Digital Marketing as a Driver of Change Towards the Circular Economy

Abstract

This study explores the emerging link between digital marketing and the circular economy, situated within the broader context of sustainability and digital transformation. Through a scientometric analysis of 276 documents indexed in Scopus, the research maps the conceptual, intellectual, and collaborative landscape of this intersection. The dataset was refined using the PRISMA standards, and the analysis was carried out with the R package bibliometrix. The findings reveal a highly fragmented body of literature, with a few prolific authors and limited international cooperation. Most of the influential publications are concentrated in a small number of sources, particularly the journal Sustainability. Thematic mapping identifies growing interest in areas such as consumer perception, sustainable development, and the adoption of digital tools by micro-enterprises, while other topics like recycling or public health remain peripheral. Although the scholarly output has surged in recent years (especially during 2023 and 2024), the field still lacks coherent theoretical frameworks or dominant research schools. Overall, digital marketing appears to be a promising driver for promoting circular practices, but further applied studies and enhanced international collaboration are essential to consolidate this line of inquiry.

1. Introduction

An old Castilian proverb says that it is very difficult “putting gates on the field” [1]. Strictly speaking, this old saying serves as a metaphor for the meaning behind the invention of the W3 by T. Berners-Lee [2,3] and R. Cailliau in 1989 [4] while both were collaborators at CERN. Such was the impact of that first digital experience on all social spheres that some authors, such as Goodman [5], speak of a “digital revolution”, while, simultaneously, during the final years of the previous millennium, terms such as “circular economy” (sometimes nominalized through the simple acronym “CE”) began to be coined. These terms reflected a set of changes in global development that aimed to be as environmentally non-intrusive as possible, promoting all kinds of policies related to raising public awareness [6]. Regarding the term CE, although its origin is still not completely clear, Winans et al. [7] suggest that it may derive from the seminal book by R. Carson, Spring [8], while the Ellen MacArthur Foundation proposes that it was defined over the years by a multidisciplinary group of researchers formed by J. Lyle, W. McDonough, M. Braungart, and W. Stahel [9]. The global awareness boost regarding the gradual reduction in pollutants and environmental protection took place in “The United Nations Framework Convention on Climate Change” (Earth Summit—1992) [10,11], whose objectives were strengthened by the Kyoto Protocol (1997), a set of measures that provided legal certainty to the decisions made at that summit by establishing more pragmatic, effective, and legally binding actions [12].

Later, at The Millennium Summit of 2000, most world leaders signed the United Nations Millennium Declaration, in which the Millennium Development Goals were defined [13], serving as a prelude to the current 17 global Sustainable Development Goals encompassed within the so-called 2030 Agenda for Sustainable Development [14]. In the face of all these major milestones, marketing as a scientific discipline has not remained on the sidelines. Its multidisciplinary nature [15] has led to developing a type of marketing entirely focused on sustainable development: sustainable marketing [16,17], in which the digital transformation process has become a cornerstone [18], decisively promoting both creativity and innovation [19]. Consequently, several trends have been followed by contemporary digital marketing [20,21], for example its joint application with AI (Artificial Intelligence) [22,23] and the emergence of Digital Neuromarketing [24], just to mention two current trends. According to the preliminary definition of the World Wide Web, Berners-Lee et al. [4] stated: “The World-Wide Web (W3) initiative is a practical project designed to bring a global information universe into existence using available technology”.

Therefore, we are discussing an initiative that currently commands multiple human activities, whose initial spirit essentially remains the same. What has changed in just 20 years are the technological means through which its processes of interconnection between a myriad of individuals via cyberspace take place [25]. Consequently, detailing how digitalization has evolved is equivalent to establishing a timeline that, grosso modo, begins with the emergence in 1998 of SEO (Search Engine Optimization), an essential tool to enhance the visibility of search engines, which remains one of the main pillars of digital marketing because it helps brands to position themselves in online search results performed by users [26,27].

Search engine advertising followed soon after (2000), especially with the launch of Google AdWords (now Google Ads) [28], enabling the Pay Per Click (PPC) model [29,30], a breakthrough in digital advertising as this model allowed companies to quickly reach their target audience and measure return on investment (ROI) much more effectively [31]. As a result, software programs were developed for “click fraud detection” [32], while, at the same time, graphic applications such as “Mobile 3D Graphics API” (also known as M3G) began to be standardized through the implementation of JSR-184 documents [33].

The period between 2003 and 2010 is considered to be the time of the rise and expansion of social networks. This includes Myspace (2003) [34], Facebook (2004) [34,35], Twitter (2006) (now “X”) [36], WhatsApp (2009) [37], Instagram (2010) [38], and other online platforms that transformed consumer behavior and interaction patterns by becoming key channels for content sharing. This context fueled the popularization of Content Marketing (2008) [39], a previously unknown area where companies, as part of their commercial strategies, began using audiovisual content, blogs, and e-books, improving SEO positioning through greater consumer trust. The rise of Mobile Marketing (2013) [40] is another milestone to highlight since, for the first time, mobile traffic surpassed desktop. Brands started developing “mobile strategies”, launching ad campaigns targeting an audience that was virtually always connected via mobile phone, tablet, or any other mobile device with Internet access.

