Getting on top of work‐email: A systematic review of 25 years of research to understand effective work‐email activity - Russell - 2024 - Journal of Occupational and Organizational Psychology - Wiley Online Library

Source: original

Journal of Occupational and Organizational Psychology

Volume 97, Issue 1 pp. 74-103Journal of Occupational and Organizational Psychology

ARTICLE

__

Open Access

Getting on top of work-email: A systematic review of 25 years of research to understand effective work-email activity

Emma Russell ____,

Corresponding Author

Emma Russell

University of Sussex Business School, University of Sussex, Falmer, UK

Correspondence

Emma Russell, University of Sussex Business School, University of Sussex, Jubilee Building, Falmer, East Sussex BN1 9SL, UK.

Email: [email protected]

Contribution: Conceptualization, Data curation, Formal analysis, ​Investigation, Methodology, Project administration, Resources, Supervision, Validation, Visualization, Writing - original draft, Writing - review & editing

Search for more papers by this author

Thomas W. Jackson __,

Thomas W. Jackson

School of Business and Economics, Loughborough University, Loughborough, UK

Contribution: Conceptualization, Formal analysis, ​Investigation, Resources, Visualization, Writing - original draft, Writing - review & editing

Search for more papers by this author

Marc Fullman __,

Marc Fullman

University of Sussex Business School, University of Sussex, Falmer, UK

Contribution: Data curation, Formal analysis, ​Investigation, Project administration, Software, Writing - original draft

Search for more papers by this author

Petros Chamakiotis __,

Petros Chamakiotis

Department of Management, ESCP Business School, Madrid, Spain

Contribution: Formal analysis, Validation, Writing - original draft, Writing - review & editing

Search for more papers by this author

Emma Russell ____,

Corresponding Author

Emma Russell

University of Sussex Business School, University of Sussex, Falmer, UK

Correspondence

Emma Russell, University of Sussex Business School, University of Sussex, Jubilee Building, Falmer, East Sussex BN1 9SL, UK.

Email: [email protected]

Contribution: Conceptualization, Data curation, Formal analysis, ​Investigation, Methodology, Project administration, Resources, Supervision, Validation, Visualization, Writing - original draft, Writing - review & editing

Search for more papers by this author

Thomas W. Jackson __,

Thomas W. Jackson

School of Business and Economics, Loughborough University, Loughborough, UK

Contribution: Conceptualization, Formal analysis, ​Investigation, Resources, Visualization, Writing - original draft, Writing - review & editing

Search for more papers by this author

Marc Fullman __,

Marc Fullman

University of Sussex Business School, University of Sussex, Falmer, UK

Contribution: Data curation, Formal analysis, ​Investigation, Project administration, Software, Writing - original draft

Search for more papers by this author

Petros Chamakiotis __,

Petros Chamakiotis

Department of Management, ESCP Business School, Madrid, Spain

Contribution: Formal analysis, Validation, Writing - original draft, Writing - review & editing

Search for more papers by this author

First published: 23 August 2023

https://doi.org/10.1111/joop.12462Digital Object Identifier (DOI)

__view metrics

Accessibility issue? Request accessibility update.

__About

__Sections

__ CITE

__Tools

Give access

__Share full text access

__

Share full-text access

Please review our Terms and Conditions of Use and check box below to share full-text version of article.

I have read and accept the Wiley Online Library Terms and Conditions of Use Shareable Link

Use the link below to share a full-text version of this article with your friends and colleagues. Learn more.

Copy URL

Share a link

Share on

Abstract

In the last 25 years, work-email activity has been studied across domains and disciplines. Yet, despite the abundance of research available, a comprehensive, unifying framework of how work-email activity positively impacts both well-being and work-performance outcomes has yet to emerge. This is a timely and significant concern; work-email is the most prominent and popular form of work communication but it is still unclear what people need to do to be effective emailers at work. To address this, we undertook a rigorous cross-disciplinary systematic literature review of 62 empirical papers. Using action regulation theory, we developed a multi-action, multi-goal framework and found four ‘super’ actions that consistently predict effectiveness (positive well-being and work-performance outcomes). These actions involve: (i) communicating and adhering to work-email access boundaries; (ii) regularly triaging emails (iii) sending work-relevant email and (iv) being civil and considerate in work-email exchanges. We found that super actions are engaged when workers have the resources to appropriately regulate their activity, and can attend to their self, task and social needs. Our framework synthesizes a broad and disparate research field, providing valuable insights and guiding future research directions. It also offers practical recommendations to organizations and individuals; by understanding and encouraging the adoption of work-email super actions, effective work-email practices can be enhanced.

Practitioner Points

INTRODUCTION

More than 45 years after the first electronic mail message was sent (Leiner et al., 2009), email has fully permeated working life. Despite the emergence of various new contenders, like Slack, Teams or Yammer, email is still the most used form of business communication, the preferred method of communication for managers and professionals, and shows little sign of being surpassed in the near future (Ofcom, 2019; Radicati Group, 2021). The continued popularity of email at work can be understood by attending to its manifold benefits. Work-email helps workers communicate, manage task lists and multiple project strands, co-ordinate information, plan and schedule meetings, keep an audit trail, build relationships and work conveniently and flexibly (Dabbish et al., 2005; Dawley & Anthony, 2003; Middleton & Cukier, 2006; O'Kane & Hargie, 2007). Yet, there exists a paradox in work-email activity as, despite its high level of functionality, a combination of poor work strategies, pressurized work cultures and design limitations means that email is also a major contributor towards impaired organizational performance and work-related stress (Barley et al., 2011; Dawley & Anthony, 2003; Middleton & Cukier, 2006).

Within academia, there exists a significant body of research into understanding the application and use of email across multiple methodologies and disciplines, including psychology, organization and management studies and information systems domains. However, to date, no unifying and organizing conceptual framework has been developed to make sense of the widely varied (and often paradoxical) findings across the multitude of studies reported. This renders both researchers and end-users somewhat confused about what constitutes effective work-email use. On the one hand, research findings outline how work-email offers a host of benefits for busy workers trying to organize and deal with their work in a convenient and flexible way. On the other hand, studies report that workers are tired of being tethered to their email, constantly interrupted by it and drowning under the weight of tasks to be achieved, and ambiguous messages to decipher (Addas & Pinsonneault, 2018; Barley et al., 2011; Dawley & Anthony, 2003; Mazmanian et al., 2013; Middleton & Cukier, 2006; Russell et al., 2017). This is concerning as it is increasingly difficult to predict how actions relating to work-email use will have positive or negative repercussions for performance and well-being as a whole. This is important because when people are able to engage in activity that enhances both their well-being and their work-performance, this is a marker of thriving, success or ‘effectiveness’ (Brown et al., 2017; Carver & Scheier, 1990; Hobfoll et al., 2018; Russell, Daniels, et al., 2021). Because we do not have a clear, organizing paradigm to help us understand when work-email activity will predict overall effective outcomes, end-users and managers are forced to rely on the prolific, but potentially unsound, advice being proffered by an abundance of self-help books, websites and email ‘gurus’.

The primary purpose of this paper is, therefore, to develop an evidence-based, unifying conceptual framework of effective work-email activity, from which future research directions and end-user recommendations can be garnered. To develop such a framework, we undertake a cross-disciplinary review of 25 years of research literature into work-email use, and ask the question, “what is effective work-email activity?” Our research is framed by action regulation theory (ART; Frese & Zapf, 1994; Hacker, 1985, 1994; Zacher & Frese, 2018), which allows us to adopt a multi-action, multi-goal perspective in understanding the various actions, outcomes and contextual factors associated with work-email activity. Since ART does not explicitly focus on the impact of action on well-being (a key measure of effectiveness), we also draw on the conservation of resources (COR) theory (Hobfoll, 1989; Hobfoll et al., 2018) to further explain its manifestation in work-email activity. By taking this approach, our paper makes three significant contributions.

