This study used an exploratory case study design to identify and characterize GSS firms² and solutions. Due to the focus on how and why GSS solutions are being introduced along with the relative novelty of the phenomenon, case study methodology was deemed appropriate, which enables rich data collection despite a small number of cases (Eisenhardt, 1989; Voss et al., 2002). Multiple cases were selected in order to prevent researcher bias and increase the external validity of generalized findings (Voss et al., 2002). Qualitative research methods that enabled an in-depth investigation of the GSS systems were utilized (Denscombe, 2007).

To identify cases, we made use of an exploratory approach to identify firms implementing GSS solutions. An initial set of firms were identified from previous research conducted by the researchers. This list was expanded through information found in vertical farming newsletters and other industry news sources. Furthermore, we conducted online searches using keywords such as “growing-as-a-service,” “farming-as-a-service,” with a combination of terms such as modular, in-store, and vertical farming. To be included in the study, the firms needed to have a business model that went beyond the sale of plants to include a service component, typically realized through the combination of hardware and software systems. In addition, an effort was made to include firms with a business-to-business focus and ones with a business-to-consumer focus in order to capture the full spectrum of GSS solutions on the market.

The data was collected between February and September 2021. Questionnaires and interviews with firms made up the primary data sources. Due to the start-up environments of the firms, the researchers decided to give a choice between completing open-ended questions via an online survey tool or a video-based interview format so that firms could respond in the manner they deemed best because of often busy schedules. The questionnaire and interviews were developed and focused on seven key areas: (1) Company background and motivations, (2) Overview of how the modular unit/system work, (3) General business model (product and services), (4) Customer experience/training, (5) Benefits of the modular systems, (6) Barriers for modular systems, and (7) Sustainability aspects of the systems. The open-ended questions in the survey and structured interview questions were aligned to enable analysis of the qualitative information whether gathered in written or spoken form. The questionnaire and interview guide employed for the data collection are provided in the Supplementary Material for further information.

Questionnaires and interview requests were sent to 16 firms through the online questionnaire system Typeform. Survey responses were received from seven firms. Two of these were determined to be invalid for the study due to insufficient information or outside the case study criteria. Two firms elected to conduct a structured qualitative interview instead of participating in the survey. Primary data provided by the questionnaires and interviews were supplemented with secondary data sources, including online media articles, videos, and podcasts in order to enhance the reliability of the study through triangulation of data (Yin, 2014). This resulted in seven cases built on primary and secondary data. In addition, despite not having answered questionnaires or being interviewed, further cases were added through the sole use of secondary materials due to the richness of online sources (Yin, 2014; Salmons, 2015). Some of the largest firms providing GSS solutions had ample information in online interviews, podcasts, and their respective websites that enabled the researchers to answer questions in the questionnaire and interview protocol outlined above. Thus, a total of 11 firms from six countries in North America, Europe, and the Middle East were included in the study. See Table 1 for a summary of the data collection and firms analyzed for the study.

As the first step in data analysis, the results of the questionnaires and interviews were compiled and reviewed as individual cases. The two interviews were recorded and transcribed to enable the compilation of data and analysis. During this initial phase of analysis, research memos were written to capture emerging themes (Saldaña, 2013). The researchers were also inspired by themes from the PSS literature (e.g., based on business models and value creation), thus an iterative process between data and literature began, which resulted in the construction of a data matrix encompassing these themes: system characteristics, general business model, benefits/drivers, barriers, and sustainability. The data collection and analysis process is illustrated in Figure 1. The matrix was used to plot information from both primary and secondary sources for all cases and enabled a systematic cross-case analysis and comparison during the second phase of analysis. The goal during this phase was to identify similarities and differences across cases (Miles and Huberman, 1994) as well as convergent or divergent views about the benefits and future needs of GSS solutions. The data matrix is not provided in the Supplementary Materials due to proprietary information and requests from the firms involved. However, anonymized information and data can be provided upon request to the corresponding authors.