Meanwhile, Email Marketing automation (2015) [41], “The Internet’s true killer application”, in the words of Dysart [42], was enhanced by tools like Mailchimp [43] and HubSpot [44], enabling an increasingly higher degree of automation capable of creating large-scale, fully personalized email campaigns. It should be noted that, due to automation, email remains one of the most effective marketing channels for increasing sales and customer loyalty. Almost simultaneously, the Video Marketing Era (2015–2017), shaped through platforms like YouTube [45], Facebook Live [46], and TikTok [47], drove content dissemination by leveraging widespread bandwidth growth and the popularity of smartphones [48]. Video content creation helped to foster greater customer engagement, becoming an almost essential element in the design of digital strategies.

With Personalization and Big Data (2018) [49,50], it became possible to efficiently analyze massive datasets in real time, allowing for an unprecedented level of consumer segmentation. Special mention should be made of the so-called Influencer Marketing Era (2016–2020), where platforms like YouTube and Instagram fostered the rise of influencers who, in many cases, ended up becoming spokespeople or brand ambassadors. In this context, it is said that “the artificial revolutionizes the reality” [51], a statement embodied by well-known influencers such as James S. Donaldson or Felix A. U. Kjellberg (also known as MrBeast [52] and PewDiePie [53], respectively), some of whom also use their fame and prestige to support purely charitable causes, completely removed from profit [52]. In the current landscape, AI is playing an essential role in the development of digital marketing by enhancing systematic automation, deepening client personalization through real-time data analysis, personalized content creation, and optimized customer service. Among these cutting-edge technology sets, we find ChatGPT [54,55,56], the Chinese initiative DeepSeek [57], the Spanish project ALIA [58], and Grok [59], owned by the controversial entrepreneur E. R. Musk, current Administrator of the Department of Government Efficiency for the U.S. government.

Despite the growing research on digital marketing [60,61], the specific role that digital marketing currently plays in supporting and consolidating the various circular economy models that have emerged over time has received relatively limited attention from the academic community. While some studies have focused on digitalization as a key enabler of circular strategies [62,63,64], few have explored the main functions of digital marketing in the context of the circular economy, such as in terms of consumer engagement, co-creation of value, or content creation. However, recent works such as Leal et al. [65] and Nobre and Tavares [66] highlight the need to align sustainability goals with marketing communication strategies. In any case, a systematic analysis that jointly examines the existing connections between digital marketing and the circular economy is still lacking.

Therefore, this study seeks to address this gap through an exhaustive bibliometric analysis, which includes, among other elements, thematic trends, the most influential contributors and university centers, as well as a strategic diagram that maps the current structure of research connecting digital marketing and the circular economy. It is also worth noting that bibliometric analyses are particularly well suited for rigorously examining this relationship insofar as, as noted by Põder [67], they are limited to efficiently describing objective processes.

The set of procedures included in the R package “Bibliometrix” (version 5.0.0) [68] has been implemented as the bibliometric toolset in this study. This package provides a substantial number of approaches for the tabulation, processing, and visualization of bibliometric data [69], also enabling content analysis techniques that, among other functions, summarize the current state of a given field of knowledge [70]. To carry out our study, we have structured it as follows. In Section 2, we define the bibliographic database used, which was refined using the PRISMA statement (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) [71]. Then, in Section 3, we present the results obtained, which are duly discussed in Section 4. This work concludes in Section 5, where we list the main findings of the research, its limitations, and, as a final note, a set of practical policy recommendations that may be implemented based on the results of this study.

2. Materials and Methods

When conducting the bibliometric analysis, the first step was to decide which scientific database to use. While it is true that the number of specialized scientific databases has increased in recent years, there are essentially two that receive the most attention from the scientific community: Web of Science (WoS) and Scopus [72,73]. In fact, most bibliometric analyses originate from these two databases [74]. In this regard, we opted to implement the Scopus database as it is the most comprehensive scientific database, with records dating back to 1788, before the French Revolution (specifically, “Transactions of the Royal Society of Edinburgh” [75]). Moreover, Scopus covers 20% more citations than WoS [76] and, likewise, includes a wider range of scientific disciplines [74], which is essential for our research when examining the multidisciplinarity of digital marketing [15] in the current context of the circular economy. Subsequently, a Boolean search was conducted within that database. However, the term digital marketing [77,78,79] is a neologism [80], that is, a very recently created word that lacks a univocal meaning for all researchers. Hence, there are several analogous terms, such as “online marketing” [81,82,83], understood as the set of marketing strategies implemented through the Internet; “Internet marketing” [84,85,86], focused on tactics and strategies using the web as the main communication channel; “electronic marketing” [87,88], which broadly covers the use of electronic media for marketing activities; and “cybermarketing” [33,89], also used to describe marketing within the digital sphere. The Boolean search conducted is outlined in Figure 1. As shown, using the Boolean operators AND–OR, a search was performed covering the 2000–2024 period for the aforementioned terms in relation to circular economy and digital marketing, within the Title, Abstract, and Keywords fields. It should be emphasized that the selection of keywords was by no means random but rather based on a consolidation of terms synonymous with digital marketing following a procedure very similar to that of Munteanu et al. [90], in this case focused on a bibliometric perspective that combines sustainable financial strategies with the circular economy. In any case, it can be observed that the keywords chosen made it possible to carry out the bibliometric analysis in a robust manner, such that removing any of these keywords would have resulted in a significantly lower number of records than obtained in this study (276), which would have made any coherent bibliometric analysis unfeasible. As with any bibliometric analysis, it is assumed that no records prior to the year 2000 satisfy the keywords derived from the Boolean search applied (see Figure 1). The upper boundary was set at 2024 given that 2025 had not yet concluded at the time of this study and its partial inclusion would have lacked analytical significance. In order to homogenize the results, the search was limited to bibliographic records in English. First of all, it is obvious to determine that most of the academic literature is written in English, a fact even more evident if we limit ourselves to the pre-existing relationship between digital marketing and the circular economy. If this search had not been restricted to English, only 7 items would have been included in the database to be implemented in Bibliometrix, six in Spanish, and one in Indonesian. Regardless of the fact that these contributions would have been residual if added to the database (around 2.54% of the records), carrying out any type of analysis between English and Spanish or Indonesian would have been entirely unproductive, by mixing, for example, co-words from different languages. That is why we established that English was used in order to homogenize the analysis of this research.