Our first contribution is the development of an up-to-date and cohesive multi-action, multi-goal framework that explains and predicts effective work-email activity. This approach is essential given the current tendency in empirical research to narrowly focus on specific work-email actions (e.g. managing interruptions, engaging in after-hours connectivity) and outcomes (e.g. email overload, work-life conflict). This narrow focus may overlook the complex and multifunctional ways in which individuals interact with work-email, as it relates to their broader work and personal objectives (Chen & Karahanna, 2018; Sun & Frese, 2013). Consequently, research examining email actions and goals in isolation has resulted in a growing body of disconnected and sometimes conflicting findings. To address this, we synthesize existing research to provide clarity and coherence. Our framework emphasizes the dynamic nature of workers' engagement with work-email and identifies the mechanisms and conditions that facilitate the simultaneous achievement of multiple goals relating to well-being and work-performance.

Our second contribution is the identification of four distinct categories of work-email actions that simultaneously promote both work-performance and well-being goals. We refer to these actions as “super actions.” This contribution enhances our understanding because most empirical studies on work-email primarily focus on either work-performance or well-being outcomes, rarely considering them together or exploring the reasons why individuals may prioritize certain actions (e.g. emailing after hours) to achieve one set of goals (e.g. increased productivity) at the expense of another (e.g. work-life balance). Both work-performance and well-being outcomes are crucial concerns for occupational and organizational psychologists, and together are often used as indicators of effectiveness that potentiate thriving at work (Brown et al., 2017; Carver & Scheier, 1990; Hobfoll et al., 2018; Quick et al., 2007; Russell, Daniels, et al., 2021). By explicating that effective work-email action involves achieving goals for well-being and work-performance, we specify, for the first time, clear categories of super actions that constitute effective work-email use.

Our third contribution is the application of action regulation theory (ART) to provide a clear definition of actions and super actions, enabling end-users and practitioners to understand the key elements of effective work-email activity. In brief, actions are quantifiable, goal-directed behaviours (rather than perceptions or feelings) that are under the control of the individual (i.e. not a reflex). Super actions are those behaviours that satisfy both work-performance and well-being goals. By leveraging established principles within ART, relating to conditions of execution, situational parameters and levels of regulation, we can predict and explain the positive and negative consequences of work-email actions on work-performance and well-being. To address the concept of well-being, which ART does not explicitly cover, we incorporate the conservation of resources (COR) theory (Hobfoll, 1989; Hobfoll et al., 2018). COR theory posits that when individuals invest in, build, and protect resources that contribute to the attainment of their valued goals, they experience a sense of well-being.1 By focusing on actions as the core units of analysis in our framework, we offer evidence-based recommendations and guidance on how workers can enhance their effectiveness in work-email activities. Our framework serves as a practical resource for researchers, policymakers and practitioners seeking to gain insights into the consequences of different actions on various outcomes, as well as how optimizing conditions and resources can ensure effective work-email activity across different situations and contexts.

In summary, this research is a timely and necessary endeavour that seeks to enhance our understanding of how the multitude of work-email actions, studied across different disciplines, influence goals related to both work-performance and well-being (referred to as “effectiveness”). By synthesizing and summarizing the factors that contribute to effective outcomes in work-email activity, we aim to guide future research and enable scholars to design studies that expand knowledge and generate predictions about work-email usage in current and future work environments.

In the upcoming section, we introduce action regulation theory (ART) by providing an overview of its key concepts and definitions. This positioning establishes ART as a suitable framework to frame our review and conceptualization of effective work-email activity. We then outline our systematic literature review (SLR) approach and utilize the findings from the SLR to present the development of our comprehensive framework, which incorporates multiple actions and goals in relation to work-email activity. In the Discussion section, we explore the implications of this research and offer suggestions for further validation and expansion of our framework. Additionally, we provide recommendations for organizations and end-users, offering practical insights derived from our findings.

Theoretical background for examining effective work-email activity

ART is a cognitive–motivational meta-theory about work that explains how people engage in processes of action regulation to achieve their valued goals as efficiently as possible (Frese & Zapf, 1994; Hirschi et al., 2019; Zacher & Frese, 2018). Goals are mental representations that regulate and motivate individual activity (Hacker, 1985). Actions (or action plans) are purposeful, individual behavioural programmes, selected from memory according to their suitability for the current context (Frese & Sabini, 1985; Hacker, 1985). If actions allow an individual to achieve personal and organizational goals at minimum cost to their personal resources then the activity is considered to be effective or ‘behaviourally economic’ (Hacker, 1985, 1994; Schönpflug, 1983, 1986). Activity involves continuous regulation. Workers must monitor how action is contributing towards goal progress; negative feedback from the environment means workers can adapt their action plans if progress is being impeded, or if goal salience changes (Frese & Zapf, 1994; Zacher & Frese, 2018).

Regulation of an action plan uses cognitive resources and is hierarchically arranged from: (i) the lowest, relatively unconscious and automatic level, requiring few resources; to (ii) a more perceptual/conceptual level that requires some conscious processing; to (iii) the highest, intellectual level that involves conscious effort and resources (Hacker, 1985). Most of the time, workers are operating at low levels of regulation, using well-practised plans to undertake their tasks with little conscious thought given to distal organizational and personal goals (Zacher & Frese, 2018). Although ART emphasizes that efficient action involves working at the lowest level of regulation (as this requires fewer costs to resources; Hacker, 1985, 1994), a reduction in cognitive effort and attention in this mode can mean that important environmental feedback or learning opportunities are missed, which then negatively impacts action execution and subsequent goal achievement. Accordingly, higher levels of regulation are required when a situation/context is unfamiliar, the work task is more complex, and/or the associated goal more salient or important (Hacker, 2003). People may also opt to work at higher levels of regulation to grow, learn and develop and to undertake regular checks as to whether they are using the best possible actions to optimize outcomes (Hacker, 1985, 1994). From a COR perspective, investing resources of time and energy, to build resources for knowledge and skills, is an effective undertaking (Hobfoll, 2001).

ART stipulates that a worker must have decision latitude or control over their work to choose the best actions and level of regulation to employ in different conditions. By scanning current ‘conditions of execution’, the worker will decide how much effort or attention is required (i.e. the level of regulation), and will select an appropriate action plan to employ. Conditions of execution include (i) situational parameters (e.g. contextual conditions, work barriers and facilitators); (ii) personal resources (e.g. knowledge, skills, energy and action styles available to the worker); (iii) object partners (e.g. the work tools and materials available to the worker) and (iv) activity partners (those with whom one might interact; Hacker, 1985). These conditions are appraised prior to and during the execution of action.

Why is ART helpful in understanding effective work-email activity?

ART goes beyond examining singular actions in relation to singular goals (Frese et al., 1987) to provide an understanding of ‘complete’ activity. This means that an array of actions and goals – and the ways in which they interact and change during an activity cycle – can be examined in relation to the person, their action history and their organizational context and goals (Zacher & Frese, 2018). ART is, therefore, a suitable theory to apply in developing a multi-action, multi-goal framework of effective work-email activity (Hirschi et al., 2019), as it allows us to consider the breadth of email actions that have been studied in the extant literature to date, and how these impact different work-performance and well-being outcomes. ART has already been successfully applied to the study of work-email activity, most notably in the domain of email interruption research (Addas & Pinsonneault, 2018; Baethge & Rigotti, 2013; Russell et al., 2007; Russell, Jackson, & Banks, 2021; Zijlstra et al., 1999), where different actions for dealing with incoming email have been explained in terms of the levels of regulation applied, according to different situational parameters, conditions and scenarios (conditions of execution).

