (2000)	Average customer’s share of wallet of the selling firm’s product.
Relationship quality	Customer’s overall evaluation of the relationship with the selling firm.	Homburg et al. (2011a, 2011b); Jap (2001)	Three-item, seven-point rating scale:(1) “Our relationship was successful so far.”(2) “Our relationship exceeded my expectations.”(3) “I am satisfied with the results of the relationship.”
Hybrid selling (independent variable)
Share of remote calls	Salesperson’s hybrid selling mix, reflected by the share of remote to total calls on customer level.		Share of remote to total sales calls (remote + face-to-face) with a customer in the 18-month period.
Contingency factors (moderators)
(1) Product complexity	Degree to which customer perceives the selling firm’s products as complex.	Burnham et al. (2003); Friess et al. (2024); Heitmann et al. (2007)	One-item, seven-point rating scale:(1) “I felt this kind of product was complex in nature.”
(2) Customer channel preference	Degree to which customer prefers to conduct remote interactions with the selling firm.	Godfrey et al. (2011)	Two-item, seven-point rating scale: “In general, to what extent do you prefer to use the following communication channels in your collaboration with the selling company?”(1) “Videoconferences”(2) “Phone”
(3)Salesperson interaction knowledge	Salesperson’s tracked details about the interaction goals, product interests, and interaction outcomes.	Joshi (2009)	Log-transformed average amount of information salesperson tracks about the customer in the CRM system per interaction in the 18-month period.
Control variables
Total calls	Salesperson’s total calls with the customer.		Sum of total sales calls with a customer in the 18-month period.
Customer firm size	Customer’s firm size.		Number of employees of the customer’s firm.
Relationship length	Customer-salesperson relationship length.		Log-transformed years of relationship length.
Customer buying experience	Customer’s experience in organizational buying tasks.		Log-transformed years of buying experience.
Customer age	Customer’s age.		Log-transformed years of age.
Salesperson ID	Salesperson ID.		Dummy variable for salesperson.
Customer industry	Industry in which the customer operates.		Dummy variable for industry.
Buying department	Formalization of buying processes in the customer’s firm.		Dummy variable for buying department in customer firm.
Product complexity variance	Customer’s perceived complexity variance of the selling firm’s products.		Dummy variable for product complexity variance.

Table 5.

Descriptive Statistics and Correlations, Study 2.

Variables	1	2	3	4	5	6	7	8	9	10	11	12
1: SOW
2: RQU	.17**
3: SRC	.05	.03
4: SRC²	−.12**	−.13**	.14**
5: CPX	.01	−.05	−.03	.02
6: CCP	.04	.14**	.09*	.02	.02
7: SIK	−.31**	−.11**	−.10*	.17**	.00	−.03
8: Total calls	.14**	.14**	.20**	−.30**	−.06	.05	−.21**
9: Customer firm size	.04*	.02	.08*	−.09*	−.03	−.02	.00	.47**
10: Relationship length	.11**	.12**	−.04	−.08*	−.06	−.06	−.01	.11**	.07
11: Customer buying experience	.00	.03	−.06	.03	−.04	−.03	−.02	−.01	.00	.73**
12: Customer age	−.03	.02	−.10**	.05	−.06	−.94	.03	−.02	.00	.66**	.84**
Mean	.41	6.00	.42	.07	2.64	5.19	5.99	9.21	15.12	2.82	2.83	3.85
SD	.18	.98	.26	.09	1.42	1.32	.71	8.54	26.64	.70	.69	.253
Min	0	2	0	0	1	1	3.42	1	1	.70	.69	3.09
Max	.82	7	1	.35	7	7	8.37	69	300	4.11	4.11	4.45

Note. Dummy variables excluded. SOW = share of wallet, RQU = relationship quality, SRC = share of remote calls, CPX = product complexity, CCP = customer channel preference, SIK = salesperson interaction knowledge.

p < .05. **p < .01.

Model Estimation and Results

We specify the models and test the hypotheses (Table 6) using structural equation modeling in STATA SE/17. We integrated the linear and quadratic terms of the share of remote calls, as well as all lower-order interactions of the moderators with the share of remote calls (and its quadratic term). Models 1 and 2 include the main effects. We integrated all moderators and interactions into the full Model 3. In line with H₁ and H₂, we identify inverted U-shaped effects of the share of remote calls on relationship quality (RQU) (b_SRC→RQU = −.042, n.s.; b_SRC²_→RQU = −1.523, p < .01) and sales performance (b_SRC→SOW = −.013, n.s.; b_SRC²_→SOW = −.087, p < .05). Share of wallet provides the primary dependent variable, reflecting its relevance as a sales performance metric in customer relationships (Sheth et al. 2000; Zeithaml 2000). Customers’ perceived relationship quality provides the qualitative metric of relationship performance, reflecting its capacity to influence actual purchase decisions (Lemon and Verhoef 2016; Palmatier et al. 2007). Model 3 affirms a significant positive effect of relationship quality on sales performance too (b_RQU→SOW = .021, p < .01). That is, the share of remote calls affects sales performance and relationship quality, and relationship quality also determines some sales performance effects. In an additional mediation analysis of the effect of the share of remote calls on sales performance through relationship quality (Model 3), we find a significant indirect effect (b_{ind_SRC}²_→RQU→SOW = −.032, p < .05). The indirect effect explains 26.9 percent of the total effect (b_{total_SRC}²_→SOW = −.119, p < .05), indicating partial mediation. These results suggest the need for simultaneous analyses of relationship quality and sales performance.

Table 6.

Model Results, Study 2.

