Digitalization and academic research: knowing of and using digital services and software to develop scientific papers Open Access

Scientific and technological advancement – a critical driving force behind economic growth and societal wellbeing (e.g. Jin and Jin, 2013; Odhiambo and Ntenga, 2016) – is the result of scientific research across various disciplinary sectors. Higher education institutions worldwide, such as research centers and universities, play their part, contributing via publications and multi/interdisciplinary partnerships to innovate in favor of – and satisfy – the needs of society.

With specific reference to public universities, scientific research quality and productivity are fundamental goals for each university in the today's academic environment (Edvardsen et al., 2017), in which resources are limited and competition is fierce. Different indicators are considered in international measurement systems to achieve impactful levels of research (quality) and a large number of publications (productivity) in both individual and department/faculty terms (e.g. Rhaiem, 2017; De Witte and López-Torres, 2017). Accordingly, to write and publish are key research-related activities for academics (Seiler et al., 2011), to which are added dissemination of such scholarly knowledge to society (i.e. “third mission” activity; Davey, 2017; Knudsen et al., 2021) and teaching (Giraud and Saulpic, 2019).

Digital transformation fits into this scenario by favoring scientific research in terms of research quality and productivity with reference to scientific projects and products. Existing literature focuses on project rather than product management, by highlighting the importance of equipping universities with resources, especially in terms of internal infrastructure or external digital services to assist their researchers (e.g. Russell-Simmons et al., 2016; Al-Maadeed et al., 2021). Several universities worldwide are also investing in alternative avenues, to help their (especially nonnative-English speaking) researchers develop and improve their project writing skills (Stephens and Campbell, 1995) – an aspect that is particularly challenging to improve (Henson, 2004; Porter, 2007). Some universities are internally creating departments or providing training courses/workshops or externally developing partnerships to offer their scholars a wide range of editorial, educational and publishing services. Past studies in project management have focused on the development of co-authorships across (multi/interdisciplinary) academics (Brocke and Lippe, 2015; Choudhury and Uddin, 2018), by outlining how universities are creating opportunities to build relationships between and among researchers in developed, emerging and developing economies (Sankaran et al., 2021). Although a publication is the last phase of a research project – and a very important one, given that ultimately, this is what provides research quality and productivity outcomes – very little attention has been paid to research product management, which is mostly examined in terms of the reviewing process (e.g. Gilmore et al., 2006). Digital transformation in support of research activities consists of the availability of software packages and/or solutions installed in the user's computer and services accessible mainly via web (online services); when such services are provided by a specialized organization – on a free, freemium or, particularly, paid-basis – we refer to as digital professional services. No empirical research to date has examined the ongoing development of digitalization in support of the research publication process. The importance of this topic, and the lack of related studies, implies it requires further research.

Following a qualitative approach, this explorative study aims to examine how digitalization affects the academic research publication process, by taking into account the perspective of management scholars. Our goal is fourfold: (1) to provide an overview of digital professional services dedicated to academic research, (2) to investigate academics' awareness of digital services and software, (3) to study the impact of these digital tools on the scientific publication process, and (4) to examine scholars' expectations regarding their future use.

The remainder of this paper is structured as follows. After presenting the theoretical background in terms of the research publication process, digitalization, and expectations, the research method is described, and results are proposed and discussed. Implications, limitations, and future lines for research then conclude the study.

As an integral part of academic life, scholars are encouraged to write and publish (Gilmore et al., 2006), and especially, to publish in top-ranked journals (Shchemeleva, 2021) – that is, journals that have a high impact factor or H-index, or three and four stars as major criteria of research output (Kuteeva and McGrath, 2014) – and, consequently, to be included in the most prestigious international bibliographic databases (e.g. Gralka et al., 2019; Shchemeleva, 2021). Adding to the pressure to publish in top journals for not English-natives is anxiety regarding writing in English instead of the local language and having to account for non-Anglophone contexts (Curry and Lillis, 2014; Plo and Péres-Llantada, 2015). This can result in anxiety so acute, academic writing is described as “a minefield, a roller coaster ride, or at least an obstacle course” (Leki, 2003, p. 136).