All firms employ principles of controlled environment agriculture in their modular farms, including closed environment, sensors, LED lighting, and circulating water systems. While most firms boast a simple “plug-and-play” system, behind the hardware of the modules are complex software components, with remote monitoring of the systems for both business-to-business and consumer options. This happens through a wifi connection and is often accompanied by an App for the customer to also track and monitor the status of the plant growth and environment. Analytics technology is typically applied from the data gathered in order to improve conditions within the modules and promote “self-learning farms.” This enables optimization of the plant environment, with little knowledge or action needed from the user of the module.

Automation is a priority for the providers of the systems in order to minimize manual labor and ensure the systems are easy to use. Most firms include automation of key aspects, including lighting, climate controls, and pumps. Aspects that require human intervention, such as harvesting and cleaning, are handled through push notifications in accompanying Apps in order to minimize planning and time spent on the module. All systems require the initial placement of seedlings or seed pods in the system, and some also separate a “nursery chamber” for young plants that requires movement to a different shelf in the system until the plants are ready for harvest. The systems themselves range from small cube-like structures to shipping containers, with many likened to a large refrigerator unit found either in a home or retail location. The main products grown in the modular systems to date include leafy greens, herbs, and microgreens, with a few offering tomatoes and one focused exclusively on mushrooms.

The results highlight that many of the firms point to undesirable aspects of the current food system, e.g., long transport needs, unpredictability, and pesticide use as drivers to develop new ways to produce and distribute food. These drivers also translate into the perceived benefits of the systems. The ability for hyper-local production is believed to reduce transportation but also give more people the opportunity to be growers, whether that means in a retail location, restaurant, office, or at home. Some firms point to the desire to expose more people to the health benefits of a green environment, especially in cities, despite the fact that the systems themselves are not limited to use in urban areas. Contributing to food resiliency and helping to increase the local food supply are also mentioned by multiple firms as motivations for developing such modular systems.

Beyond these systematic ambitions are also business-specific drivers and benefits. As identified by several firms, the size of the systems opens up possibilities, both for location and revenue diversification. The size of the module systems also enables firms to distribute growing across locations and avoid the strict zoning and building needs of larger farming systems. But aside from these benefits, the systems also play an important role in marketing the firms and their technology. From a firm level, several firms identified that the systems bring visibility to the farms and the use of hydroponics and technology in food production. For business-to-business clients, the firms providing the GSS systems believe there are benefits to the visual appeal of the systems in stores and restaurants. In fact, one of the firms in the study initially envisioned the systems being placed in the back of the house in restaurants. However, the restaurant owners themselves began to demand well-designed systems that could be used in the front of the restaurant as a kind of art installation. This is also apparent in consumer models, where design is a key element of the systems to ensure its integration into the home where space is limited. See Figure 2 for a depiction of different types of GSS modular farms.

Seven of the firms in this study focus exclusively on the business-to-business (B2B) market, though the users of the systems vary from food retailers, restaurants, offices, and public institutions such as elderly care homes. Two of the firms focus exclusively on the consumer market, i.e., business-to-consumer (B2C), while the remaining two have deployed both B2B and B2C models. One of these firms had plans to launch a B2C module and accelerated those plans when the Covid-19 pandemic hit. The remaining firm remains focused chiefly on B2B customers but launched a B2C solution during the pandemic as a pivot when many restaurants in its area were shut down due to restrictions.

The majority of the firms suggested that their GSSs are, or are becoming, more sustainable. This is often related to environmental sustainability, where many of the firms suggest that the modular farms offer resource-efficiency advantages, primarily through reduced water and fertilizer consumption in the horticultural production methods employed.

As mentioned previously, location was identified as a key benefit of these systems. The production of hyper-local foods is often recognized by the different firms as a sustainable advantage, providing reduced transportation through shorter distances in their supply chain, and bringing the crops closer to the consumer. Connected to this are the many suggestions of increased freshness and shelf-life, which in turn are suggested to add to the quality of the product. Furthermore, this proximity to the end-users is suggested to reduce waste along the supply chain by a number of firms.