		Main Effects Model	Full Model
		Model 1:	Model 2:		Model 3:		Model 4:
Dependent Variables		SOW	SOW	RQU	SOW	RQU	SOW	RQU
Independent Variables	Hypothesis	Estimates (SE)	Estimates (SE)	Estimates (SE)	Estimates (SE)	Estimates (SE)	Estimates (SE)	Estimates (SE)
Main effects
SRC		−.005^n.s. (.021)	−.005^n.s. (.021)	−.006^n.s. (.153)	−.013^n.s. (.018)	−.042^n.s. (.130)	−.159^n.s. (.161)	−.124^n.s. (1.28)
SRC²	H₁/H₂	−.111** (.049)	−.083* (.046)	−1.33*** (.481)	−.087** (.040)	−1.523*** (.561)	−.082** (.040)	−1.490*** (.561)
RQU			.022*** (.005)		.021*** (.047)		.021*** (.047)
Main effects moderators
CPX					.004^n.s. (.003)	−.020^n.s. (.029)	.004^n.s. (.003)	−.020^n.s. (.029)
CCP					−.002^n.s. (.014)	.084*** (.024)	−.002^n.s. (.003)	.08*** (.024)
SIK					.002^n.s. (.014)	−.032^n.s. (.074)	.001^n.s. (.014)	−.040^n.s. (.076)
Moderation effects
SRC × CPX	H_3a/H_3b				−.023** (.011)	−.326*** (.099)	−.023** (.012)	−.326*** (.099)
SRC² × CPX					.020^n.s. (.032)	−.146^n.s. (.289)	.020^n.s. (.032)	−.142^n.s. (.288)
SRC × CCP	H_4a/H_4b				.009^n.s. (.012)	.310** (.141)	.009^n.s. (.011)	.313** (.143)
SRC² × CCP					−.012^n.s. (.030)	.820** (.411)	−.001^n.s. (.030)	.837** (.410)
SRC × SIK	H_5a/H_5b				.055*** (.014)	.035^n.s. (.155)	.058*** (.015)	.062^n.s. (.164)
SRC² × SIK					−.009^n.s. (.053)	.464^n.s. (.553)	−.009^n.s. (.052)	.465^n.s. (.560)
Control variables
Total calls		.003*** (.001)	.003*** (0.001)	.011^n.s. (.006)	.003*** (0.001)	.009^n.s. (.006)	.003*** (0.001)	.008^n.s. (.006)
Customer firm size		.001^n.s. (.004)	−.002^n.s. (.004)	−.026^n.s. (.021)	.002^n.s. (.004)	−.031^n.s. (.023)	.002^n.s. (.004)	−.030^n.s. (.023)
Relationship length		.008^n.s. (.012)	.003^n.s. (.012)	.197*** (.060)	.003^n.s. (.011)	.207*** (.060)	.004^n.s. (.011)	.213*** (.063)
Customer buying experience		−.003^n.s. (.014)	−.0004^n.s. (.013)	−.121^n.s. (.133)	−.0003^n.s. (.013)	−.132^n.s. (.133)	−.0001^n.s. (.013)	−.134^n.s. (.133)
Customer age		.011^n.s. (.037)	.013^n.s. (.036)	−.079^n.s. (.288)	.001^n.s. (.037)	−.085^n.s. (.282)	.008^n.s. (.037)	−.099^n.s. (.282)
Product complexity variance (dummy)		Included	Included	Included	Included	Included	Included	Included
Buying department (dummy)		Included	Included	Included	Included	Included	Included	Included
Salesperson fixed effects		Included	Included	Included	Included	Included	Included	Included
Industry fixed effects		Included	Included	Included	Included	Included	Included	Included
Endogeneity correction
Gaussian copula for SRC							−.038^n.s. (.035)	−.286^n.s. (.284)
R²		.585	.597	.099	.604	.146	.605	.147
Overall R²			.629		.654		.655

Note. We report unstandardized coefficients (robust standard errors in brackets are clustered on salesperson level), SOW = share of wallet, RQU = relationship quality, SRC = share of remote calls, CPX = product complexity, CCP = customer channel preference, SIK = salesperson interaction knowledge.

p < .10.**p < .05. ***p < .01. n.s. = not significant (p > .10). Two-tailed tests of significance.

We also uncover moderating effects of product complexity (CPX) on the impact of the share of remote calls on both sales performance (b_{SRC×CPX→SOW} = −.023, p < .05; b_SRC²_×CPX→SOW = −.020, n.s.) and relationship quality (b_{SRC×CPX→RQU} = −.326, p < .01; b_SRC²_×CPX→RQU = −.146, n.s.). Product complexity shifts the inverted U-shaped effect of the share of remote calls on relationship quality and sales performance to the left, in support of H_3a and H_3b. The moderating effect of customer channel preference (CCP) on the effect of the share of remote calls on relationship quality (b_{SRC×CCP→RQU} = .310, p < .05, b_SRC²_×CCP→RQU = .820, p < .05) also allows us to confirm H_4b, but we find no significant moderation of the effect on sales performance, in contrast with H_4a. Customers with strong channel preferences seemingly perceive the salesperson’s greater customer orientation, which can indirectly increase sales performance (Homburg et al. 2011a, 2011b). In an additional mediation analysis, we find indications for this reasoning as customer channel preference moderates the sales performance effect indirectly, through relationship quality (b_{ind_SRC×CCP→RQU→SOW} = .006, p < .10, b_{ind_SRC}²_{×CCP→RQU→SOW} = .170, n.s.). Finally, we find support for H_5a in the moderating effect of the salesperson’s interaction knowledge (SIK) on the impact of the share of remote calls on sales performance (b_{SRC×SIK→SOW} = .055, p < .01; b_SRC²_×SIK→SOW = −.009, n.s.), such that interaction knowledge shifts the inverted U-shaped sales performance effect to the right. At very low levels of interaction knowledge, increasing the share of remote calls even might harm sales performance. However, we must reject H_5b, because we find no significant moderating effect of interaction knowledge on relationship quality. We present the simple slope analyses for these findings in Figure 3.

Figure 3.

Simple slope plots, Study 2.

Robustness Checks and Post Hoc Analyses

Robustness of the Inverted U-Shape

To assess the robustness of the inverted U-shaped relationship between the share of remote calls and the performance outcomes, we followed four steps recommended by Haans et al. (2016). First, we examined whether the curve is sufficiently steep at both ends. In Table 7, the estimated slopes at the minimum and maximum levels of the share of remote calls are significant, indicating meaningful performance changes across the full range. Second, we probed the simple slopes at several values for the share of remote calls (0%, 25%, 50%, 75%, and 100%) and across low and high moderator levels (mean ± SD). The results confirm the hypothesized curve shifts and parallel the patterns in the simple slope plots (Figure 3). Third, to test the functional form, we added cubic terms for the share of remote calls. These cubic effects were nonsignificant (b_SRC³_→RQU = −1.024, n.s.; b_SRC³_→SOW = .056, n.s.), while the quadratic effects remained robust and significant (b_SRC²_→RQU = −1.124, p < .05; b_SRC²_→SOW = −.096, p < .10.). Fourth, we calculated turning points and their 95% confidence intervals across contexts, to reflect the possibility that performance outcomes remain statistically similar across small variations in the share of remote calls around a turning point. These turning points indicate optimal hybrid selling mixes for our research context. However, it is important to note that they likely vary in size across different industrial contexts.