ART principles of behavioural economy can help us to understand why people appear to employ work-email actions that have paradoxical impacts. For example, a worker who responds immediately and reactively to an incoming email interruption may be doing so because they are operating at a low level of regulation, applying an oft-used and well-practised action that in the past has satisfied work-performance goals related to being responsive to others. Employing a familiar action plan to a familiar situation requires little effort or ‘cost’ but affords an immediate performance ‘benefit’. However, attending rapidly and automatically to work-email can promote ‘always on’ responding, resulting in longer term problems for well-being. In this example, it could well be more behaviourally economic for the worker to engage a higher level of regulation to evaluate whether an alternative approach (e.g. occasional but regular email checks) could better serve well-being and work-performance goals. ART, thus, acknowledges that more conscious and deliberate regulation of action can sometimes be more beneficial in the long term (Hacker, 1985, 1994).

Whilst ART does not explicitly refer to well-being in its discussion of regulation and behavioural economy, COR theory elucidates this well. According to COR, people strive to conserve and acquire resources (e.g. time, energy, secure work, stable relationships) that support the achievement of valued goals and protect against stressors (Halbesleben et al., 2014; Hobfoll et al., 2018). In COR theory, people are said to invest resources in order to protect against (or recover from) resource loss, or to gain resources (Hobfoll, 2001). This explains why it can be helpful to operate at higher levels of regulation in acting because – despite the resource expense involved in monitoring action at higher levels – securing work and well-being goal success, deeper learning and skill acquisition are all appealing resource gains. COR theory explains that those who already have good resource reserves are better able to invest more energy and effort in acting, and have a better chance of securing positive work and well-being outcomes (Hobfoll et al., 2018).

In this paper, ART provides a useful theory for understanding the multitude of actions that agentic workers adopt, regulate and evaluate and the extent to which these help them to achieve their valued goals (Frese & Zapf, 1994; Hacker, 1994; Sun & Frese, 2013). COR theory is used to help us to understand how resources are involved in this activity, and why some actions may be deployed – despite being resource intensive – if they may also help to build resources. To summarize, in this research our aim is to build a multi-action, multi-goal conceptual framework of effective work-email activity, using ART and COR theory to frame and make sense of findings from a cross-disciplinary systematic review of the empirical academic literature. Our methodology is now outlined below.

THE SYSTEMATIC LITERATURE REVIEW METHOD

We conducted a cross-disciplinary systematic literature review (SLR) to undertake a robust analysis of the multi-paradigm and multi-method research into work-email activity using a transparent, process-driven and scientific approach. An SLR approach is especially suitable when the aim is to synthesize and advance knowledge using cross-disciplinary methodologies and analytical approaches, as it equalizes the contribution of both qualitative and quantitative research and accommodates contextual, sampling and time differences across studies (Pati & Lorusso, 2018; Rojon et al., 2021; Rousseau et al., 2008).

There were five objectives to this SLR: (i) categorize the plethora of work-email actions that have been studied to date, (ii) identify the categories of work-performance and well-being goals towards which work-email actions are directed, (iii) examine the direction of relationships between each action category and each goal category, (iv) uncover the boundary conditions and influential mechanisms that appear to dynamically drive or moderate action-goal relationships and (v) isolate the action categories that consistently predict positive outcomes of work-performance and well-being, as indicators of effectiveness.

Research design

We planned our SLR approach before beginning any searches, as per advice given by Rousseau et al. (2008). A review protocol was set up (Briner & Denyer, 2012; Petticrew & Roberts, 2008: see Appendix S1; protocol registered at: osf.io/wzqke), and a steering committee of seven ‘experts’ in the field2 was formed to advise on decisions (including choice of databases, use of search terms, inclusion and exclusion criteria). We sourced literature from psychology, management and information systems databases, along with practitioner reports, conference proceedings and other ‘grey’ studies3 (Briner & Denyer, 2012). PsycINFO, Business Source Premier (including conference proceedings), Web of Science and ETHOS databases were accessed, using search terms for ‘e*mail’; ‘electronic mail’; ‘smartphone’, with outcomes relating to our criteria of effectiveness: Work-Performance and Well-being. The full list of work-performance and well-being related search terms can be found in Appendix S1. This list was reduced from a longer list, following a scoping study. We concentrated on papers published between 1995 (when email became a significant work presence for most workers) and May 2021. After the initial scoping study, our database search was completed within 2 weeks.

Screening procedure

Pre-specified inclusion and exclusion criteria for sifting and screening the papers were agreed (Denyer & Tranfield, 2009). Papers that met inclusion criteria explicitly focused on (i) working adults using work-email, (ii) sending, receiving and managing work-email communications, (iii) related to well-being or work-performance outcomes (iv) with positive or negative repercussions. Papers were excluded if they (i) used solely student/non-working samples, (ii) were not empirical studies (e.g. reviews or opinion pieces), (iii) did not explicitly refer to work-email actions (e.g. could not be separated from voicemail or instant messenger results),4 (iv) were not written in US or UK English and (v) did not show a link between actions and outcomes. See Appendix S1 ‘Eligibility’ section for more details.

The first and third authors initially sifted the papers by title. An example paper sifted out at the title stage includes ‘Smartphone and social media addiction: Exploring the perceptions and experiences of Canadian teenagers’ (Adorjan & Ricciardelli, 2021) because the paper was clearly not about work-email use (exclusion criteria iii) and also did not refer to working adults (inclusion criteria i). The third and fourth authors then conducted high-level screening on article abstracts. An example paper sifted out at the abstract stage includes ‘E-mail infatuation’ (Parker, 2009) because the abstract clearly refers to the paper being a “satirical critique” and a “poetic exposition” (p. 663), demonstrating that this was not an empirical study linking work-email actions with outcomes (exclusion criteria ii, iii and v). Finally, the first and second authors conducted a secondary high-level screen on the retained abstracts, followed by a deep-level read and evaluation of the remaining papers.5 Papers were sifted further if necessary, on this basis. Following the full paper review, additional ‘cited’ papers (30) that had not been returned from the initial searches were identified to be considered, as per Rojon et al. (2011). These papers were subject to the same inclusion/exclusion criteria checks and sifts.

At the ‘deep-level read’ stage, we extracted data from the papers to review. We set up an Excel spreadsheet with a range of different column header fields to organize the data. The header fields asked for: the paper details (e.g. authors, paper title, journal, date, etc.); details about nine criteria to be analysed as part of the review (study aim, study type, population, sample and sampling method, study design, analysis approach, results, treatment of missing data and ethics) – see Appendix S2; a summary of whether the paper met the inclusion/exclusion criteria; and fields relating to the paper quality assessments, conclusions and additional notes. Although the spreadsheet is too large to include here, Figure 1 below presents an example of the data we extracted under some of the header fields for a sample paper (Belkin et al., 2020).

Details are in the caption following the image

FIGURE 1

Open in figure viewer __PowerPoint

Example of data extracted when reviewing a paper.