Data and Methodology

From the previous search, a bibtex bibliographic file was obtained [91], suitable for bibliometric analysis, which subsequently had to be refined according to the PRISMA 2020 statement, as shown in the flowchart inserted in Figure 2, allowing for the elimination of incomplete, incorrect, duplicate items [92], or those not directly related to the object of study.

Regarding the choice in bibliometric tool, we start from the premise that the proliferation of bibliometric applications has been enormous in recent times [93], paralleling the growing interest of the scientific community in this field of knowledge. We could have chosen from among the many available, such as Bibexcel [94], CiteSpace [95,96], Gephi [97], HistCite [98,99], Pajek [100], SCIMAT [101], SCI2 [97,102], UCINET [103], or VOSviewer [98,104,105]. However, in our case, the goal was not to determine which bibliometric application is better or worse for analyzing a given dataset derived from a scientific database but rather which one provides a simultaneous overview or “helicopter view” of the greatest number of dimensions. This expression is widely extended in the scientific literature (see, i.e., [106,107]), which serves as an epitome intended to describe a generalist overview [108] of a given phenomenon.

Therefore, we chose to implement the set of bibliometric procedures contained in Bibliometrix (version 5.0.0) [68], which has become a standard in scientometric research as it jointly analyzes eight dimensions that can a priori be considered essential: (1) overview, (2) sources, (3) authors, (4) documents, (5) clustering by coupling (i.e., per author), (6) conceptual structure, (7) intellectual structure, and (8) social structure. It is worth noting that this tool has been reliably used in many other fields of knowledge, such as Cooperative Economy Research [96], Green Finance [109], Image Recognition [95], Business Organization [110], and Post-Disaster Reconstruction [111], to name just a few examples. The core of the research is presented in

Table 1. Bibliographic database analyzed: main information and characteristics.

Timespan	2001–2024	AUTHOR COLLABORATION
Sources (journals, books, etc.)	185	Single-authored docs	56
Documents	276	Co-authors per doc	3.11
(*) Annual growth rate %	21.55	International co-authorships %	17.03
Time for duplication of bibliographic production in accordance with (*)	3.55 years	DOCUMENT TYPES
Most prolific country in this area of knowledge	Indonesia
Document average age	3.43	Articles	144 (0.5217)
Number of citations	2501	Books	17 (0.0616)
Average citations per doc	9.062	Book chapters	48 (0.1739)
DOCUMENT CONTENTS		Conference papers	42 (0.1522)
Keywords plus (ID)	696	Conference reviews	10 (0.0362)
Author’s keywords (DE)	847	Editorials	1 (0.0036)
AUTHORS		Letters	1 (0.0036)
Authors	826	Reviews	12 (0.0435)
Authors of single-authored docs	43	Short surveys	1 (0.0036)

Source: own elaboration. (*) Annual percentage growth in the number of publications.

, covering the analysis period from 2001 to 2024 across a total of 185 different sources.

The 276 items ultimately selected after applying the PRISMA flowchart (see Figure 2) represent, at first glance, an adequate sample size for the dataset analyzed as Seglen [112] establishes a minimum range of 50–100 items, while Glänzel and Moed [113] estimate that the minimum threshold should be around 50 for the results to be considered minimally consistent. In the absence of a previous benchmarking study to contrast the resulting bibliographic database, it is striking that single authorship predominates. There is still a very low level of transnational cooperation in multi-authored works (just over one-sixth), and, although articles are the primary type of document, they account for just over 50% of the dataset. From this, we infer that the joint analysis of digital marketing and the circular economy is still in its early stages, an entirely reasonable conclusion given the novelty implicit in both terms.

This is supported by the average age per document, which is just over three years, and the relatively low average number of citations per document (for a total of 2501), which stands at around 9 citations. Another noteworthy fact derived from Table 1 is that not a developed country but rather a developing one, Indonesia, holds the highest scientific production during the analyzed period. Clearly, this demonstrates that applications and variants of digital marketing within the context of the circular economy represent a global phenomenon that today attracts, or is beginning to attract, the attention of practically all countries worldwide.