At the beginning of the 1900s, writing an academic paper was similar to writing an essay or prose with no structure. After the Second World War, specialization in different fields, especially scientific ones, brought more standardized papers, which must now follow the so-called IMRAD structure – that is, Introduction, Materials and Methods, Results, and Discussion (Wu, 2011). Nowadays, writing an academic paper is not like writing an essay (e.g. West, 1992); such activity requires much time (Aguinis and Vaschetto, 2011), commitment, and various aspects of research expertise (Neuhauser et al., 2000) during the whole of the publication process (e.g. Caniato et al., 2018; Adams, 2014). Some studies (e.g. Troutt, 1998) have emphasized the importance of quality improvement principles with reference to the scientific article writing process, especially by offering advice to PhD students who want to publish their research in peer-reviewed academic journals (Paltridge and Starfield, 2016).

The research publication process follows a complex and structured pattern (e.g. Levitt et al., 2018) that incorporates several typical steps (Bordens and Abbott, 2002; Hair et al., 2007), from planning the scope and content to final manuscript production (Perry et al., 2003). During this process, each author develops an individual approach to planning, preparing, writing, and submitting articles to an academic journal (Gilmore et al., 2006).

Before the advent of computers and the internet, academics used to submit typewritten manuscripts through snail mail to journals, facing all kinds of difficulties from corrections, quantitative calculations, linguistic translation, and proofreading (e.g. Turk and Björk, 2008). With the rise of computers, new specific pieces of software have been developed, and starting from the 1990s, web-based services have also thrived (Humphries, 2007). Nowadays, digitalization is widely integrated into scholars' professional life, many of whom began to use technology for its ease of use and their need to conduct research supported by information and communication technologies. As Davis et al. (1989, 1992) argued, in one of the various extensions of the technology acceptance model (TAM) – a model derived from the psychology-based theory of reasonable action and theory of planned behavior – human behavior toward potential acceptance or rejection of a technology depends on perceived ease of use, perceived usefulness, and behavioral intention. Actually, several software solutions and online services are helping researchers in many ways, from advanced proofreading, content enhancement, dataset provision, and statistical analysis to post-publication marketing. However, to the best of the authors' knowledge, no previous studies have examined which technologies (digital research tools and professional services) authors use for each phase of the research writing and publishing process.

Writing and publishing a scientific paper, hence, is a task that can fit well – at least to some degree – with online digital services and software. More, given the rapid development of Artificial Intelligence (AI) technology across many fields (Dwivedi et al., 2021; Raisch and Krakowski, 2021), a further, even more provocative, question arises: could online digital services and software based on AI conduct research and write an entire academic paper? The answer from the academic literature is (for the moment) apparently negative, however some scholars have opened a debate especially on the Internet. For instance, the Massachusetts Institute of Technology (2019) claimed how a team of MIT scientists had developed a neural network, which could perform some work of a science writer. For a start, this system can read specialized journal papers and understand how to gather all the data and information needed for readership, even making the language more straightforward for anyone. The system can analyze many papers, dramatically reducing the time to review all of them and provide a preliminary answer to possible topics. The MIT team had some difficulties with the neural network in correlating a long string of data, but it seems this system could ease the initial stage of choosing a topic, writing a simple summary and introduction of the paper, and performing the literature review. Similarly, McCook (2017) claimed that there are already pieces of AI software able to generate a first draft ready to be revised by the researcher. However, these pieces of software are not capable of writing the discussion section so far, which is considered the most important and original part of a paper. Mindzak (2020) discussed the pros and cons of using AI for writing academic papers. The author claimed that we are not sure of what AI will be able to do for the research field; regardless, there are already applications and systems heading in this direction. For instance, in 2005, another MIT team developed an algorithmic language generator that wrote some papers accepted by some predatory journals. According to the author, this could lead to an automation of publication and problems related to plagiarism, originality, and academic ethics.

In a “publish or perish” era – where a scholar's quest for and pursuit of productivity is continually exacerbated by intense international competition between academic institutions – online digital services and software could strongly affect research quality and productivity by helping obtain more research outputs and maximizing the metrics used to measure the impact of a research product. Although expertise in scientific article writing and academic publishing comes with practice, online digital services and software could enable authors to dramatically increase their efficiency and ease of research (Rubin, 2003). Specifically, they could help the researcher improve the research and its outputs, relieving the author of repetitive and – to some extent – non-value-added (but crucial for a successful publication) activities such as corrections, finding sources, reviewing and comparing literature, finding the right journal, collecting data, and performing calculations. On the other hand, all these advanced services could introduce issues that could affect the entire scientific endeavor. For instance, a dramatic increase in research productivity could result in the mass production of similar papers, only to maximize the metrics used to measure the impact of a research product.