To establish the ‘quality’ of a paper, each of the nine review criteria was allocated a score based on whether the information provided was sufficient and appropriate for this review (see Appendix S2). We scored this for each criterion, whereby 1 = sufficient and appropriate, .5 = partially sufficient and appropriate and 0 = Neither sufficient nor appropriate. An overall quality score for the appropriateness of the information provided by each paper was calculated (ranging from one (low quality) to nine (high quality)),6 and papers with a score of 6/9 or more were retained. This follows guidance from Briner and Denyer (2012), Rojon et al. (2011) and Valentine (2019). To illustrate, for the Belkin et al. (2020) Paper 5 (see Figure 1 and footnote 5), this paper achieved a quality score of 8/9 because it provided satisfactory coverage of eight out of the nine fields where we require information (see Appendix S2), but did not include information about the ethical treatment of participants. However, 8/9 is considered a good enough score for the study to be included, with enough detail for us to review the paper in relation to our research aims. An example of a paper excluded from the study on the grounds of quality (i.e. not providing sufficient, relevant and appropriate information for us to review it on three or more fields) can be found in Appendix S2, with a full explanation.

Quality scores from the first and third authors showed an agreement level of kappa .66. Where there was disagreement, the authors discussed whether the paper should be included until an agreement was reached. Based on this process, 62 papers were selected, on which the SLR is based (see Figure 2 for PRISMA flowchart).

Details are in the caption following the image

FIGURE 2

Open in figure viewer __PowerPoint

A summary of papers returned at each stage of the SLR process.

Analysing and synthesizing the literature from the SLR

An explanatory/interpretive narrative synthesis approach was used to analyse the data compiled in our Excel spreadsheet (as per Briner & Denyer, 2012; Rousseau et al., 2008). Explanatory/interpretive synthesis gives equal status to qualitative and quantitative findings from a diverse research literature, and allows data across these studies to be coded and categorized together, to explain the phenomena of interest. Explanatory/interpretive narrative synthesis means that robust studies from multiple fields and epistemologies are unlikely to be overlooked as a result of researcher bias or ‘cherry picking’ (see Rousseau et al., 2008 for a detailed discussion). A summary of the wide range of methods and analysis approaches used in the papers returned from the SLR can be found in Table 1.

TABLE 1. Characteristics of the papers returned from the SLR.

Methods used in the SLR papers1 | Number of papers returned

Quantitative2 | 493

Qualitative5 | 116

Mixed methods | 5

Analysis Approach Used in the SLR Papers | Number of papers returned

Quantitative7

Qualitative

A structural coding approach was used to make sense of the data across these studies (Saldana, 2011). We sought to define categories and codes of actions and outcomes (goals). First, we decoded the data using a bottom-up approach to establish initial, discrete codes from the articles. This was independently conducted by the first and third authors. Then, we encoded the data by aggregating the decoded discrete themes into broader overall categories, consolidated via discussion.

For each paper, we listed the reported email actions and outcomes in a spreadsheet. We used ART principles to agree that an action is (i) a behavioural execution on the email system (as opposed to a feeling or thought/perception about email), (ii) with a purpose or goal direction,7 (iii) quantifiable to a greater or lesser extent and (iv) under one's control (i.e. the actor/user had to have a degree of choice/decision latitude over it). By using these four criteria, some email actions reported in the literature were not included in our spreadsheet (e.g. addictive responding to email was removed as it implies a lack of choice or control over the action – point iv). Some actions were reconfigured to meet our criteria (e.g. being a recipient of uncivil email indicates a lack of choice or control over an action, so this ‘action’ category was reframed to refer to the perpetration of uncivil email behaviours as actions under the actor/user's control). Finally, only actions that were reported to have an impact on the work-performance or well-being goals of the actor/user were included. Although many actions will impact the goals of others (e.g. team members), and goals beyond the work domain (e.g. the family), a consideration of this was beyond the scope of the present study.8

Having listed the email actions, we then coded these into broader themes. We created 13 action category codes (1.1–4.4), which were organized according to four broad action themes (see Table 2). The four broad action themes were as follows: Timing Actions – where the purpose is to control the temporal impact of work-email; Organizing Actions – where the purpose is to control how work-email is managed, organized, stored or retrieved; Crafting Actions – where the purpose is to craft and construct a work-email message; and Communication Actions – where the purpose is to preference email as a communication channel in the context. Each of the 13 action categories was considered to be classifiable into one of the four possible action categories with a low likelihood of a secondary loading or overlap.

TABLE 2. Action themes and categories in work-email activity.

Action themes | Purpose/goal | Action categories | Quantifiable as: | Behavioural execution? | Actor has choice/control?

1. Timing Actions | Purpose is to control the temporal impact of email | 1.1 Uses work-email out-of-work hours | The extent to which a person connects to work-email outside of contracted working hours | Y | Y 1.2 Reactive or frequent checking and responding to work-email | The extent to which a person checks work-email immediately on alert or with a high level of frequency (e.g. compared to others) | Y | Y 1.3 Allows continuous work-email notifications | The extent to which a person allows work-email to interrupt them – that is by keeping notifications on and keeping send/receive server download times to a short (frequent) checking period | Y | Y 1.4 Communicates and adheres to boundaries for work-email access times | The extent to which a person informs others as to when they are available for dealing with work-email (e.g. response time estimates, out-of-hours access) | Y | Y 2. Organizing Actions | Purpose is to control how email is managed, organized, stored and retrieved | 2.1 Keeps work-email in live inbox (vs. deleting or using system of files and folders) | The extent to which the inbox is used as the place where all incoming work-email is kept (rather than deleted, or moved to folders) | Y | Y 2.2 Regularly triages work-email | The extent to which a person triages their email to fit with current priorities by regularly reading, responding to and clearing messages, monitoring actions required and adjusting their own strategies and priorities as appropriate | Y | Y 2.3 Uses work-email for managing multiple project strands (compared to single strands or use of other methods) | The extent to which a person uses email to manage and negotiate multiple project strands | Y | Y 3. Crafting Actions | Purpose is to craft and construct work-email messages | 3.1 Sends email with high relevance/criticality to work | The extent to which a person sends email that focuses on work agendas, goals and priorities, as opposed to sending personal email or email that is not work-congruent to the receiver | Y | Y 3.3 Uses uncivil or inconsiderate actions towards others when sending work-email | The extent to which someone engages in uncivil or inconsiderate email actions, such as delegating tasks without negotiation, using an abrupt, rude tone, emailing in company (absent–presence or phubbing), putting pressure on people for an immediate response, sending ambiguous, lean or complicated content, inappropriate use of cc/bcc/reply-to-all | Y | Y 4. Communication Actions | Purpose is to prefer email as the communication channel in the context | 4.1 Uses work-email to connect with more diverse people | The extent to which a person regularly uses email to connect with higher numbers and diversity of people | Y | Y 4.2 Uses work-email for remote working | The extent to which someone chooses email to communicate with others in a remote working environment, compared with other people, activities and methods | Y | Y 4.3 Uses work-email at a proportionately high level | The extent to which a person demonstrates an overall high level of email use, compared to use by other people, activities and methods | Y | Y 4.4 Using work-email to communicate with team | The extent to which someone uses email to communicate with their team, compared with other people, activities and methods | Y | Y

We then took the list of email outcomes to create goal category codes. We focused on outcomes associated with the well-being or work-performance of the individual perpetrating the action. The majority of empirical findings relating to well-being were reported in terms of resource depletion or stress outcomes. Categories of ‘well-being’, therefore, primarily refer to negative indicators. Negative well-being indicators were sub-categorized as: (a) strain, burnout, pressure or exhaustion or – reverse direction – positive affect, coping and feeling in control; (b) work disengagement and withdrawal, or – reverse direction – work satisfaction, commitment and engagement9; (c) Work-life conflict and (d) overload (includes work, technology, email and information overload outcomes). Work-performance outcomes were given sub-categories relating to: (a) job/task output; (b) email performance and (c) enhanced work collaborations. It is worth reiterating here that whilst these categories are all ‘goals’ (or negative indicators of goals), when people apply an action, it is only at high levels of regulation that they are likely to be consciously aware of working towards these outcomes as ‘goals’. In the most part, actions are regulated at low levels and the aforementioned goal categories may be distal and sub-conscious (Sun & Frese, 2013).