Table 7.

Probing the Inverted U: Slopes, Confidence Intervals, and Turning Points.

Share of Remote Calls (SRC)			Moderators
		Unconditional Main Effect	Product Complexity		Customer Channel Preference		Salesperson Interaction Knowledge
		Unconditional Main Effect	Low	High	Low	High	Low	High
Turning point(optimum)	Estimates	35.85%	47.95%	16.34%	29.35%	40.81%	18.14%	51.76%
Turning point(optimum)	CI	[.186; .531]	[.367; .593]	[−.400; .726]	[−.018; .606]	[.224; .592]	[−.118; .481]	[.372; .663]
Minimum(SRC = 0%)	Estimates	.085*	.140**	.030^ns	.060^ns	.110**	.041^ns	.129*
Minimum(SRC = 0%)	CI	[−.007; .177]	[.016; .265]	[−.110; .170]	[−.072; .193]	[.001; .219]	[−.060; .141]	[−.004; .263]
Low(SCR = 25%)	Estimates	.026^ns	.067**	−.016^ns	.009^ns	.043^ns	−.015^ns	.067*
Low(SCR = 25%)	CI	[−.028; .079]	[.001; .134]	[−.100; .068]	[−.064; .082]	[−.024; .109]	[−.072; .041]	[−.008; .141]
Medium(SCR = 50%)	Estimates	−.034*	−.006^ns	−.061**	−.042*	−.025^ns	−.071***	.004^n.s.
Medium(SCR = 50%)	CI	[−.069; .002]	[−.040; .028]	[−.115; −.007]	[−.087; .002]	[−.072; .023]	[−.113; −.030]	[−.032; .041]
High(SCR = 75%)	Estimates	−.093***	−.079**	−.106**	−.094**	−.092**	−.128***	−.058^ns
High(SCR = 75%)	CI	[−.154; −.032]	[−.155; −.003]	[−.190; −.022]	[−.180; −.007]	[−.164; −.020]	[−.204; −.051]	[−.130; .014]
Maximum(SCR = 100%)	Estimates	−.152***	−.152**	−.152**	−.145*	−.159**	−.184***	−.121*
Maximum(SCR = 100%)	CI	[−.253; −.051]	[−.288; −.017]	[−.292; −.012]	[−.293; .003]	[−.274; −.045]	[−.308; −.059]	[−.251; .010]

Note. We probe the simple slopes of the inverted U-shaped effect for the whole possible range of share of remote calls. The range spans minimum = 0%, low = 25%, medium = 50%, high = 75%, and maximum = 100% levels. We report 95% confidence intervals in italized brackets. CI = confidence interval; SRC = share of remote calls.

*p < .10.**p < .05. ***p <.01. n.s. = not significant (p > .10). Two-tailed tests of significance.

Relationship Length

Remote interactions may be less effective in early relationship stages than in later ones, because they are limited in building trust, as is essential in early stages (Moffett et al. 2021; Shi et al. 2024). Thus, relationship length might moderate the effect of the share of remote calls. However, our research context involves existing customers with long-standing relationships (M = 19.65 years; SD = 11.23) and high relationship quality (M = 6.01 on a seven-point scale; SD = .96), as is true in many B2B industries (e.g., machinery, chemicals, pharmaceuticals). The length of the customer relationships in our data is uncorrelated with the share of remote calls (r = −.04, n.s.; Table 6), so we expect the risk of confounding to be low. Still, we tested for potential moderation of relationship length (Appendix B). The results indicate no significant moderation of the relationship quality or sales performance effects. We expect that this finding reflects the nature of our sample, featuring mature, high-quality relationships, such that potential ceiling effects might arise. For our research context, this finding supports our inclusion of relationship length as a control variable rather than as a central contingency factor.

Nested Data Structure

The customer-level observations are nested within twenty-seven salespeople and multiple industries. In the main analysis, we include salesperson and industry fixed effects, then correct the standard errors for clustering at the salesperson level. As a robustness check, we reestimated Model 3 for both relationship performance dimensions, using mixed-effects specifications with random intercepts for salesperson and industry assignments (Appendix C). For both the main and interaction effects, the mixed-effects model yields similar results, suggesting the robustness of our findings.

Endogeneity

Underlying factors (e.g., relationship length, customer value) might relate systematically to the share of remote calls, which would cause the observed effects to reflect artifacts of these variables rather than the causal impact of this share (Becker et al. 2022). We consider this risk relatively slight though for two reasons. First, our sample of long-term customer relationships exhibits little variation in relationship stage, which minimizes its potential impact. Second, the share of remote calls is not correlated with relationship length or share of wallet (r = .05, n.s.). The inverted U-shaped effect also suggests that the highest performance levels occur at a moderate share of remote calls.

To strengthen the robustness of our findings, we again follow recommendations by Park and Gupta (2012). First, our rich data modeling approach includes multiple control variables. Second, we applied Gaussian copulas to address potential endogeneity by modeling the correlation between the regressor and error term. Both assumptions of the copula method are met (Kolmogorov–Smirnov p < .001; Shapiro-Wilk .995, p < .05). Including the copula term in Model 4 left our results substantively unchanged. Thus, we conclude that endogeneity is unlikely to bias our results.

Discussion

Implications for Theory and Research

To the best of our knowledge, this study provides the first empirical examination of hybrid selling in B2B contexts. It makes four main contributions to sales and service research (Kindermann et al. 2024; MacInnis 2011). First, we establish hybrid selling as an organizational frontline phenomenon by conceptualizing and empirically investigating its relational and financial outcomes. Organizational frontlines research that examines the interfaces where employees, customers, and technologies meet to co-create value (Singh et al. 2017) has shown that frontline employees’ behaviors drive performance (Schepers and Van der Borgh 2020) and that new technologies (e.g., AI, remote interaction formats) are transforming or disrupting such dynamics (Arnold and Marinova 2023; Huang and Rust 2018; van Doorn et al. 2017). As our results show, value creation at the frontline depends on not just the availability of multiple (technological and in-person) interaction formats but also their careful orchestration in dyadic relationships. We highlight frontline employees as integrators of human- and technology-based interfaces that need to carefully balance efficiency with effectiveness when selling hybrid.