3. Results

First, Figure 3 details the main characteristics of the dataset analyzed. Regarding the evolution of total scientific production—Figure 3a—it is noted that, between 2001 and 2012, the number of works was merely anecdotal or sporadic, while, between 2013 and 2014, practically exponential growth began, observable until the end of the time horizon studied.

In turn, the number of citations—Figure 3b—was also rather low until the aforementioned period (2013–2014), improving somewhat from then on. According to Lotka’s law [114]—Figure 3c—a significant fact in the dataset can be illustrated: scientific production is mainly concentrated in a relatively small group of very prolific authors, while the vast majority of authors contribute a rather limited number of bibliographic items. The ranking of scientific production by country, based on the origin of the respective corresponding authors—Figure 3d—confirms some stylized facts previously indicated in Table 1. Indeed, if we break down the scientific production carried out in each country based on this criterion (TCP = Total Country Production by corresponding author) into two parts, SCP and MCP (Single Country Production vs. Multiple Country Production), it can be seen that publications carried out among authors from more than one country (that is, the degree of intra-country collaboration) are still quite limited. Likewise, in this top ten, we can observe that, alongside Indonesia in first place and the developed countries, there are also several emerging or developing countries interested in analyzing the degree and nature of the relationship between digital marketing and the circular economy (i.e., India, Thailand, and Malaysia).

Figure 4 outlines Bradford’s law [115], a prototypical bibliometric procedure based on subdividing bibliographic datasets into groups or clusters of equal size so that the number of records increases geometrically, resulting in three clearly differentiated zones: (1) the peripheral zone, composed of contributions on the subject that appear completely sporadically, (2) the intermediate zone, composed of a much broader group of records that publish more specialized works but in much lower quantity, and, finally, (3) core sources, represented by a relatively small number of highly specialized contributions on the object of study, identifying the most relevant sources for the research. In our specific case, Figure 4 summarizes this core, highlighting sources such as Sustainability (mainly), IOP Conference Series, Springer Proceedings in Business and Economics, Emerald Emerging Markets Case Studies, and Lecture Notes in Networks and Systems.

Next, Figure 5 deepens the analysis of the countries that contributed most. For example, Figure 5a outlines the top ten ranking of countries with the highest scientific production regardless of the origin of the corresponding author, while Figure 5b shows the temporal evolution of the scientific production of the top five countries in this ranking: Indonesia, China, India, United States, and Greece.

Again, we can verify that, at the beginning of the period, the contributions were quite sporadic (limited to the USA and India), with the rest of the countries joining soon after: Greece in 2014, Indonesia in 2019, and Portugal in 2020. In this sense, two circumstances should be noted: the great surge in scientific production occurred during the last two years of the study period (2023–2024) to the extent that more than 50% of the total production is concentrated in that biennium. On the other hand, as indicated previously, the 2013–2014 subperiod marks the beginning of the acceleration in the growth of scientific production. This turning point may likely be due to the emergence of seminal works in this area, such as Van Mierlo [116], Andreopoulou et al. [117], and Li et al. [118], focused, respectively, on digital health, the duality between sustainable development and rural tourism, and paid search marketing. This diversity, of course, serves to emphasize the multidisciplinary nature of digital marketing, even more so when confronted with the circular economy. Finally, Figure 5c presents a ranking of countries based not on the origin of the corresponding author (as in Figure 5b) but on the number of citations received by each country. Here, the panorama is entirely different from that of Figure 5b as the ranking is led by developed countries renowned for digitalization, such as China, Canada, Spain, the USA, and the United Kingdom, all of which show very high average citation numbers (20.5, 131.5, 30, 17.1, and 19.8, respectively).

For its part, Table 2 lists the 10 most cited contributions in the dataset, disaggregated by total citations, total citations per year, and normalized total citations, that is, total citations divided by the average number of citations per year and field. Considering that this top ten accumulates over half the citations in the dataset and that one source, Sustainability (Switzerland), overwhelmingly dominates the rest, this confirms what was previously stated in the core sources of Bradford’s law (see Figure 4) as a very limited number of publications end up becoming the most important sources in the field. Consequently, Table 3 displays a triple ranking of authors (Panel A), sources (Panel B), and affiliations (Panel C), using six key indicators: H index, G index, M index, TC (total citations), NP (number of publications), and PY Start (publication year start). The H index [119] is defined as a hypothetical maximum number h for a given entity (researcher, institution, etc.) that has published h bibliographic items that have been cited at least h times.

The G [120] and M [121] indexes are derived from the former: while the first represents the maximum number g such that the most cited works together received at least g² citations, the latter is calculated by dividing a given H index by the total number of years since the first publication. Regarding the author ranking (Panel A), apart from the exceptions represented by Z. Andreopoulou [117], D. Chhabra [122,123], and C. Koliouska [117], whose initial contributions date from 2014 to 2015, the rest of the works were carried out quite recently, or even contemporaneously (2020–2023). Clearly, the authors in this ranking present very similar metrics (i.e., H index, G index, and NP all equal to 2). Although very small, these figures confirm Lotka’s law (see Figure 3c) inasmuch as the scientific output is concentrated in a relatively small number of authors on this topic. Likewise, as shown in the source ranking (Panel B), we once again confirm the pattern established by Bradford’s law [115] as only one source, Sustainability (Switzerland), accumulates the highest number of publications in the dataset, far ahead of the rest. Indeed, this journal leads in the number of academic items published in this field (35 in total), with H index and G index values of 15 and 23, respectively.