After categorizing actions and goals, we returned to our spreadsheet to conduct a comprehensive review of the papers to determine the relationships between each action category and goal category. This analysis involved examining the key findings and results of the studies, considering statistical relationships in quantitative papers and reported/interpreted relationships in qualitative papers. Positive and negative relationships were identified, and in quantitative papers, non-significant relationships were also noted. Additionally, we took note of any boundary conditions or interaction effects that were involved in these relationships. Appendix S3 shows how these data were captured in a detailed table providing a paper reference for each relationship (e.g. [P5 10] for Belkin et al., 2020). Table 3 is a simplified version of this, showing the positive (+), negative (−) and (in quantitative papers) non-significant (NS) relationships between each work-email action category and goal, and the number of papers (in brackets) that evidenced each relationship. We provide an illustration of how the data from [P5] was coded (see Figure 1), and how this is represented in Appendix S3 and then Table 3, in the footnote.11

TABLE 3. Work-email action categories and relationships with outcome categories by number of papers.

Email action category | Negative well-being (resource depletion) | Work-performance

Strain, burnout, pressure or exhaustion (vs. positive affect, control and coping) | Disengagement and withdrawal (vs. satisfaction, engagement and commitment) | Work-life conflict | Overload (incl. work, technology, email and information) | Job/task outputs | Email performance | Enhanced work collaborations 1.1 Uses work-email out-of-work hours | + (7); − (9); NS (3) | − (2) | + (5); − (1); NS (1) | + (2); − (1); NS (2) | + (6); − (1) | + (3) | + (1) 1.2 Reactive or frequent checking and responding to work-email | + (8); − (7); NS (1) | + (1); − (1) | + (1); NS (1) | + (5); − (2); NS (3) | + (7); − (3); NS (1) | + (1); − (2) | + (4); − (1) 1.3 Allows continuous work-email notifications | + (2); − (3) | − (1) | + (2) | + (3); − (5) | + (2) | + (2) 1.4 Communicates and adheres to boundaries for work-email access times | − (6) | − (1) | − (1) | + (1) 2.1 Keeps work-email in live inbox (vs. deleting or using system of files and folders) | + (4); − (2); ns (1) | + (7); − (1); NS (3) | + (2); − (3) | − (5) | − (1) 2.2 Regularly triages work-email | − (8) | − (8) | + (4) | + (3) 2.3 Uses work-email for managing multiple project strands (compared to single strands or use of other methods) | + (1) | − (1) | + (2) | + (1); − (2) | − (1) | + (1) 3.1 Sends email with high relevance/criticality to work | − (1) | − (1) | − (3); NS (1) | + (9) | + (4) | + (1) 3.2 Uses uncivil or inconsiderate actions towards others when sending work-email | + (4) | + (1) | + (3) | − (7) | − (10) 4.1 Uses work-email to connect with more, diverse people | + (6); − (6); NS (2) | + (2); − (2) | + (5) | + (1); − (2) | + (10); − (1) | + (1); − (1) | + (10); − (2) 4.2 Uses work-email for remote working | + (1) | + (2) | + (1) 4.3 Uses work-email at a proportionately high level | + (13); − (4); NS (3) | + (2); − (2); NS (1) | + (1); NS (1) | + (13); NS (6) | + (12); − (4) | + (6); − (4) | + (10); − (3) 4.4 Uses work-email to communicate with team | + (1) | + (1) | + (3) | + (2); − (2) | + (4); − (1)

Because the work-email action and goal categories are not orthogonal, we do not consider our categorization of actions or goals to constitute a taxonomy (Nickerson et al., 2013). We note that, as per ART, goals can influence each other, and thus are not independent. For example, achieving work goals can often result in feelings of well-being (Barrick et al., 2013; Carver & Scheier, 1990; DeShon & Gillespie, 2005); equally, negative well-being (overload or work-life conflict) can negatively impact work-performance (Hobfoll et al., 2018).

FINDINGS OF THE SLR

An overview of the papers used in the SLR can be found in Appendix S4. These are referenced separately in the SLR References section of this manuscript. As seen, in Table 3, for some action categories, there was a high volume of papers providing empirical evidence to support relationships with goal outcomes (e.g. category 4.3), but for other categories far fewer papers provided support (e.g. category 4.2). The SLR approach allows for this, as there will always be trends in research that distort the availability of evidence (Rousseau et al., 2008). It is also worth acknowledging that action categories differ in how broad or discrete they are. For example, ‘2.1 Keeps work-email in a live inbox’ is a very discrete action pertaining to a singular behavioural expression, whereas ‘2.3 uses work-email for managing multiple project strands’ is broader and could encompass a range of different behaviours. In ART, actions and goals are arranged, enacted and pursued across different levels. Key to our research is the premise that, at whatever level the action is specified, the action's purpose is clear and is mapped to specific goal-related outcomes.12

As a result of the SLR, we found four themes relating to how multiple actions and goals dynamically interact to produce either effective or ineffective work-email activity. These four themes are now discussed below.

Theme 1: The categorization of actions and ‘super actions’

To our knowledge, within action regulation theory (ART), there has been no comprehensive classification of different types of actions (Hirschi et al., 2019; Zacher & Frese, 2018). This may be due to the specificity of actions in particular contexts, making generalization challenging. However, for our SLR, we needed clear criteria for defining actions in ART terms. We outline these criteria on page 14. Having established what an action is, the key aim of this research was to identify which actions positively predict both work-performance and well-being goals, as a proxy for ‘effectiveness’. Using Table 3, we identified which action categories (i) had a positive relationship with at least one goal outcome from the work-performance categories, (ii) additionally had a positive relationship with at least one goal outcome from the well-being (reversed resource depletion) categories and (iii) did not have any negative relationship reported in relation to work-performance nor well-being (reversed resource depletion).

We identified three action categories that met these criteria: ‘3.1 sends email with high relevance/criticality to work13’, ‘1.4 communicates and adheres to boundaries for work-email access times’ and ‘2.2 regularly triages work-email’. Additionally, one action category consistently showed a negative relationship with work-performance and well-being outcomes: ‘3.2 uses uncivil or inconsiderate actions towards others when sending work-email’. The reverse action, ‘uses civil or considerate actions towards others when sending work-email’, was associated with effectiveness. These four action categories represent overall effective work-email activity and are referred to as ‘super’ action categories in our framework, acknowledging ART's focus on ‘superworkers’ (Frese & Zapf, 1994).

To understand the theoretical basis for the ‘super’ effect of these action categories, we turned to ART. A relatively under-explored area in ART is how workers focus their regulation efforts. Zacher and Frese (2018) suggest that actions are regulated by attending to the task, social context and self. Effective workers consider these factors, potentially in a hierarchical order, to assess the progress and likelihood of goal success. The four super actions suggest the worker is able to address the requirements of the task, social context and self (see Appendix S5 for a detailed breakdown per action category). When these three foci are overlooked, actions tend to fall into non-super action categories. This appears to be because neglecting the needs of the self negatively affects well-being, overlooking the needs of the task impairs work-performance and ignoring the social context can negatively impact both well-being and work-performance. Thus, for any action to be considered ‘super’ in the context of work-email, it must be regulated with a focus on these three elements together. This supports Zacher and Frese (2018) and aligns with key premises of COR theory, which emphasizes building resources related to social, task, and self-considerations for effective functioning (Hobfoll et al., 1990). Further research is needed to explore if this combination of regulation foci extends to other contexts, informing the understanding of super actions in different domains.