Second, in relation to multiformat communication research, we empirically test the fit logic proposed by communication theories in a B2B selling context. Media richness theory suggests that effective communication depends on matching format richness with task complexity (Daft and Lengel 1986). Prior conceptual studies have predicted that salespeople should deploy formats flexibly, depending on relational contexts and interaction requirements (Ahearne et al. 2022; Moffett et al. 2021). Our findings demonstrate that the outcomes of combining remote and face-to-face calls follow an inverted U-shape based on diminishing positive effects of interaction quantity and amplifying losses in interaction quality. By detailing this conceptual mechanism, we extend communication research and clarify that relationship performance effects depend on balancing formats.

Third, we link hybrid selling to sales channel research by specifying the concept of within-salesperson channel hybridization. In hybrid sales structures or inside–outside sales rep dyads, salespeople in different roles or channels manage the same customer (Shi et al. 2024; Sleep et al. 2020). But when one salesperson manages both face-to-face and remote channels with the same customer, he or she balances their impact on relational trust and knowledge that stem from the interaction quantity–quality trade-offs.

Fourth, we explicate relevant contextual boundary conditions that shape the effectiveness of hybrid selling. Product complexity, customer channel preference, and salesperson interaction knowledge all can shift the inverted U-shaped curve. We identify specific conditions in which hybrid selling enhances or harms relationship performance based on the developed interaction quality-quantity tradeoff. Future research might build on these insights and examine other contingencies, such as interaction sequences, psychological mechanisms, or negotiation contexts (Ahearne et al. 2022). For example, on the one hand, higher interaction quantity with digital interaction modes might induce technostress along with increased workload (Tarafdar et al. 2014). On the other hand, hybrid selling might offer new possibilities for increasing sales performance with improved quota achievement. Salesperson-level investigations could shed light into when and how hybrid selling benefits or harms not only customer relationships but also the well-being and performance of frontline salespeople.

Implications for Practice

When designing a strategic hybrid selling approach, selling firms and their executives must balance the expected efficiency gains against customer demands, while also convincing their salespeople of the potential benefits. This study provides some actionable implications that should help sales managers develop targeted hybrid selling approaches and show salespeople how and when to optimize their interactions to enhance their performance.

We provide some illustrative guidelines for hybrid selling and their performance effects, derived from Study 2 (Table 8), that suggest how to set target mixes for several customer interaction contexts. These contexts reflect specific combinations of moderator values (±standard deviation, mean). For each scenario, we calculate a baseline scenario (100% face-to-face), predict the optimal share of face-to-face calls with confidence intervals, and identify the expected performance increase relative to the baseline scenario. For example, the optimal share of face-to-face calls for a customer with average interaction requirements is 64.1 percent. For a customer interaction context defined by low product complexity, strong remote channel preferences, and strong salesperson interaction knowledge, a share of face-to-face calls of 37.9 percent enhances sales performance by 27.3 percent in comparison with pure face-to-face selling. Customers marked by higher complexity, lower preferences, and weaker salesperson interaction knowledge demand a higher face-to-face share to enhance performance.

Table 8.

Managerial Guidelines for Hybrid Selling.

			Baseline Condition		Optimal Hybrid Selling Mixes and Performance Effects
Customer Interaction Contexts			Share of Face-to-Face Calls	Share of Wallet	Share of Face-to-Face Calls (Optimum)	Share of Wallet (Maximum)	Share of Wallet Increase (Relative to Baseline Condition)
1	Product complexity	Mean	100%	28.6%	64.1%	30.01%	+5.3%
	Customer channel preference	Mean
	Salesperson interaction knowledge	Mean			[.469; .814]
2	Product complexity	Low (−SD)	100%	26.2%	52.1%	29.6%	+12.9%
	Customer channel preference	Mean
	Salesperson interaction knowledge	Mean			[.407; .634]
3	Product complexity	Low (−SD)	100%	25.4%	49.1%	29.6%	+16.5%
	Customer channel preference	High (+SD)
	Salesperson interaction knowledge	Mean			[.348; .635]
4	Product complexity	Low (−SD)	100%	23.8%	37.9%	30.3%	+27.3%
	Customer channel preference	High (+SD)
	Salesperson interaction knowledge	High (+SD)			[.172; .587]
5	Product complexity	High (+SD)	100%	30.9%	83.7%	31.2%	+0.8%
	Customer channel preference	Mean
	Salesperson interaction knowledge	Mean			[.274; 1.400]
6	Product complexity	High (+SD)	100%	31.7%	96.7%	31.7%	+0.03%
	Customer channel preference	Low (−SD)
	Salesperson interaction knowledge	Mean			[−.223; 2.158]
7	Product complexity	High (+SD)	100%	33.3%	100%	33.3%	+0.0%
	Customer channel preference	Low (−SD)
	Salesperson interaction knowledge	Low (−SD)			[−.1.032; 3.609]

Note. Low/high value of moderator reflects ±1 standard deviation (SD) from the mean value. The contexts refer to conditions in which the combinations of all mentioned moderator conditions are simultaneously true. To facilitate a comparison between pure face-to-face selling and hybrid selling in the context of our specific company case, we present the share of face-to-face calls in place of the share of remote calls (Studies 1 and 2). The turning points are reported with confidence intervals. We report 95% confidence intervals in italized brackets.

Remote calls also free up salespeople’s capacity, by reducing travel time and average call duration (Schmitz et al. 2021), which might additionally enhance their interaction quantity (note that Table 6 indicates a statistically significant and positive correlation between the share of remote calls and total calls, with r = .20, p < .01). A hybrid salesperson managing an average customer portfolio (scenario 1 with optimal face-to-face share of 64.1%) could increase their total calls by up to 365 per year, compared with the 300 calls that a purely face-to-face interacting salesperson could conduct. This results in more than 20 percent increase in interaction quantity, which likely supports salespeople’s relationship management activities. However, for customers that are characterized by high product complexity, low channel preference, and limited salesperson interaction knowledge (e.g., scenario 7), relying solely on remote calls risks strong performance declines of up to 32.4 percent. Thus, our findings caution against high shares of remote calls in complex relationships.