To the best of the authors' knowledge, the background dedicated to digital services and software is mainly comprised of observation of increasing tools and services in the market rather than academic literature. Even if these services are broadly used, it is surprising that there is such scarce investigation. The lack of research on these digital tools and, especially, on scholars' awareness, impact and expectations regarding these services suggest a need for further research. Before filling this gap, in the next section, this study summarizes the expectations concept, to support the specific analysis approach of this research.

The term expectations indicates customer wants. Wants, in this sense, differ from needs, as several studies in service management literature have argued (Parasuraman et al., 1985). In particular, expectations are conscious and accessible (Schneider and Bowen, 1995) while needs reside deeply in the human psyche, and are unconscious and global. Accordingly, needs underlie expectations (Chiu and Lin, 2004).

Many studies have proposed different classifications of expectations (Boulding et al., 1993; Walker, 1995; Hubbert et al., 1995; Lee et al., 2000; Ojasalo, 2001), which, essentially, highlight their multidimensional nature. The multi-expectation framework (Zeithaml et al., 1993), for instance, distinguishes between what customers consider acceptable (i.e. the minimum level of service performance that they can receive without being disappointed; adequate service level), and what they want or hope to receive, or what they believe can and should be provided to them according their personal needs (desired service level). A tolerance zone develops between these two levels of expectation. When the quality of a service falls below the adequate level, customers are dissatisfied; on the contrary, if it increases beyond the desired service level, customers are more than satisfied, and may in fact be “delighted.” This concept is well studied in both TQM and the service quality sector. For instance, Hiles (1994) was one of the first authors who examined the importance of service quality measurement using effective quality metrics and credible service quality monitoring as key factors for success. Mal Kong and Muthusamy (2011) studied the relationship between the perceptions of service performance and the service gaps in private higher education institutions. They found that the relationship can be used to map the quality attributes into four quality factors, namely satisfier, critical, dissatisfier, and neutral. Rönnbäck and Witell (2008) and Talib (2013) explored TQM concepts applied to service systems. They claimed there is an increased focus on the implementation of TQM principles in service organizations and especially in delivering high-quality service to customers. This can be achieved even using mechanization and automation at every organization level.

To meet customers' expectations, service providers “should offer” rather than “would like to offer” something to their customers (Parasuraman et al., 1988, p. 17). This study aims to provide practical recommendations to service providers of digital services and software to satisfy scholars' expectations in using these digital tools for their research.

In line with the exploratory nature of this study, we follow a qualitative approach (Creswell and Creswell, 2018) by making direct observations and conducting in-depth semi-structured interviews. Direct observations were employed to provide an overview of digital professional services dedicated to academic research. The interviews were organized with Italian and international management scholars to investigate academics' awareness of digital services and software, study their impact on the scientific publication process, and examine scholars' expectations regarding the future use of these digital tools.

Previous literature has considered both these methods suitable for conducting explorative research. Indeed, direct observation is a non-intrusive qualitative method that allows the researcher to understand a new phenomenon (Pantano and Vannucci, 2019; Grove and Fisk, 1992), while interviews provide the researcher with the opportunity to obtain a wide range of ideas and impressions regarding the subject by understanding individuals' perceptions and attitudes toward a phenomenon (e.g. Krueger and Casey, 2000). Further, interviews provide participants with the freedom and flexibility to communicate their ideas and beliefs in their own words and use storytelling to converse in a relaxed way (Creswell and Creswell, 2018). During the data analysis, an inductive approach was used (Saldaña, 2015).

This study adopted both purposeful and snowball sampling techniques (Miles et al., 2014) to recruit knowledge-rich informants and capture their experiences with reference to digital professional services and research tools. These sampling techniques enable researchers (authors) to group participants according to preselected criteria. Based on the research purpose, this study has chosen the five following criteria: (1) high annual research productivity, (2) high research quality in terms of publications' impact, (3) different seniority of work, (4), to be digital technology user in (each of) the stages of the scientific publication process, and (5) academics who are operating (retired scholars excluded) in international research teams. Given that digitalization affects the research publication process especially of younger academics, we arbitrarily involved a greater number of researchers between 30 and 40 years in the composition of the sample. The remainder of the sample was distributed between the other ranges of age, with a slight prevalence of age from 40 to 50 years Table 1 presents the profile of the respondents.