Theme 2: Attaining work-performance and well-being goals in unison

Super actions are highly efficient because they encompass specific categories of actions that simultaneously fulfil well-being and work-performance goals, potentiating a state of thriving (Brown et al., 2017). It is acknowledged that individuals can engage in multiple action categories when using work-email, possibly to compensate for any negative effects caused by certain actions. For instance, if a goal is not attained (or even disadvantaged) by one action, another action might be adopted to rectify this. However, from the perspective of ART, using multiple actions to achieve goals is not behaviourally economical (Hockey, 1997, 2000; Schönpflug, 1986). Therefore, while a combination of work-email actions can lead to the attainment of well-being and work-performance goals, only super actions that singularly enable the achievement of both major goal categories in unison are considered effective.

The well-known paradox of work-email use is evident in our framework (Dawley & Anthony, 2003; Middleton & Cukier, 2006), as the same action category can have both positive and negative impacts on different goals. For example, while ‘1.1 uses work-email out-of-hours’ negatively affects well-being [P7], it positively influences task/job outputs [P45]. We found that these paradoxes tended to emerge across studies rather than within them. When papers examine how an action affects multiple goals within the same study, this paradox is largely absent.14 Particularly in studies employing mediation analysis, the effects of actions on multiple goals tend to follow the same direction (i.e. actions that negatively impact well-being in some categories also negatively affect other well-being or work-performance categories, and vice versa). This aligns with the traditional premise that actions taken to achieve one goal generally support the attainment of other goals (Locke & Latham, 1990) and reflects the resource gain or loss spirals described in COR theory (Hobfoll et al., 2018).

These findings suggest that when goals are examined independently in separate studies, an action that positively impacts one goal in one study may paradoxically have a negative impact on another goal in a different study. However, when the impact of an action on multiple goals is observed within the same study (e.g. through mediation analysis), such paradoxes are less likely. This insight contributes significantly to our understanding of work-email activity and how goal outcomes are measured in ART research. To further elucidate these dynamics, more complex research designs, such as contemporaneous mediation analysis, are needed to investigate the simultaneous or sequential achievement of joint goals in work activities.

Theme 3: The essential interplay between influential mechanisms in ART

It is evident that specific conditions and mechanisms play a crucial role in enabling effective action-goal relationships within work-email activities. Indeed, we found that conditions of execution, levels of regulation and principles of behavioural economy were involved in both the deployment of super actions, and in moderating super action–goal relationships. When an action category was not classed as super, this was due to a lack of consideration of at least one of the above mechanisms, whereby workers: (i) operated at inappropriate (usually low) levels of regulation, (ii) misapplied principles of behavioural economy and (iii) experienced conditions of execution that did not provide resources required to attain relevant goals. We elaborate further in the sections below.

Levels of regulation

The level of regulation that workers apply appears to influence action–goal relationships in relation to work-email. For example, the action category for ‘1.2 reactive or frequent checking and responding’ emerged as a result of people responding to incoming email with little conscious thought. This suggests that a low level of regulation can be beneficial in reducing overload [P15, P57] and engendering task accomplishment [P15, P26, P32, P47, P52, P31, P49]. However, several studies also indicated that the low level of regulation resulted in problematic outcomes across several goals [P3, P6, P9, P15, P35, P47, P38, P39, P59] as workers failed to identify changing needs in the situation. It seems that operating at low levels of regulation can be efficient for workers when dealing with usual day-to-day email exchanges – it keeps the inbox clear, reduces overload and gives a sense of task achievement. However, if this results in habitual or addictive behaviours, or means that people fail to consider changing conditions, this is problematic; hence, why action category 1.2 had paradoxical effects on outcomes of work-performance and well-being.

In comparison, when higher levels of regulation are applied, as per the super action category ‘2.2 regularly triages work-email’, universal benefits are clear. Within this category workers reported to be consciously and flexibly applying actions according to changing priorities; considering the best way to deal with each email, with positive outcomes for both work and well-being goals. Although some triaging actions may be automated over time and with familiarity [P26, P49], effective activity involves occasionally switching to a higher regulation level to proactively check action appropriateness and adapt as necessary [P28, P47]. Without doing so, workers are more likely to succumb to habitual, compulsive, addictive or inappropriate activity, which will negatively impact work and well-being. In summary, effective work-email activity involves being able to appropriately apply the correct level of regulation to deal with the changing demands of the situation. Our research indicates that to effect this, people must have control over their work-email [P54], and it is likely that workers with more resources/capacity in the first instance are also those who act at higher levels of regulation and adapt appropriately to changing demands in relation to work-email activity [e.g. P21, P24, P28, P50, P31, P49] (Hobfoll et al., 2018).

Behavioural economy

Behavioural economy also appears to have been applied in each of the four super action categories. For example, although ‘3.2 (reversed – see page 18-19) uses civil or considerate actions towards others when sending work-email’ may require more effort in terms of time and energy resources [P10, P29], it appears to be a worthwhile cost, as positive outcomes are consistently related to work-performance and well-being. The failure to consider behavioural economy appropriately can be seen in the non-super action category ‘1.1 uses work-email out-of-work hours’. In the most part, actions in this category have a positive impact on work-performance [P12, P11, P19, P32, P45, P57, P31]. but there are also clear negative repercussions for well-being [P4, P6, P7, P11, P13, P18, P35, P37, P43, P59], which can then impact performance negatively too [P11]. Similar contradictory goal outcomes are seen across the other non-super categories.

In summary then, effective work-email activity involves ensuring that behavioural economy principles apply to any course of action, so that neither well-being nor work-performance is unduly compromised. Being able to choose a course of action that benefits both work-performance and well-being necessitates that the worker has decision latitude and the resources at his/her disposal to make appropriate decisions about email action.

Conditions of execution

Each type of condition of execution, outlined by ART, appears to require consideration if actions are to successfully lead to the attainment of work-performance and well-being goals. The role of the four different conditions of execution, reported in the literature on work-email activity, is outlined now below.

Situational parameters

As ‘Work-demand’ situational parameters fluctuated (e.g. working under a deadline), super emailers changed the action that they deployed to deal with email interruptions [P47, P60]. Further, ‘Contextual’ parameters such as work cultures and group norms were important action prompts. Cultures that promoted an expeditious ‘pressure to respond’ norm directly influenced actions, such as ‘1.2 reactive or frequent checking and responding’ [P26, P35] or ‘3.2 uses uncivil or inconsiderate actions towards others’ [P58]. Cultures of ‘trust’ also directly influenced work-email actions. When trust is low within a work culture people engage in negative actions such as ‘3.2 uses uncivil or inconsiderate actions towards others’ [P6, P35]. If trust within a culture is high, then people engage in super actions such as ‘1.4 communicates and adheres to boundaries for work-email access times’ [P35]. Work culture also appeared to moderate or influence the relationship between work-email actions and outcomes. Those who engage in actions related to ‘4.3 uses work-email at a proportionately high level’ experience better outcomes if they also hail from ‘work critical’ email cultures [P19, P46, P57]. High ‘pressure to respond’ cultures generally exacerbated or mediated actions with negative outcomes [P6, P9, P45, P52].