Sales executives should also establish strong organizational foundations to support salespeople’s hybrid selling efforts. If customers express a strong preference for remote communication channels, salespeople should accommodate them. Through their CRM systems and data analytics, selling organizations might identify these customers. By integrating AI-driven algorithms, they also might be able to personalize their customer interactions more precisely, in accordance with the unique characteristics and expectations of each customer (Bharadwaj and Shipley 2020). Such alignments promise better perceptions of relationship quality and sales performance. To foster salespeople’s interaction knowledge, selling firms should implement comprehensive CRM systems and encourage the adoption of such systems by salespeople and other users. Our partner firm offers an illustrative example. It incentivizes salespeople to maintain the quality of CRM records to preserve interaction knowledge across the sales force, which in turn creates a better foundation for effective hybrid selling. By harnessing such data-driven insights, salespeople can anticipate customer needs and preferences with greater accuracy. Furthermore, if salespeople leverage real-time feedback loops and sentiment analysis tools during remote calls, based on their existing interaction knowledge, they can address concerns promptly and fine-tune their communication strategies (Bharadwaj and Shipley 2020; Luo et al. 2020). Such proactive, data-driven approaches should help mitigate the potential negative effects of remote calls, while also fostering trust and enhancing performance outcomes.

Limitations and Avenues for Further Research

Some limitations of this study offer avenues for further research. First, our study’s research context is based on relationships involving existing customers. Noting the need for rapid relationship development and the lack of trust in early relationship stages (Moffett et al. 2021), the financial and relational effects of hybrid selling likely differ substantially for customer acquisition efforts. Furthermore, while conceptually discussed in our hypotheses, we had no possibility to directly measure trust due to the missing availability of data. Future research could analyze the crucial role of trust as a mediating mechanism of the hybrid selling-performance path, as well as differences of hybrid selling effectiveness in customer acquisition settings to establish more comprehensive insights into mechanisms and contingencies.

Second, we were not able to investigate broader dynamics of hybrid selling as we conceptualized hybrid selling as the mix of face-to-face and remote calls. Such dynamics could reflect sequencing of interaction formats (e.g., temporal factors, such as interaction sequences along the customers buying cycle) and matching of formats to task-specific needs of a selling situation (e.g., message factors, such as need recognition, technical specification of requirements, explanation of technical or economic uses or price negotiation) (see multiformat communication theory by Moffet et al. 2021). Investigating such contingencies could add more depth into the broader dynamics of hybrid selling and additionally help salespeople to understand how to adjust their interaction formats to specific selling situations over time when selling hybrid.

Third, the customer samples are large in both of our reported studies. However, they include only 27 male salespeople of a European manufacturer, within whom all the customers are nested. While an overrepresentation of male salespeople is common in the industrial context that we investigated, we believe that potential gender-induced communication differences or format impact (see work on gender differences in Zoom Fatigue by Fauville et al. 2022) might affect hybrid selling effectiveness. Methodologically, we account for this by including salesperson and industry fixed effects, clustering at the salesperson level in our main analysis, and calculating mixed-effects models with random intercepts for salesperson and industry clusters in robustness checks (Rogers 1994). However, further research could include a larger set of more (gender-) diverse salespeople to account for the potential heterogeneity among salespeople and also enhance the representativeness of our findings.

Fourth, despite the inclusion of salesperson fixed effects, we cannot analyze if and how salesperson experience might influence hybrid selling effectiveness. Prior research cites other contingencies of the effectiveness of remote interactions, such as customer relationship tenure, span of control, competitive intensity, and message or dyadic factors (Ahearne et al. 2025; Mangus et al. 2024; Moffett et al. 2021; Shi et al. 2024). Salesperson experience could be another important influence on the adoption and temporal or dynamic coordination of interactions (e.g., how salespeople schedule remote vs. face-to-face interactions to optimize efficiency and customer availability).

Fifth, working closely with a single industrial organization enhanced our ability to collect rich data across a wide range of customers. However, it also limits the potential generalizability of the results, beyond settings marked by firms that sell a broad portfolio of technical products, using a field sales force, whose salespeople rely on strong in-person relationships with their customers. Considering how common such settings are in B2B industries, our results might transfer to other B2B contexts, but they cannot offer prescriptions. Rather, they should serve as guidelines for other firms.

While acknowledging these limitations, the conceptual framework and findings presented herein should prove useful for salespeople and executives seeking guidance for developing effective hybrid selling approaches. We hope these insights into the critical role of the salespeople in achieving sales performance through hybrid selling also lead to further research in this important area.

Footnotes

Appendix

Appendix C.

Mixed-Effects Model Calculations to Account for Hierarchical Data Structure.

		Mixed Effects Regression Models
		Model 7:	Model 8:
Dependent Variables		Share of Wallet	Relationship Quality
Independent Variables	Hypothesis	Estimates (SE)	Estimates (SE)
Main effects
Share of remote calls		−.010^n.s. (.017)	−.042^n.s. (.113)
Share of remote calls²	H₁/H₂	−.082** (.041)	−1.250** (.487)
Relationship quality		.020*** (.005)
Main effects moderators
Product complexity		.004^n.s. (.003)	−.024^n.s. (.029)
Customer channel preference		−.002^n.s. (.004)	.098*** (.023)
Salesperson interaction knowledge		−.003^n.s. (.014)	−.072^n.s. (.047)
Moderation effects
Share of Remote Calls × Product Complexity	H_3a/H_3b	−.022* (.012)	−.315*** (.093)
Share of Remote Calls² × Product Complexity		.019^n.s. (.033)	−.114^n.s. (.285)
Share of Remote Calls × Customer Channel Preference	H_4a/H_4b	.010^n.s. (.012)	.279** (.133)
Share of Remote Calls² × Customer Channel Preference		−.012ⁿˢ (.028)	.785* (.408)
Share of Remote Calls × Salesperson Interaction Knowledge	H_5a/H_5b	.057*** (.015)	−.010^n.s. (.151)
Share of Remote Calls² × Salesperson Interaction Knowledge		−.009^n.s. (.055)	.430^n.s. (.600)
Control variables
Total calls		.003*** (.001)	.007^n.s. (.005)
Customer firm size		−.001^n.s. (.004)	−.019^n.s. (.018)
Relationship length		.005^n.s. (.012)	.179*** (.054)
Customer buying experience		−.003^n.s. (.013)	−.111^n.s. (.132)
Customer age		.012^n.s. (.036)	−.020^n.s. (.267)
Product complexity variance (dummy)		Included	Included
Buying department (dummy)		Included	Included
Salesperson ID random intercept		Included	Included
Industry ID random intercept		Included	Included
Akaike information criterion		−844.67	1774.78
Bayesian information criterion		−741.66	1873.30
N		651	651