A saturation criterion was used to stop interviews; more precisely, the researchers continued to interview additional participants until saturation of interest field (i.e. until nothing new was being heard).

The Italian and international scholars that participated in this study were contacted via email to ask about their availability for an interview. The email contained a cover letter that clearly described the study's purpose. No financial incentives were offered for participation. Data confidentiality and participants' anonymity were guaranteed.

A list of possible digital professional services available for academics was derived by direct observation. In particular, the research identified 23 websites for digital professional research services through the Google search engine. To find these services, we used keywords such as “editing services,” “manuscript editing services,” “scientific manuscript writing services,” “proofreading services,” and “publication support services.”

Data were collected through researcher (author) observation of each website. The researcher observed the available digital services directly to understand their functions and possible benefits to academics. To ensure the correct recording data and limit collection bias, the researcher observed each website drawing on a research protocol based on website name, nationality, academic or non-academic source, type of service offered, main features, and stage of the research process. Observations were made between March and November 2021, with each observation lasting 20 minutes. Data for each website were systematically tabulated through an Excel file that further allowed a comparison among the websites. Finally, to assure validity and reliability of data collection, the dataset was shared with two independent researchers, who validated the research protocol and data collection procedure.

Based on participants' availability, appointments were set for face-to-face semi-structured in-depth interviews with each respondent to obtain detailed information on their experiences regarding the digital research tools and professional services that they were using. The scholars were interviewed face-to-face or via telematics (Microsoft Teams or Zoom platform). Interviews were conducted between September and December 2021, and lasted between half an hour and one hour. During data collection, the informants were encouraged to provide more detail about the research technologies used during their professional life.

The interview protocol was designed based on the literature review. Before data collection, a pretest – based on open-ended conversations with three scholars – was undertaken to ensure that the meaning of the questions was clear and that the wording was unambiguous. No changes were suggested, and the designed protocol was submitted to the identified scholars. Briefly, interviewees were asked to present the online digital services and software that they know and use, their needs and expectations regarding these tools, and their effects on the scientific production process in terms of quality and productivity.

The interviews were audio-recorded and later transcribed and checked for accuracy for data analysis. Data were analyzed using thematic analysis through NVivo 11 software (Braun and Clarke, 2006). Following a comparison and discussion of coding generated separately by the researchers, a consensus on common codes was reached. The codes were clustered in more general analytical themes for the analysis.

This section presents how digitalization affects the publication process from the perspective of management scholars, by highlighting the following: (1) digital professional services dedicated to academic research, (2) academics' awareness of digital services and software, (3) impact of these digital tools on the scientific publication process, and (4) scholars' expectations regarding the future use of digital services and software.

Based on direct observations of the digital professional services, a range of essential services are available to support scholars in their research work. Table 2 summarizes these services and their main characteristics, in addition to presenting the four phases of the research process (researching, writing, submission and review, and publication) and the activities to develop a scientific paper, which emerged from the joint analysis of the direct observations and the literature review. In relation to the research phase, multiple activities have to be undertaken by scholars to develop a scientific paper. In this regard, different digital professional services are offered on the market such as a service to carry out surveys independently, to obtain (reliable and tested) datasets, mailing list samples, surveys, and complete statistical analysis from the professionals. Some digital professional services select literature dedicated to the topic and even write some sections of the paper.

With reference to the writing phase, services offered again enable authors to operate independently – using websites to automatic real-time corrections – or with the support of professionals who check the manuscripts' technical or other rigor, to make it more readable, consistent, and academic.

With regard to the submission and review phase, digital professional services help the authors both to submit their manuscript according to the journal guidelines and support them in re-submitting their research work according to reviewers' comments.

Finally, publication services support the authors in the marketing of the published paper to increase scientific performance indicators (e.g. number of citations and h-index).