Table 2. Top ten of the ten most relevant contributions in this specific field of knowledge.

Paper	Total Citations	*TCs per Year	Normalized *TCs
Van Mierlo [116]	257	21.417	2.511
Bassamzadeh and Ghanem [124]	89	9.889	2.918
Kapoor et al. [125]	85	21.250	7.296
Low et al. [126]	84	14.000	4.145
Shen and Chou [127]	84	21.000	7.210
Ivars-Baidal et al. [128]	81	16.200	7.505
Harries et al. [129]	70	5.385	2.838
Diez-Martin et al. [130]	61	8.714	3.459
Dumitriu et al. [131]	59	8.429	3.345
León-Quismondo et al. [132]	51	8.500	2.517

Source: own elaboration. *TC: Total Citations.

Table 2. Top ten of the ten most relevant contributions in this specific field of knowledge.

Paper	Total Citations	*TCs per Year	Normalized *TCs
Van Mierlo [116]	257	21.417	2.511
Bassamzadeh and Ghanem [124]	89	9.889	2.918
Kapoor et al. [125]	85	21.250	7.296
Low et al. [126]	84	14.000	4.145
Shen and Chou [127]	84	21.000	7.210
Ivars-Baidal et al. [128]	81	16.200	7.505
Harries et al. [129]	70	5.385	2.838
Diez-Martin et al. [130]	61	8.714	3.459
Dumitriu et al. [131]	59	8.429	3.345
León-Quismondo et al. [132]	51	8.500	2.517

Source: own elaboration. *TC: Total Citations.

It also spans a relatively long period (2018–2024), with the initial work by Jankowski et al. [133] on viral marketing and more recent studies such as Piras [134], which conducts a thorough review of technological innovation in the wine industry.

Interestingly, within the ranking of the ten most productive academic institutions in this research field (Panel C), none of the countries represented appear more than once. Moreover, each institution has produced a relatively small number of contributions, ranging from three to six. This can largely be attributed to the fact that this research field is still emerging (barely over 20 years old) and remains highly fragmented both in terms of country distribution and thematic focus when linking digital marketing with the circular economy. Regarding the characterization and evolution of the different topics and research lines analyzed, Figure 6 presents their current taxonomy through a strategic diagram. This method implements Fuzzy Sets Theory (FST) [135,136] in bibliometric analysis using co-word analysis [137,138] to study various themes based on two parameters: “Callon’s density” and “Callon’s centrality” [137,138]. In this sense, centrality measures the degree of interaction of a theme (grouped in a given cluster) with other themes, being an indicator of its importance or external relevance within the scientific field analyzed, while density reflects the degree of internal cohesion of a theme (grouped in a given cluster), determining how developed and specialized that thematic cluster is. It should also be noted that the strategic diagram is oriented toward making inferences and analyzing the evolution of a scientific domain, being much more than a merely descriptive or exploratory analysis [139].

Table 3. Top ten authors, sources, and affiliations in this specific field of knowledge.

Panel A: Top Ten Authors
Author	$\mathit{H}$ Index	$\mathit{G}$ index	$\mathit{M}$ Index	TC	NP	PY Start
ALMANSOUR M	2	2	0.500	53	2	2022
ANDREOPOULOU Z	2	2	0.167	32	2	2014
BALAJI MS	2	2	0.400	100	2	2021
CHHABRA D	2	2	0.182	7	2	2015
GIANNAKOPOULOS NT	2	2	0.667	10	2	2023
KANELLOS N	2	2	0.667	10	2	2023
KOLIOUSKA C	2	2	0.167	32	2	2014
MENON S	2	2	0.333	23	2	2020
NASIOPOULOS DK	2	2	0.500	14	2	2022
SAKAS DP	2	2	0.667	10	2	2023
Panel B: Top Ten Sources
Source	$\mathit{H}$ Index	$\mathit{G}$ Index	$\mathit{M}$ Index	TC	NP	PY Start
SUSTAINABILITY (SWITZERLAND)	15	23	1.875	581	35	2018
INTERNATIONAL JOURNAL OF DATA AND NETWORK SCIENCE	3	4	0.600	63	4	2021
IOP CONFERENCE SERIES: EARTH AND ENVIRONMENTAL SCIENCE	3	4	0.600	18	7	2021
JOURNAL OF ECOTOURISM	3	3	0.143	56	3	2005
ADMINISTRATIVE SCIENCES	2	2	0.667	8	3	2023
COGENT BUSINESS AND MANAGEMENT	2	3	1.000	24	3	2024
EMERALD EMERGING MARKETS CASE STUDIES	2	2	0.154	8	5	2013
GLOBAL APPLICATIONS OF THE INTERNET OF THINGS IN DIGITAL MARKETING	2	2	0.667	12	2	2023
INNOVATIVE MARKETING	2	3	0.500	19	3	2022
JOURNAL OF GLOBAL SCHOLARS OF MARKETING SCIENCE: BRIDGING ASIA AND THE WORLD	2	2	0.500	51	2	2022
Panel C: Top Ten Affiliations
Affiliation						Papers
AGRICULTURAL UNIVERSITY OF ATHENS (Greece)						6
SZÉCHENYI ISTVÁN UNIVERSITY (Hungary)						6
UNIVERSITAS PADJADJARAN (Indonesia)						5
CHULALONGKORN UNIVERSITY (Thailand)						4
KRISTU JAYANTI COLLEGE -AUTONOMOUS- (India)						4
AHLIA UNIVERSITY (Bahrain)						3
INSTITUTE OF GEOGRAPHY (Romania)						3
JAMES COOK UNIVERSITY (Australia)						3
UNIVERSIDAD SEÑOR DE SIPÁN (Peru)						3
UNIVERSIDADE DE LISBOA (Portugal)						3

Source: own elaboration.