Note. We report unstandardized coefficients (robust standard errors in brackets). SOW = share of wallet, RQU = relationship quality, SRC = share of remote calls, CPX = product complexity, CCP = customer channel preference, SIK = salesperson interaction knowledge. We mean-centered all focal model variables.

p < .10.**p < .05. ***p < .01. n.s. = not significant (p > .10). Two-tailed tests of significance.

Acknowledgements

We thank Gary L. Lilien for insightful discussions and his valuable feedback on earlier drafts of this paper.

ORCID iD

David Ergun

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Author Biographies

David Ergun is a post-doctoral researcher at the Sales Management Department of the Ruhr University Bochum, Germany. His research focuses on customer-salesperson interactions, sales management, and sales channels.

Christian Schmitz is a professor of Sales Management and the Chair of the Sales Management Department at the Ruhr University Bochum, Germany. His research focuses on sales management, personal selling, and business-to-business markets.

References

Ahearne

Michael

Atefi

Yashar

Lam

Son K.

Pourmasoudi

Mohsen

. 2022. “The Future of Buyer-Seller Interactions: A Conceptual Framework and Research Agenda.” Journal of the Academy of Marketing Science 50, no. (1): 22–45.

Ahearne

Molly

Pourmasoudi

Mohsen

Habel.

Johannes

2025. “Inside Sales Structures and Firm Performance.” Journal of the Academy of Marketing Science 53, no. (1): 1356–1378.

Arnold

Todd

Marinova

Detelina

. 2023. “Disrupting the Organizational Frontlines.” Journal of Service Research 26, no. (3): 303–309.

Becker

Jan-Michael

Proksch

Dorian

Ringle

Christian M.

2022. “Revisiting Gaussian Copulas to Handle Endogenous Regressors.” Journal of the Academy of Marketing Science 50, no. (1): 46–66.

Bendapudi

Neeli

Leone

Robert P.

2002. “Managing Business-to-Business Customer Relationships Following Key Contact Employee Turnover in a Vendor Firm.” Journal of Marketing 66, no. (2): 83–101.

Bharadwaj

Neeraj

Shipley

Garrett M.

2020 “Salesperson Communication Effectiveness in a Digital Sales Interaction.” Industrial Marketing Management 90: 106–12.

Boyer

Stefanie L.

Jap

Sandy D.

2022. “The Big Spaces in Sales Negotiation Research.” Journal of Personal Selling & Sales Management 42, no. (2): 181–92.

Brown

Brian P.

Zablah

Alex

Bellenger

Danny N.

Johnston

Wesley James

. 2011. “When Do B2B Brands Influence the Decision Making of Organizational Buyers? An Examination of the Relationship between Purchase Risk and Brand Sensitivity.” International Journal of Research in Marketing 28, no. (3): 194–204.

Burnham

Thomas A.

Frels

Judy K.

Mahajan

Vijay

. 2003. “Consumer Switching Costs: A Typology, Antecedents, and Consequences.” Journal of the Academy of Marketing Science 31, no. (2): 109–26.

10.

Cencioni

Paolo

Gilbertini

Jacopo

Sprengel

David

Yanni

Martina

. 2024. “Next-Gen B2B Sales: How Three Game Changers Grabbed the Opportunity.” McKinsey, (March 2024), https://www.mckinsey.com/capabilities/growth-marketing-and-sales/our-insights/next-gen-b2b-sales-how-three-game-changers-grabbed-the-opportunity#/

11.

Daft

Richard L.

Lengel

Robert H.

1986. “Organizational Information Requirements, Media Richness and Structural Design.” Management Science 32, no. (5): 554–71.

12.

De Wulf

Kristof

Odekerken-Schröder

Gaby

Iacobucci

Dawn

. 2001. “Investments in Consumer Relationships: A Cross-Country and Cross-Industry Exploration.” Journal of Marketing 65, no. (4): 33–50.

13.

Donchak

Lisa

Mcclatchy

Julia

Stanley

Jennifer

. 2022. “The Future of B2B Sales Is Hybrid.” McKinsey, (January 2025), https://www.mckinsey.com/capabilities/growth-marketing-and-sales/our-insights/the-future-of-b2b-sales-is-hybrid

14.

Donchak

Lisa

Klein

Yair

Stanley

Jennifer

. 2023. “Future of B2B Sales: Building the right team and talent to drive growth in an uncertain environment.” McKinsey, (January 2026), https://www.mckinsey.com/capabilities/growth-marketing-and-sales/our-insights/future-of-b2b-sales-building-the-right-team

15.

Dotzel

Thomas

Shankar

Venkatesh

. 2019. “The Relative Effects of Business-to-Business (Vs. Business-to-Consumer) Service Innovations on Firm Value and Firm Risk: An Empirical Analysis.” Journal of Marketing 83, no. (5): 133–52.

16.

English

Larry

. 2022. “Hybrid B2B Sales Will Be the Norm by 2024. How to Embrace the Future of Sales Now.” Forbes, (August 2023), https://www.forbes.com/sites/larryenglish/2022/08/23/hybrid-b2b-sales-will-be-the-norm-by-2024-how-to-embrace-the-future-of-sales-now/

17.

Fauville

Géraldine

Luo

Mufan

Queiroz

Anna C. M.

Bailenson

Jeremy N.

Hancock

Jeffrey

. 2021. “Nonverbal Mechanisms Predict Zoom Fatigue and Explain Why Women Experience Higher Levels than Men.” SSRN Electronic Journal 3820035: 1–18.