Personal resources

COR suggests that people who have more resources are better able to build and protect their resources (Hobfoll et al., 2018). Across the papers reviewed, evidence consistently found that possessing personal resources relating to job-related capacity, action style/ personality and available skills or knowledge all positively influence action–goal relationships. First, when ‘Job-related capacity’ was lacking, when demands exceeded resources [P32, P46], there were impediments to goal progress [P3, P9, P43, P46, P47, P48]. This was offset when people reported greater capacity to cope [P57]. Second, in some studies, ‘Action styles and personality’ impacted action use directly. Conscientiousness negatively predicted ‘1.2 reactive or frequent checking and responding’ [P48], and personality also predicted actions leading to the attainment of trait-relevant goals [P49]. Interaction effects revealed that trait self-control having both an advantageous and disadvantageous effect on job/task progress and well-being [P46, P48]. Third, ‘Available skills and knowledge’ mattered as email training was found to positively impact choice of email super action categories [P10, P23]. Knowledge of email software and functionality was also found to influence action–outcome relationships [P6, P16, P19, P28, P54].

Object partners

The impact of technology on actions is evident in relation to portability of email and software capabilities. In relation to the former, many of the most recent papers (2015–2021) examined ‘1.1 Out-of-hours email activity’ [P4, P6, P7, P12, P11, P13, P18, P5, P31], with studies specifically focusing on how devices like smartphones facilitated this [P13, P21]. Engaging in more out-of-hours email activity detrimentally impacted well-being outcomes [P4, P11, P35, P37, P59] but often improved work-performance outcomes, because of the mobile, extendable quality of email [P12, P11, P32, P57, P19, P37]. In relation to software capabilities, as email infrastructure and software has evolved, research suggests that work-email activity is no longer considered to be an ‘add-on’ to other work tasks [P57, P45], but inherent to work and project management (category 2.3), networking (category 4.1) and organizing team activity (category 4.4) [P2, P14, P36, P16, P57]. As people use work-email more (category 4.3) they become more adept at it, using the functionalities of email software packages to improve email effectiveness [P24, P28, P62].

Activity partners

Effective engagement with activity partners was found in the super action category ‘3.1 sends email with high relevance/criticality to work’. Here, managers' work-performance and effectiveness improved because they modelled effective, critical, work-relevant email actions, which improved the way colleagues responded [P8, P43, P46, P20]. Consideration of one's activity partners is essential to effective work-email activity. When this is overlooked, and a lack of respect is shown to others' priorities, needs and goals, work relationships can suffer, negatively impacting task outputs and collaborations. Individual email activity does not exist in a vacuum, being mindful of how actions towards others can negatively impact personal goals needs to be considered (Zacher & Frese, 2018).

Theme 3 summary

Super actions are adopted, and outcomes of both well-being and work-performance are achieved, when conditions of execution are favourably inclined towards providing goal-relevant resources that allow for action-goal regulation at an appropriate level, and a behaviourally economic response. Thus, it is evident that the three influential mechanisms highlighted in ART matter in predicting effective work-email activity. Having the ‘right’ influential mechanisms in place is one part of the equation in facilitating effective multi-action, multi-goal pathways. However, in our synthesis of the 62 papers in the SLR, we further identified key regulatory processes that appear to encourage greater actions-goal alignment towards effective work-email activity. We outline these processes below.

Theme 4: Assessing priority parity and regulating errors in action-goal progress

By identifying that super actions meet multiple goals, our SLR revealed cognitive and behavioural considerations, in choosing and regulating effective activity. In the sections below, we outline examples of how ‘priority parity’ analysis and the ‘regulation of errors’ in action-goal progress are important considerations in our framework.

Priority parity

In our appraisal of the SLR papers, it appears that when people adopt actions that demonstrate high parity with (i) their own goals, (ii) those of others and (iii) the valued goals of their culture, then effective activity is more likely to ensue. At an individual level (i), when the action for dealing with work-email facilitates priority goals of the actor/user (e.g. building effective work collaborations), this also has positive repercussions for other outcomes (such as well-being [P37]). When the actions of the individual align with the goals of their team or activity partners (ii), then the action-goal pathway is better realized, presumably because others did not resist these actions, and worked with the actor to assist in goal attainment. For example, in the super action category of ‘3.1 sends email with high relevance/criticality to work’, other members of the team reported to appreciate such activity and responded more readily to work-critical email [P2, P57, P58]. Finally, when an actor's goals align with the values of the culture or climate (iii), they are more likely to adopt actions that will encourage effective outcomes, and vice versa. For example, ‘pressure to respond’ cultures deprioritize people's well-being goals, and can result in negative well-being outcomes for those who are acting to fit with the culture (e.g. ‘1.3 allows continuous work-email notifications’) even though this does not fit with personal preference [P48, P52].

Regulating errors in action

ART purports that a continuous feedback cycle involves an actor testing if the current action is meeting desired outcomes, by building a mental representation of success towards goal progress (Frese & Zapf, 1994; Zacher & Frese, 2018). ART acknowledges that people engage in flawed behaviours and that this is not necessarily a hindrance to success if people recognize these flaws and make plans to correct them. In correcting errors, workers achieve a sense of motivation and satisfaction, which builds their resources (Zacher & Frese, 2018). We found evidence of actors engaging in regular action regulation checks across the SLR studies, especially in longitudinal or qualitative papers. It was reported that actions changed in order to better facilitate goals when progress is stymied [P19, P25, P38, P47, P49]. More longitudinal studies, and greater use of mediation analysis, are needed in future research to more directly test how attending to errors can have long-term benefits in regulating activity.

A multi-action, multi-goal framework of effective work-email activity

Figure 3 presents our multi-action–multi goal framework of effective work-email activity. The figure places super work-email actions at its core (Box B) as actions that led to the achievement of both work-performance and well-being goals – indicative of effective activity. The figure indicates how ART mechanisms (Box A) impact the use of super work-email actions and moderate the relationship between actions and outcomes (Boxes C and D).

Details are in the caption following the image

FIGURE 3

Open in figure viewer __PowerPoint

A multi-action, multi-goal framework of effective work-email activity.

DISCUSSION, CONCLUSIONS AND FUTURE RESEARCH

The purpose of this paper was to organize the cross-disciplinary extant research literature on work-email activity and provide a comprehensive framework of what constitutes effective work-email activity when multiple actions and goals are considered. We identified four super categories of action that predict effective outcomes, along with the specific influential mechanisms that drive and influence these. We hope that the production of this framework will allow researchers, organizations and end-users to understand and predict what actions will result in effective work-email activity. In the sections below, we outline the main theoretical and practical contributions of this research.

Theoretical contributions

Our first contribution is the development of a comprehensive framework that explains and predicts effective work-email activity by considering multiple actions and goals simultaneously. We stipulated that well-being and work-performance goals need to be achieved in unison for activity to be deemed effective, and believe that ours is the first framework of complete activity to examine both goals together (addressing calls from Hirschi et al., 2019; Sun & Frese, 2013). By adopting a multi-action, multi-goal approach, we observe that achieving multiple goals concurrently has positive cascading effects on each other, suggesting a synergistic relationship between work-performance and well-being goals. Interestingly, contradictions in goal attainment were rarely observed when studies examined contemporaneous multi-goal pursuit (e.g. in mediation studies). This highlights the importance of considering both work-performance and well-being goals within a holistic activity cycle to understand effectiveness at work and address the paradoxes of email use.

Our second contribution builds upon action regulation theory (ART) by establishing clear criteria for categorizing “actions” within the context of work activity. This allowed us to define “super” actions, which simultaneously fulfil work-performance and well-being goals across categories. We identified four categories of super actions that predict effective work-email activity and identified the influential mechanisms that facilitate their engagement. Super actions are optimized when individuals have access to goal-relevant resources, regulate their activity at an appropriate level, adapt to changing situational parameters and employ behaviourally economic strategies. This enables them to satisfy the needs of the self, task and social context. Our research provides insights into the nature of super email actions and how their engagement can be promoted.