Table 3. Top ten authors, sources, and affiliations in this specific field of knowledge.

Panel A: Top Ten Authors
Author	$\mathit{H}$ Index	$\mathit{G}$ index	$\mathit{M}$ Index	TC	NP	PY Start
ALMANSOUR M	2	2	0.500	53	2	2022
ANDREOPOULOU Z	2	2	0.167	32	2	2014
BALAJI MS	2	2	0.400	100	2	2021
CHHABRA D	2	2	0.182	7	2	2015
GIANNAKOPOULOS NT	2	2	0.667	10	2	2023
KANELLOS N	2	2	0.667	10	2	2023
KOLIOUSKA C	2	2	0.167	32	2	2014
MENON S	2	2	0.333	23	2	2020
NASIOPOULOS DK	2	2	0.500	14	2	2022
SAKAS DP	2	2	0.667	10	2	2023
Panel B: Top Ten Sources
Source	$\mathit{H}$ Index	$\mathit{G}$ Index	$\mathit{M}$ Index	TC	NP	PY Start
SUSTAINABILITY (SWITZERLAND)	15	23	1.875	581	35	2018
INTERNATIONAL JOURNAL OF DATA AND NETWORK SCIENCE	3	4	0.600	63	4	2021
IOP CONFERENCE SERIES: EARTH AND ENVIRONMENTAL SCIENCE	3	4	0.600	18	7	2021
JOURNAL OF ECOTOURISM	3	3	0.143	56	3	2005
ADMINISTRATIVE SCIENCES	2	2	0.667	8	3	2023
COGENT BUSINESS AND MANAGEMENT	2	3	1.000	24	3	2024
EMERALD EMERGING MARKETS CASE STUDIES	2	2	0.154	8	5	2013
GLOBAL APPLICATIONS OF THE INTERNET OF THINGS IN DIGITAL MARKETING	2	2	0.667	12	2	2023
INNOVATIVE MARKETING	2	3	0.500	19	3	2022
JOURNAL OF GLOBAL SCHOLARS OF MARKETING SCIENCE: BRIDGING ASIA AND THE WORLD	2	2	0.500	51	2	2022
Panel C: Top Ten Affiliations
Affiliation						Papers
AGRICULTURAL UNIVERSITY OF ATHENS (Greece)						6
SZÉCHENYI ISTVÁN UNIVERSITY (Hungary)						6
UNIVERSITAS PADJADJARAN (Indonesia)						5
CHULALONGKORN UNIVERSITY (Thailand)						4
KRISTU JAYANTI COLLEGE -AUTONOMOUS- (India)						4
AHLIA UNIVERSITY (Bahrain)						3
INSTITUTE OF GEOGRAPHY (Romania)						3
JAMES COOK UNIVERSITY (Australia)						3
UNIVERSIDAD SEÑOR DE SIPÁN (Peru)						3
UNIVERSIDADE DE LISBOA (Portugal)						3

Source: own elaboration.

Thus, by plotting these parameters along coordinate axes (x-axis: Callon’s centrality; y-axis: Callon’s density), we obtain a strategic diagram chart that visually classifies [140] the research into four types of topics depending on their quadrant within the bidimensional space [138,141,142,143,144,145]. This method allows us to distinguish the following types of themes [145]:

(1) Upper-right quadrant (“motor themes”), well-developed and essential for structuring a specific field of research.

(2) Upper-left quadrant (“niche themes”), characterized by strong internal links but weak external ones. These are highly specialized topics and therefore of limited or peripheral importance to the overall field.

(3) Lower-left quadrant (“emerging or declining themes”), poorly developed and thus marginal, either emerging or fading topics.

(4) Lower-right quadrant (“basic themes”), important to the field but still underdeveloped, often transversal, being overly generic or foundational. Accordingly, Figure 6 presents this thematic classification, displaying in each quadrant the clusters grouped using the walktrap algorithm [146] across a total of 250 keywords. Specifically, the following themes have been identified per quadrant. Each cluster’s size is proportional to the number of keywords it contains; the cluster title corresponds to the most relevant keyword; and, for the larger clusters (perception, sustainable development, and marketing), the five most relevant terms are included as follows:

• Quadrant I: Motor Themes

              •

  Cluster #1 (Perception): “perception”, “Internet”, “consumption behavior”, “humans”, and “agriculture”.

              •

  Cluster #2 (Micro-Enterprises): “micro-enterprises”, “business models”, and “digital technologies”.

              •

  Cluster #3 (Sustainable Development): “sustainable development”, “digital marketing”, “commerce”, “electronic commerce”, and “online marketing”.