All the respondents communicated that they are aware of multiple online digital services and software packages for academic research. However, they admitted that they do not know all the services offered on the market, but rather, only a limited section. This cognitive limitation is fundamentally related to the personal habit of using specific digital tools. Precisely, most interviewees said that they frequently and systematically use the same digital services and software for their research. Fewer interviewees do not use these services and software; they have only heard that these services exist or have seen them used by other colleagues. Therefore, their knowledge is purely abstract, without any technical expertise. In addition, a generational difference among interviewees emerged as a further cognitive limitation, as shown in the following two interview excerpts:

Although the interviewees revealed an openness to improving their knowledge of these digital tools, difficulties emerged in the interviews. Specifically, scholars said that they only occasionally discovered the existence of digital services, as this interviewee highlighted:

The typologies of online digital services and software that scholars use for their studies are multiple. As Table 3 shows, the digital tools that emerged from the interviews are mainly used during the research phase to develop the theoretical framework and literature review, and to collect and examine the data in the empirical phase. Digital linguistic services and software, especially for nonnative-English speakers, are widely used during the writing phase, as this interviewee noted:

A few scholars reported contracting surveys to external specialized digital services and purchasing data, especially data aimed at a specific research goal, such as profile customers to administer questionnaires or organize experiments.

Manuscript submission as well as copyediting and proofing of papers are usually used through digital services offered by the single publisher.

After publishing their research, few scholars invested their time or effort in disseminating research results through social media or digital services, or to developing relationships with other researchers in the same research area.

All the interviewees argued that digital services and software are beneficial in supporting scientific research work; in some cases, also preventing human error. They considered these digital tools as a means, not as an end, to conduct research, as this respondent said:

These digital tools enable scholars to improve the quality and to increase the number of research outputs, as these interviewees noted:

The quality/quantity relationship was at the center of many of the interviews. Different opinions emerged, which can be summarized in terms of the importance of researchers' abilities rather than digital tools. Two views in this regard were proposed:

In addition to improved research quality and increased productivity, a further positive impact, and also negative effects, of using digital services and software for academic research emerged from the analysis (see Table 4).

Online digital services and software have not radically changed scholars' research approach because the reasoning, mental path, research setting, and evaluation of the results have not been modified. They have, however, changed habits applied in drafting papers and sometimes even scheduling research activities, as these scholars affirmed:

A difference for this topic emerged in terms of generational aspects. Younger researchers who started working in academia more recently found that digital services were widely diffused, and could not say if they had affected their research as they had always used them and learned to use them during their PhDs.

The interviewees believed that there would be good opportunities for digital software and services development in the future. They believed that these tools would increase in number and use, especially across younger researchers. Expectations regarding the future amount of digital tools were conflicting – some expected growth, while others predicted convergence toward a few packages used by many, instead of plentiful but smaller software packages. According to some, software such as R would take over because it allows an excellent level of customization through the programming language and co-creation, requiring a lower cost (some packages are even free).

Regardless of this, the positive development of these digital tools fuels multiple expectations that can be distinguished into the following two categories: (1) the improvement of digital services features and functionalities, and (2) the overcoming of specific difficulties faced by scholars in carrying out their studies. Referring to the first, not all respondents contributed to this topic because of their limited awareness of the use of digital tools. Scholars who replied expected that digital services and software would become more flexible, user-friendly, intuitive, integrated, customizable, have more automated procedures, and be interconnected with other – not just academic but managerial – software. Some of the interviewees noted specific expectations, for example, to refine the search for academic papers:

In addition, digital services and software could increase/improve their functionalities to detect topics relevant for the academic literature and not discussed in other papers, to identify similarities or differences in comparison with content already included in other papers, and to collect specific topics or theories. All this would help scholars define a framework via which to understand topics not yet encountered and any research gaps, as this scholar stated:

In relation to the end of the publication process, although some online copyediting services exist, some desired a more significant presence and efficiency in the online channel. Further, some suggested that a forum grouping scientists may be helpful to collect ideas and prevent research rejection (i.e. gathering suggestions and feedback).

Finally, scholars expected that academic journals and related publishers should improve their offering of digital services for managing sources, previous papers, and editing/formatting of papers both in the paper development phase and in the proof editing following acceptance of the manuscript. Some scholars further desired that automated services for grammar correction (such as Grammarly) improve their service for academic writing to reduce the “human” revision work done by proofreaders.