Practical contributions and future pathways

To address the lack of clarity in the field of email research and counteract the prevalence of unsubstantiated advice found in popular outlets, one of the key objectives of this study was to provide evidence-based recommendations to managers, policymakers and end-users on how to optimize work-email use. Our third contribution is the development of a framework that enables researchers and practitioners to understand and predict which actions are associated with effective work-email use, thus facilitating the generation of research-driven recommendations. The range of recommendations derived from our framework places a specific emphasis on the four categories of super actions. These recommendations, along with their rationale, are presented in Table 4. Wherever possible, we have incorporated practical suggestions based on the advice provided by authors cited in the bibliography of the SLR (papers referring to these suggestions are included in squared brackets).

TABLE 4. Recommendations to facilitate use of ‘super actions’ and optimize effective work-email activity.

Super action | Related recommendation and rationale

1.4 Communicates and adheres to boundaries for work-email access | Use automatic replies, shared inboxes and signatures to communicate access

Use ‘delay send’ to communicate with colleagues out-of-hours

Organizations to remove expectations that put pressure on people to respond to work-email out of hours

2.2 Regularly triages work-email | Regularly check and deal with email

Turn off notifications

Provide workload allocations for dealing with work-email

3.1 Sends email with high relevance/criticality to work | Offer evidence-based email training

Use email for work-purposes only

Provide extra email workload allocation

3.2 Uses civil or considerate actions towards others when sending work-email | Be clear in content

Develop email guidance

It is important to emphasize that the recommendations in Table 4 should be regarded as a starting point for practitioners and researchers. Whilst these evidence-based recommendations are supported by studies referenced in the SLR, providing a solid foundation for our suggestions, future studies can test these recommendations by implementing interventions aimed at training workers in optimal work-email use. We recommend granting workers sufficient time to implement and practise new work initiatives. Additionally, it should be acknowledged that these recommendations may be challenging to apply when workers lack control or decision latitude. Therefore, we encourage organizations and job designers to foster an environment that promotes complete activity at work whenever feasible.

Furthermore, it is worth noting that our research focused primarily on actions related to individual actors/users and their goals and outcomes. However, the studies identified in the SLR indicated that an individual's actions can also impact the work-performance and well-being outcomes of those with whom they interact. There is now an opportunity for future research to expand the framework and explore crossover effects on activity partners. Extending the framework to encompass the broader context of teams, organizations and society will provide insights into the effectiveness of collective work-email activity and shed light on when, why and how it can be optimized.

Limitations and future directions

Across the papers reviewed in the SLR, we identified five key features that posed challenges in terms of operationalization, understanding of actions and study design focus. First, there was a lack of longitudinal and causal designs, particularly in earlier papers. This limited our ability to establish predictive relationships between actions and outcomes. We emphasize the need for studies that examine causation over different time points to demonstrate the causal and potentially reciprocal relationships between actions, outcomes and their precursors.

Second, while some studies utilized objective measures such as physiological responses or email logs, there was an over-reliance on self-report data in the empirical studies. This reliance on subjective measures may overlook important predictors such as objective workload or email volume, potentially confounded by attitudes towards email. We recommend future research to incorporate a balanced combination of subjective and objective measures to provide a comprehensive understanding.

Third, there was a lack of consistency in the operationalization of variables across studies, particularly concerning the concept of “overload.” Different studies measured overload in various ways, making it challenging to establish equivalence. However, in all cases, overload refers to a depletion of resources that negatively impacts well-being. Researchers should carefully select the most appropriate measure of overload for their specific research context and clearly specify the form of overload being examined.

Fourth, there were problems with specificity. For example, researchers examining out-of-hours working often failed to specify whether this involved digital communications, and/or did not specify communication mode. As such, many papers were rejected as we could not isolate work-email as the digital work tool being used. This is important; when researchers separate out modes of working, evident differences are found regarding impact on workers [e.g. P22, P39, P11, P13]. All of the papers that we included here specifically isolated work-email mode, and researchers would be encouraged to follow this practice.

Last, when studying a phenomenon like work-email that has evolved across different technology epochs, researchers should report the timeframe and system versions of their data collection. Unfortunately, only a limited number of studies in our SLR included such information. Reporting the time of data collection and the email system used by participants is essential for understanding the conditions of execution. Future studies should include this information to enhance the interpretability and relevance of their findings.

Despite these limitations, we observed an upward trend in the average quality rating of papers investigating work-email activity. The number of papers meeting inclusion and exclusion criteria has increased over the years, indicating growing interest and rigour in this field. We remain optimistic that with our recommendations and the continuous improvement in research quality, understanding of work-email activity will continue to advance and provide valuable insights into this important work phenomenon.

A note about using SLR approaches

The SLR approach is a robust and transparent process that avoids the possibility of researchers' cherry-picking studies according to bias or personal agenda. However, its success and comprehensiveness rest on the choice of search terms and databases in the first instance. To reduce bias and subjectivity, consulting a steering committee of subject experts is always recommended. Yet, we acknowledge that in our research a committee comprising different members might have made different selections. The advantage to using a SLR protocol (Appendix S1; registered at osf.io/wzqke) is that researchers can identify exactly what decisions were made, thus providing clear boundaries and scope to any project, and allowing future researchers to extend or develop this according to their own research priorities.

Our SLR approach did not prioritize some (the better evidenced) results over others. Whilst techniques such as meta-analysis do this, the SLR potentially allows for a broader opportunity to theorize, as researchers are less likely to be limited by ‘trends’ in research (e.g. that favour publication of papers relating to some topics over others), and can also give qualitative data equity alongside quantitative data, in producing conjecture (Rousseau et al., 2008). We encourage researchers to be mindful of Table 3, where we specify the number of papers returned to support the different action–goal relationships. To develop understanding of effective work-email activity further, we suggest that a dearth of papers could create opportunity for researchers to explore such areas in the future.

Final reflections

This paper brings together 25 years of cross-disciplinary research on work-email to develop a multi-action, multi-goal framework of effective work-email activity. By applying ART and COR principles, our framework allows us to explain and predict how actions related to work-email use will impact work-performance and well-being outcomes. We identified four super action categories that consistently predicted both work-performance and well-being. Super actions are supported by a specific set of conditions that optimize resources and activity in relation to the task, the self and the social context. Non-super actions had differentially positive and negative outcomes on work-performance and well-being but did not predict both in unison. This serves to remind us that when practitioners and policymakers implement guidance to promote a particular work-email action (e.g. banning access out of work hours) this can be beneficial for some outcomes (e.g. reducing work-life conflict and overload) but could be detrimental to other outcomes (e.g. work effectiveness or promoting effective work collaborations). Our framework allows scholars and practitioners to predict whether any action for dealing with work-email is likely to be ‘effective’ (i.e. ‘super’) or not. We, therefore, encourage organizations, policymakers and end-users to consider the likely outcome of any work-email action- by using our framework.

AUTHOR CONTRIBUTIONS

Emma Russell: Conceptualization; data curation; formal analysis; investigation; methodology; project administration; resources; supervision; validation; visualization; writing – original draft; writing – review and editing. Thomas W. Jackson: Conceptualization; formal analysis; investigation; resources; visualization; writing – original draft; writing – review and editing. Marc Fullman: Data curation; formal analysis; investigation; project administration; software; writing – original draft. Petros Chamakiotis: Formal analysis; validation; writing – original draft; writing – review and editing.

CONFLICT OF INTEREST STATEMENT

We have no known conflict of interest to disclose.

DATA AVAILABILITY STATEMENT

Data can be made available on reasonable request to the authors. On publication, data will be uploaded to the project registration site at the Open Science Framework (osf.io/wzqke).

Supporting Information __

REFERENCES __

Systematic Literature Review References __

Citing Literature __

Volume97, Issue1

March 2024

Pages 74-103