              •

  Cluster #4 (Performance Assessment): “performance assessment”, “building”, “energy efficiency”, “knowledge-based systems”, “machine learning”, and “Pacific Northwest”.

• Quadrant II: Niche Themes

              •

  Cluster #5 (Small and Medium Enterprises): “small and medium enterprises”, “international trade”, and “knowledge management”.

• Quadrant III: Emerging or Declining Themes

              •

  Cluster #6 (Recycling): “recycling” and “supply chains”.

              •

  Cluster #7 (Data Mining): “data mining” and “research focus”.

              •

  Cluster #8 (Public Health): “public health” and “United States”.

              •

  Cluster #9 (Local Government): “local government”.

• Quadrant IV: Basic Themes

              •

  Cluster #10 (Marketing): “marketing”, “sustainability”, “social media”, “strategic approach”, and “business development”.

4. Discussion

The results of this research define it as a field of knowledge still in its early stages, highly atomized in terms of the various topics it addresses, given that it is just over 20 years old. Indeed, it is a multidisciplinary (if not multifaceted) field that encompasses a wide range of themes. However, the approach of digital marketing associated with the circular economy represents an emerging area that has not yet been fully explored. In other words, the potential of digital marketing strategies to advance circular business models is truly promising, but, in light of our results, it would require more practical studies and direct applications to ensure its consolidation. Likewise, regarding scientific output, there has been a substantial increase in research on digital marketing vs. the circular economy, especially in the 2023–2024 biennium. This degree of growth highlights the rising global interest in this field of knowledge, considering that, while most contributions come from developed countries, nations like Indonesia are leading the research, indicating a potential global expansion of the topic under study.

Despite the steady growth in the number of publications, cross-border collaboration remains relatively scarce, with only a limited number of works involving co-authorship between institutions from different nations. This indicates that, although the research activity is expanding, the field has yet to attain a truly global scope, which could initially limit both its influence and evolution. Likewise, as this area of study is still in its early stages, distinct “Schools of Thought” (understood as well-established academic groups or research institutions focusing on particular theoretical or methodological lines) have not yet formed, in contrast to more mature disciplines (see, for instance, Parrales Choez et al. [147] in the scientometric study of Credit Unions). In terms of academic dissemination, journals such as Sustainability have become key venues for publishing relevant research, serving as focal points for scholarly exchange regarding the intersection of digital marketing and the circular economy.

If we evaluate the findings of the strategic diagram (Figure 6) in light of the academic literature, we observe that the results are conclusive. Indeed, Quadrant I (high centrality and high density, motor themes) represents key areas with the greatest potential for impact and therefore indicates future research directions. This is the case of “consumer perception” regarding sustainable products and services, which is crucial for the strengthening of the circular economy, with digital marketing playing a fundamental role in facilitating such perception [148]. A similar case is that of “micro-enterprises”, which tend to integrate circular business models linked with digital marketing [149], or “sustainable development”, which is reinforced by the educational potential that digital marketing can have on consumers, encouraging fully sustainable purchasing decisions [150]. As for “performance assessment”, this term is essential for measuring the effectiveness of digital marketing strategies on the circular economy while also optimizing resources and improving their effectiveness [151]. In contrast to Quadrant I, Quadrant III (low centrality and low density, emerging or declining themes) includes those topics that may either disappear or become future opportunities over time. These are themes such as “recycling” [152] or “data mining”, which help to identify sustainable consumption patterns in order to personalize digital marketing strategies [153]. Within this quadrant of marginal topics, we can also find themes like “public health”, considering that digital marketing can influence the perception of sustainable products as beneficial for the health of broader society [154], or “local government”, bearing in mind that these governmental entities may implement digital tools aimed at adopting a circular economy within the public sector [152].

Meanwhile, Quadrant II (low centrality and high density, niche themes) is based on highly specialized topics that still have limited impacts in academia. This is the case, for example, for “small and medium enterprises”, which apply circular economy practices in a fragmented way depending on the sector and access to digital resources [155]. Lastly, Quadrant IV (high centrality and low density, basic themes) identifies works that are fundamental to the field but still underdeveloped, such as the term “marketing”, understanding that its digital implementation is essential for the development and consolidation of the circular economy, although it currently requires greater specialization and a higher degree of practical implementation.

On the other hand, it should be noted that research in the field of digital marketing and the circular economy remains limited, indicating the need to explore this emerging area. The results show that the core sources for publication are less diverse. Marketing and digital marketing journals appear to have shown limited interest thus far in publishing research related to the circular economy. On the flip side, increased interdisciplinarity will enhance the importance of the field, attracting marketing scholars to study the relationship with sustainability and the circular economy. This can broaden the discourse about sustainability marketing to contribute to a circular economy.

Furthermore, the results revealed the most active nations and institutions publishing on digital marketing and the circular economy. These findings encourage future researchers to find opportunities for collaboration while filling the research gap by investigating the topic in different countries and organizational contexts. Moreover, the cluster analysis revealed that the most relevant keyword was ‘perception’, followed by ‘sustainable development’ and ‘marketing’. Perception in terms of consumer behavior is related to digital marketing and the circular economy. Consumer decision-making is shaped by marketing activities and plays a crucial role in the transition to a circular economy. Consumers’ sustainable lifestyle, environmentally conscious behavior, and sustainable consumption patterns support the circular economy.