Concerning the overcoming of specific difficulties faced by scholars, some outlined how administrative procedures – e.g. payments and accounting procedures – that are needed to purchase online services slow down the development of these services because of non-alignment with current guidelines of universities. This mainly occurs when service providers are foreign companies, as this involves greater complexity in terms of communication and administrative agreements across universities. In addition to expecting these obstacles would be overcome, scholars highlighted their wish that service companies provide digital services and software at lower prices. There are universities and research centers with significant financial resources, but there are cases of researchers forced to buy these services and software at their own expense. As one respondent put it:

During the interviews, some scholars emphasized that in the future, an increasingly heavy intervention of machine learning and all related forms of artificial intelligence may increase, as these testimonials claimed:

However, it was commonly agreed that scholars' skills are important, as these respondents argued:

Findings are summarized in Table 5 

This study explored how digitalization affects the academic publication process by taking into account the perspective of management scholars. Overall, digitalization emerged as pervasive throughout the entire scientific research publication process, given that it can be used in each phase of this process. This study, thus, confirms that digitalization helps authors overcome the different kinds of difficulties inherent in publishing a scientific paper (e.g. Turk and Björk, 2008) and, at the same time, favors the authors themselves in each phase of the publication process (Rubin, 2003).

In relation to the first research objective (i.e. to provide an overview of the digital professional services dedicated to academic research), some already well-known and taken-for-granted in daily practice emerged, such as translations and grammar checking, proofreading, improvement of a paper's structure and readability, and sample generation and data collection and analysis services. Further, a growing offering of services, resulting in de facto outsourcing of some core research activities, such as data analysis, data presentation, and report generation, can now also be found. In particular, Table 3 highlights how some of the digital professional services can be considered advanced, simulating the peer-review process by suggesting the kind of journal for submission and revising the paper according to reviewers' comments. Therefore, we consider these services far from basic corrections or proofreading, and they may ultimately substitute for the core activities of research.

With reference to the investigation of academics' awareness (the second objective), this study highlights that the scholars interviewed did not know all the professional digital services and software that could help them in developing a scientific paper. As Table 4 shows, they mainly used digital tools with regard to the following two phases of the research process: (1) research to develop a literature review, data collection, and data analysis, and (2) writing to manage the manuscript's references, translation, and proofreading. This last service is particularly adopted by scholars that are nonnative-English speakers to reduce their anxiety in writing in English and to make their manuscripts more reliable, as argued in literature (Leki, 2003; Curry and Lillis, 2014; Plo and Péres-Llantada, 2015). Essentially, this analysis outlines that digital services and software are not used in relation to the entire research process not because they are not useful, but because they are unknown.

Regarding the impact of these digital tools on the publication process (the third objective of this research), the findings show that these technologies cover practically all phases of the research process, and that they have a profound effect on it – greater in some aspects than others, but certainly not neutral. In line with previous studies, this explorative research confirms that scholars have benefited from these digital tools to improve research quality and increase their productivity (e.g. De Witte and López-Torres, 2017; Edvardsen et al., 2017; Rhaiem, 2017). Fundamentally, this study outlines that digital tools do not substitute for, but rather, support scholars' research work. This aspect is shared across all generations of academics interviewed, though the younger researchers perceived these tools as fundamental to develop a scientific paper.

With regard to the last objective of this research – to examine scholars' expectations regarding the future use of digital tools – this study highlights a common opinion regarding the positive development of these tools, in that the scholars had formulated expectations especially regarding improvements to features and functionalities of digital services. The interviews also provided an opportunity to discuss the overcoming of difficulties face by scholars in carrying out their studies. There was agreement that scholars are essential to state hypotheses, to explain results, to develop theoretical models, and to present novelties by enhancing the scientific debate. However, the wide development of digital research services and software have accentuated the trend of writing more standardized papers following the IMRAD structure (Wu, 2011), making cognitive work highly fragmented and impersonal. Although this opinion is not particularly developed across young researchers, multiple scholars believed that future research could end up with modular and very standardized papers that provide insignificant advances and few groundbreaking impacts, differing from each other only marginally. A phenomenon that could be identified and named “mass science” surely needs to be investigated. Engber (2017) named this phenomenon “sameness,” and claimed that this is strictly connected with the use of AI, which could make all papers similar in their structure, with no errors, with no “prose,” though surely more efficient and effective in terms of “publishability.”

An even more delicate issue is related to outsourcing to specialized academic writers or AI applications to obtain a complete draft of a paper, possibly including the more valuable and original parts of the article. We did not find services advertising that they could write an entire paper; however, Tatalovic (2018) wrote about an AI academic writer service able to write complete drafts of papers, though it was not clear from the website of the provider or the authors the level of completeness achieved. This opens a debate about loss of author control and dilution of his or her contribution and establishing “unfair competition” among researchers. Outsourcing more and more parts of the work needed to complete a paper – and ever more substantial parts – makes the author more an assembler of research pieces than an author in the proper – or at least traditional – sense. A question emerges as to what is left of authorship when a paper collects research pieces that the author did not personally conduct, but rather, bought from various providers.