18.

Fitzsimons

Gavan J.

Lehmann

Donald R.

2004. “Reactance to Recommendations: When Unsolicited Advice Yields Contrary Responses.” Marketing Science 23, no. (1): 82–94.

19.

Friess

Maximilian

Haumann

Till

Alavi

Sascha

Oproiescu

Alexandru Ionut

Schmitz

Christian

Wieseke

Jan

. 2024. “The Contingent Effects of Innovative Digital Sales Technologies on B2B Firms’ Financial Performance.” International Journal of Research in Marketing 41, no. (4): 703–23.

20.

Godfrey

Andrea

Seiders

Kathleen

Voss

Glenn B.

2011. “Enough Is Enough! The Fine Line in Executing Multichannel Relational Communication.” Journal of Marketing 75, no. (4): 94–109.

21.

Haans

Richard F. J.

Pieters

Constant

Zi-Lin

. 2016. “Thinking About U: Theorizing and Testing U - and Inverted U-Shaped Relationships in Strategy Research.” Strategic Management Journal 37, no. (7): 1177–95.

22.

Heitmann

Mark

Lehmann

Donald R.

Herrmann

Andreas

. 2007. “Choice Goal Attainment and Decision and Consumption Satisfaction.” Journal of Marketing Research 44, no. (2): 234–50.

23.

Homburg

Christian

Müller

Michael

Klarmann

Martin

. 2011a. “When Does Salespeople’s Customer Orientation Lead to Customer Loyalty? The Differential Effects of Relational and Functional Customer Orientation.” Journal of the Academy of Marketing Science 39, no. (6): 795–812.

24.

Homburg

Christian

Müller

Michael

Klarmann

Martin

. 2011b. “When Should the Customer Really Be King? On the Optimum Level of Salesperson Customer Orientation in Sales Encounters.” Journal of Marketing 75, no. (2): 55–74.

25.

Hornik

Jacob

. 1992. “Tactile Stimulation and Consumer Response.” Journal of Consumer Research 19, no. (3): 449–58.

26.

Huang

Ming-Hui

Rust

Roland T.

2018. “Artificial Intelligence in Service.” Journal of Service Research 21, no. (2): 155–72.

27.

Jap

Sandy D.

2001. “‘Pie Sharing’ in Complex Collaboration Contexts.” Journal of Marketing Research 38, no. (1): 86–99.

28.

Joshi

Ashwin W.

2009. “Continuous Supplier Performance Improvement: Effects of Collaborative Communication and Control.” Journal of Marketing 73, no. (1): 133–50.

29.

Keiningham

Timothy L.

Perkins-Munn

Tiffany

Evans

Heather

. 2003. “The Impact of Customer Satisfaction on Share-of-Wallet in a Business-to-Business Environment.” Journal of Service Research 6, no. (1): 37–50.

30.

Kindermann

Bastian

Wentzel

Daniel

Antons

David

Salge

Torsten-Oliver

. 2024. “Conceptual Contributions in Marketing Scholarship: Patterns, Mechanisms, and Rebalancing Options.” Journal of Marketing 88, no. (3): 29–49.

31.

Lemon

Katherine N.

Verhoef

2016. “Understanding Customer Experience Throughout the Customer Journey.” Journal of Marketing 80, no. (6): 69–96.

32.

Luo

Xueming

Qin

Marco

Fang

Zheng

Zhe

. 2021. “Artificial Intelligence Coaches for Sales Agents: Caveats and Solutions.” Journal of Marketing 85 (2), 14–32.

33.

Macdonald

Emma K.

Kleinaltenkamp

Michael

Wilson

Hugh N.

2016. “How Business Customers Judge Solutions: Solution Quality and Value in Use.” Journal of Marketing 80, no. (3): 96–120.

34.

MacInnis

Deborah J.

2011. “A Framework for Conceptual Contributions in Marketing.” Journal of Marketing 75, no. (4): 136–54.

35.

Mangus

Stephanie M.

Shi

Huanhuan

Folse

Judith A. G.

Jones

Eli

Sridhar

Shrihari

. 2024. “Communicating with B2B Buyers After ‘Dropping the Ball’ Using Digital and Non-Digital Communication Formats to Recover from Salesperson Transgressions.” International Journal of Research in Marketing 41, no (2): 194–219.

36.

Marinova

Detelina

Singh

Sunil K.

Singh

Jagdip

. 2018. “Frontline Problem-Solving Effectiveness: A Dynamic Analysis of Verbal and Nonverbal Cues.” Journal of Marketing Research 55, no. (2): 178–92.

37.

Moffett

Jordan W.

Folse

Judith A. G.

Palmatier

Robert W.

2021. “A Theory of Multiformat Communication: Mechanisms, Dynamics, and Strategies.” Journal of the Academy of Marketing Science 49, no. (3): 441–61.

38.

Mullins

Ryan

Agnihotri

Raj

. 2022. “Digital Selling: Organizational and Managerial Influences for Frontline Readiness and Effectiveness.” Journal of the Academy of Marketing Science 50, no. (4): 800–21.

39.

Mullins

Ryan R.

Ahearne

Michael

Lam

Son K.

Hall

Zachary R.

Boichuk

Jeffrey P.

2014. “Know your Customer: How Salesperson Perceptions of Customer Relationship Quality Form and Influence Account Profitability.” Journal of Marketing 78, no. (6): 38–58.

40.

Palmatier

Robert W.

2008. Relationship Marketing. Marketing Science Institute.

41.

Palmatier

Robert W.

Scheer

Lisa K.

Houston

Mark B.

Evans

Kenneth R.

Gopalakrishna

Srinath

. 2007. “Use of Relationship Marketing Programs in Building Customer–Salesperson and Customer–Firm Relationships: Differential Influences on Financial Outcomes.” International Journal of Research in Marketing 24, no. (3): 210–23.

42.

Park

Sungho

Gupta

Sachin

. 2012. “Handling Endogenous Regressors by Joint Estimation Using Copulas.” Marketing Science 31, no. (4): 567–86.

43.

Patterson

Paul G.

Johnson

Lester W.

Spreng

Richard A.

1997. “Modeling the Determinants of Customer Satisfaction for Business-to-Business Professional Services.” Journal of the Academy of Marketing Science 25, no. (1): 4–17.