Consumers’ perception of sustainable products and sustainable brands reinforces sustainable corporate behavior. Consumer attitudes and demand for products and services produced using circular economy principles influence companies’ decisions regarding sustainable production processes. In this respect, sustainable consumption and production contribute to the circular economy. Designing products that are repairable and recyclable but also promoting sustainability practices and encouraging product reuse and recycling will reduce waste, prevent the depletion of natural resources, and protect the environment. In this sense, digital marketing impacts consumption patterns and shapes consumer sustainability behavior. Augmented digitalization ensures public outreach through digital channels to raise awareness of circular economy principles and sustainability issues. Thus, digital marketing can be used to foster the circular economy’s aims of eliminating waste and pollution, extending product lifespans, and regenerating natural resources.

5. Conclusions

This research has analyzed the growing relationship between digital marketing and the circular economy, a field of knowledge that is still in its early stages and has begun to attract the interest of the scientific community over the past 20–25 years. Interestingly, both the rise of sustainability awareness in the business environment [156] and the emergence of bibliometric analysis procedures [93] have occurred almost simultaneously. In other words, this research has focused on a novel field of study, applying equally novel analytical tools, which, at first glance, reveal a domain characterized by high thematic dispersion, accompanied by a low concentration of cited works that also exhibit very limited international collaboration. In this sense, it has been verified that, although there is certain interest among researchers in exploring how digital marketing can give rise to strategies and sustainable models within the scope of the circular economy, the body of knowledge generated through scientific production lacks sufficient robustness to assert that a cohesive theoretical framework truly exists. Hence, the link between digital marketing and the circular economy has not yet been sufficiently explored despite its remarkable potential [157].

In any case, throughout this article, thematic areas have been identified that, both due to their conceptual depth and practical projection, have begun to consolidate as strategic cores within the conceptual framework of the digital marketing vs. circular economy relationship. Relevant topics have been detected, such as consumer perception regarding the emergence of sustainable products, or the gradual incorporation of micro-enterprises into circular production and consumption models. These themes indicate that digital marketing is more than just a communication channel; it can also become a key element in creating and promoting responsible consumption environments.

With regard to the geographical distribution of the field of knowledge analyzed, the results show a plural, and somewhat unexpected, scenario in which, although developed economies lead the research, several emerging countries are beginning to stand out by contributing an increasing number of publications.

In summary, this article has helped to clarify the academic debate on the role that digital marketing can play in the transition towards circular economic models based on the premise that, although interest in this synergy has grown significantly in recent years, a higher level of theoretical integration is still necessary, as well as the adoption of more homogeneous methodological approaches—although such issues may stem from the multidisciplinary nature of this specific field of knowledge.

As with any other research, this study also presents certain limitations. Primarily, as previously noted, the absence of a universally accepted definition of digital marketing—coupled with the fact that, according to Kirchherr et al. [155], there are 114 different definitions—poses a challenge that a priori may constrain the establishment of a one-to-one relationship between the two key elements of this study.

Another plausible limitation lies in the fact that we did not employ an ad hoc procedure for selecting keywords, such as a systematic literature review, prior frequency analysis, or consultation with experts. While these strategies could potentially have enhanced the rigor of our research, they would also have considerably delayed the process. Moreover, none of them guarantee a “perfect” keyword selection. In this regard, the argument of Chen and Xiao [158] is reinforced: to date, no method exists that can automatically identify keywords optimally. Nevertheless, the development of a standardized keyword selection procedure represents a valuable line of future research.

A further limitation concerns the use of a single database (Scopus) without incorporating others, such as Web of Science. Although using multiple databases could indeed have enriched our study, this represents one of the most significant challenges in bibliometric research. Beyond the need to deal with different metadata structures across platforms, there are also substantial operational difficulties given the persistent differences in journal coverage across scientific domains and the variations in citation indexes due to the differing indexing policies [159].

Future studies can focus on the digital marketing channels that are best for promoting circular economy principles. Moreover, studies can address how digital marketing affects consumer perceptions of sustainability and the circular economy, which further influence business decisions. It is noteworthy that societal expectations regarding sustainability can exceed companies’ abilities to adapt to the circular economy. Consequently, innovative solutions should be found through future research. Companies using digital marketing can create realistic expectations and communicate with stakeholders, disclosing information about their achievements towards a sustainable future. Therefore, future studies should analyze digital marketing strategies for companies to implement sustainability goals and communicate their achievements. Furthermore, international cooperation in research on the relationship between the circular economy and digital marketing will enrich the relevant literature by adding insights from different perspectives.

To effectively promote international research in digital marketing and the circular economy, collaborative research projects that bridge geographical and disciplinary boundaries should be encouraged. Providing funding and resources for international research teams and creating platforms to share findings will connect researchers across different geographical locations and academic disciplines. Thus, future studies should delve deeper into examining strategies to promote international cooperation in the field of digital marketing and the circular economy.

Finally, it is worth emphasizing that, based on the strategic diagram developed, it has been possible to draw a series of inferences regarding the categorization of the pre-existing research lines between digital marketing and the circular economy. However, the results should be interpreted with caution within a short-term context as any type of prediction (especially those based on the long term) will always be rather uncertain and, in any case, not sufficiently meaningful [160].