This study makes several theoretical contributions. First, despite the growing popularity of digital professional services and software dedicated to academic research, there is not much scientific literature published on this topic. In fact, most of the available research has focused on scientific projects (e.g. Russell-Simmons et al., 2016; Al-Maadeed et al., 2021) rather than research product management.

Second, this study contributes to advancing the knowledge on this topic by investigating how digitalization affects the academic research publication process. In this regard, this research takes into account the perspective of not reviewers, who are already widely examined in the literature (e.g. Gilmore et al., 2006), but the management scholars, which was hitherto unexplored.

Third, the interviewees of this research confirmed in general terms the TAM model (Davis et al., 1989, 1992), because perceived ease of use and perceived usefulness were reported as the drivers of the adoption of such digital services and software. Thus, a contribution has been made to expectation theory, in the sense of having expanded the theory in a special field of customers – management scholars.

The results of this study provide suggestions both to scholars who professionally deal with academic research and digital services and software providers. In terms of scholars, this analysis has made explicit and hopefully created greater awareness around the usefulness of using a range of services that are easily and quickly adopted, but on which attention is not always sufficiently focused. To move in this direction, scholars should learn to use these technologies. As some scholars proposed during the interviews, it may be useful to create quality circles among academics to circulate knowledge and acquire sharing regarding the best use of these tools. Essentially, this study invites scholars to make good use of digital tools for their research, by continuing to develop individual critical skill to reflect. In particular, younger researchers are encouraged not to rely too much on such services but, first of all, to gain sound epistemological and methodological basis in order to be in a position of mastering such digital instruments and not to be mindlessly dependent on them. Research at its core is essentially a matter of curiosity, of scientific acquaintance and of desire to contribute to improve people's quality of life, all things that cannot come with a software or a digital tool.

As for digital services providers, it is suggested to disseminate information about their services to optimize their use; often, it seems researchers use just a fraction of the features these services include. In this respect, a kind of community of practice could be built around the most popular services. In addition, customization is emerging, and service providers could invest in developing more personalized versions of their software. However, it is also important to note that the multiple positive effects of using digital services and software for academic research that emerged from this analysis (Table 5) are accompanied by many negative effects, which service providers should examine to design improved digital tools.

Equally important are the social implications since academic research ultimately produces benefits for the whole society. There are several points to note in this regard. First, the possible negative consequence of “sameness” could also operate at an aggregate level; “sameness” or “mass science” would have repercussions not only in the specific field or discipline but in general, thus making the entirety of scientific activity less effective and meaningful.

The second issue concerns using public funds to acquire these services for conducting research and publishing papers. The question of ownership of research carried out by researchers who are public employees is already debated; if publicly employed researchers use additional resources to publish their work, the matter becomes even more complicated.

This paper has, no doubt, several limitations. First, given that a well-established base of scientific knowledge on the topic was not available, an exploratory research was undertaken with a small number of scholars interviewed, which limits the generalizability of the results. Second, only a few international scholars were interviewed, and not all generations were considered for this study. This research would have benefited from a larger number of interviewees worldwide and the inclusion of researchers of different ages.

Given the exploratory nature of the study, it may be a starting point for future academic research and may indeed lead to practical implications for digital service providers and act as a pathfinder for a topic that is still little explored.

Avenues for future research can include the following. Although these services are thought to be widely used and widespread, there are no surveys available to measure their degree of diffusion more precisely. Research on this would, therefore, undoubtedly be helpful for a better understanding of it. A second aspect worthy of further study is related to the judgment that researchers who employ digital research services made toward them. Use does not necessarily imply satisfaction or even inner intellectual approval. Further, this type of analysis should be conducted not in an aggregate manner with a generic reference to all digital research services but analytically, concerning each of them. Relatedly, it would be interesting to know why researchers who do not use digital research services have made this choice. The reasons for non-use may reveal aspects that are not necessarily the opposite of their use. Further, it would be essential to understand the effects that can be observed in scientific production assisted by digital research services from a quantitative point of view and, if possible, also from a qualitative point of view. Finally, as often happens when it comes to digital technologies – and even more so in technologies that involve AI – it would be interesting to analyze the implications of an ethical nature that emerge.