44.

Plotkin

Candace L.

Stanley

Jennifer

Harrison

Liz

de la Torre

Victor

. 2024. “Five Fundamental Truths: How B2B Winners Keep Growing.” McKinsey, (September 2024), https://www.mckinsey.com/capabilities/growth-marketing-and-sales/our-insights/five-fundamental-truths-how-b2b-winners-keep-growing#/

45.

Rogers

William

. 1994. “Regression Standard Errors in Clustered Samples.” Stata Technical Bulletin 3, no. (13): 19–23.

46.

Ramos

Carla

Claro

Danny P.

Germiniano

Renato

. 2023. “The Effect of Inside Sales and Hybrid Sales Structures on Customer Value Creation.” Journal of Business Research 154: 1–19.

47.

Reinartz

Werner

Thomas

Jacquelyn S.

Kumar

2005. “Balancing Acquisition and Retention Resources to Maximize Customer Profitability.” Journal of Marketing 69, no. (1): 63–79.

48.

Schepers

Jeroen J.

der Borgh

Michel van

. 2020. “A Meta-Analysis of Frontline Employees’ Role Behavior and the Moderating Effects of National Culture.” Journal of Service Research 23, no. (3): 255–80.

49.

Schmitz

Christian

Huckemann

Matthias

Ergun

David

Hohmann

Mara

. 2021. “Hybrid Selling: Mehr Vertriebsproduktivität Durch Synchronisierte Vor Ort Und Online Besuche.” (March 2024), https://smd.rub.de/hybrid-selling/

50.

Schmitz

Christian

Lee

You-Cheong

Lilien

Gary L.

2014. “Cross-Selling Performance in Complex Selling Contexts: An Examination of Supervisory- and Compensation-Based Controls.” Journal of Marketing 78, no. (3): 1–19.

51.

Sheth

J. N.

Sisodia

R. S.

Sharma

2000. “The Antecedents and Consequences of Customer-Centric Marketing.” Journal of the Academy of Marketing Science 28, no. (1): 55–66.

52.

Shi

Huanhuan

Sridhar

Shrihari

Grewal

Rajdeep

. 2024. “Building Effective Inside-Outside Sales Rep Dyads: A Collaboration Perspective.” Journal of the Academy of Marketing Science 52, no. (3): 835–58.

53.

Short

Williams

Christie

1976. The Social Psychology of Telecommunications. Wiley.

54.

Singh

Jagdip

Brady

Michael

Arnold

Todd

Brown

Tom

. 2017. “The Emergent Field of Organizational Frontlines.” Journal of Service Research 20, no. (1): 3–11.

55.

Singh

Jagdip

Bridge

R. G.

2023. “Interfaces, Interactions, Time, and the Frontline Nexus: Foundational Constructs and Focus for the Field of Organizational Frontlines.” Journal of Service Research 26, no. (3): 310–29.

56.

Sleep

Stefan

Dixon

Andrea L.

DeCarlo

Thomas

Lam

Son K.

2020. “The Business-to-Business Inside Sales Force: Roles, Configurations and Research Agenda.” European Journal of Marketing 54, no. (5): 1025–60.

57.

Spiro

Rosann L.

Weitz

Barton A.

1990. “Adaptive Selling: Conceptualization, Measurement, and Nomological Validity.” Journal of Marketing Research 27, no. (1): 61.

58.

Tarafdar

Monideepa

Pullins

Ellen B.

Ragu-Nathan

T. S.

2014. “Examining Impacts of Technostress on the Professional Salesperson’s Behavioural Performance.” Journal of Personal Selling & Sales Management 34, no. (1): 51–69.

59.

Thaichon

Park

Surachartkumtonkun

Jiraporn

Quach

Sara

Weaven

Scott

Palmatier

Robert W.

2018. “Hybrid Sales Structures in the Age of E-Commerce.” Journal of Personal Selling & Sales Management 38, no. (3): 277–302.

60.

Travasoni

Andrea

Fiandro

Luca

Atluri

Venkat

Labianca

Marco

. 2024. “The Future of Telcos: Mapping the Routes to Renewed Success.” McKinsey, (December 2024), https://www.mckinsey.com/industries/technology-media-and-telecommunications/our-insights/the-future-of-telcos-mapping-the-routes-to-renewed-success

61.

van Doorn

Jenny

Mende

Martin

Noble

Stephanie M.

Hulland

John

Ostrom

Amy L.

Grewal

Dhruv

Petersen

J. Andrew

. 2017. “Domo Arigato Mr. Roboto: Emergence of Automated Social Presence in Organizational Frontlines and Customers’ Service Experiences.” Journal of Service Research 20, no. (1): 43–58.

62.

Venkatesan

Rajkumar

Kumar

2004. “A Customer Lifetime Value Framework for Customer Selection and Resource Allocation Strategy.” Journal of Marketing 68, no. (4): 106–25.

63.

Vickery

Shawnee K.

Droge

Cornelia

Stank

Theodore P.

Goldsby

Thomas J.

Markland

Robert E.

2004. “The Performance Implications of Media Richness in a Business-to-Business Service Environment: Direct Versus Indirect Effects.” Management Science 50, no. (8): 1106–19.

64.

Walther

Joseph B.

1996. “Computer-Mediated Communication: Impersonal, Interpersonal, and Hyperpersonal Interaction.” Communication Research 23, no. (1): 3–43.

65.

Webster

Frederick E.

Wind

Yoram

. 1972. “A General Model for Understanding Organizational Buying Behavior.” Journal of Marketing 36, no. (2): 12.

66.

Juan

der Borgh

Michel van

Nijssen

Edwin J.

Lam

Son K.

2022. “Why Salespeople Avoid Big-Whale Sales Opportunities.” Journal of Marketing 86, no. (5): 95–116.

67.

Zeithaml

V. A.

2000. “Service Quality, Profitability, and the Economic Worth of Customers: What We Know and What We Need to Learn.” Journal of the Academy of Marketing Science 28, no. (1): 67–85.

68.

Zoltners

Andris A.

Sinha

Prabhakant

Lorimer

Sally E.

Curry

. 2022. “Where Do Salespeople Fit in the Digital World?” (August 2023), https://hbr.org/2022/01/where-do-salespeople-fit-in-the-digital-world