Sustainability was also suggested by many firms to be a priority in their work. Some firms are looking into reducing supply chain sustainability impacts by reducing shipping distances for consumables needed in the modular farms. Several firms also addressed the sustainability of their packaging materials, suggesting they are moving away from conventional plastics, and have been using, or experimenting with new materials. Furthermore, circularity was also discussed, as several firms are taking steps to include more circular approaches in their production. This includes reusing materials, developing new fertilizers, and improving the integration with urban environments and their building hosts. Nearly all firms were aware of the impacts of energy use, mentioning electricity and its negative environmental impacts. As such many of the studied cases highlight their purchasing of renewable energy or optimization developments to reduce energy consumption; see also discussions above on key characteristics.

The barriers and challenges outlined by the firms can be categorized into broader industry barriers and firm-specific challenges. From an industry perspective, the cost of technology is considered a current barrier, though many admit the costs are decreasing. The variety of products grown in the systems is also seen as a challenge for long-term growth and demand generation. Overall, the efficiency and sustainability factors of the systems are noted as an area that needs to be improved. In addition, one firm also identified the need for better business models in order to achieve economic sustainability of the modular offering. This includes aspects of the contracts and ensuring the long-term use of the systems so they are not seen as just marketing or display tools that are frequently changed out for other product displays, as floor space is often limited and/or expensive. Due to the novelty of the systems and hydroponic growing in general, supply and demand management is also difficult for most firms at this stage.

From a firm-specific perspective, the cases seem to be at different stages of development or concentrating on different concerns. In general, most of the firms are focused on bringing greater efficiency to the hardware/software interaction in order to further decrease the work required by the customer. As noted by one B2B-focused firm, the customers do not want to be farmers, so improving automation and services are seen as vital. Others are focused on increasing the variety of plants that can be grown in the systems and/or the mix within one unit. For B2C-focused firms, the initial costs of the units are seen as a barrier, as they may be considered luxury products in the current market. In addition, space is a concern, especially for city apartments. One firm mentioned the development of smaller units and units with less technology included to bring different price options to the consumer segment.

While there has been an extensive expansion of larger-scale centralized production systems for vertical farming (Butturini and Marcelis, 2020; Kotsier, 2020), our results highlight an expanding smaller-scale modular system for vertical farming. It was found that the novel approach to food provisioning in urban areas is being conducted worldwide, and encompasses a number of different products and systems. As highlighted in previous studies, there is a growing market for such solutions (Jürkenbeck et al., 2019; Butturini and Marcelis, 2020; Renmark, 2021). Our findings imply that these systems are being offered as novel, or niche, approaches, and in B2B environments, as an expansion of the vertical farming firms' own business portfolio. It was found that several firms are combining modular farms with conventional larger scale vertical farms; either starting directly with modular units or starting from larger farms and exploring the use of modular units. Once again, this approach has been highlighted as a way to differentiate from competitors in the market; aligning with previous studies on vertical farm market development, (e.g., Bustamante, 2020). As such, these tailored systems can create customized products to increase competitiveness and a unique profile in the retail market; (see e.g., Pine and Gilmore, 2014; Charters et al., 2017; Jürkenbeck et al., 2019; Sjölander-Lindqvist et al., 2020). A few of the firms in the study highlighted the ability to increase the types of products grown in the systems as an important area for expansion, which would address previous criticism of vertical farming in general (Cox, 2016; Pinstrup-Andersen, 2018); though others argue this limitation is more about economics than system ability (Banerjee and Adenaeuer, 2014).

Jurkenbeck et al. (2020) also found that the transparency provided by such modular solutions, which are directly visible to consumers, greatly improves their acceptance of such systems. Nonetheless, research has shown that consumers may be reluctant to consume foods from these more “technical” or less “natural” solutions (Siegrist, 2008; Coyle and Ellison, 2017; Grebitus et al., 2020), due in part to the lack of knowledge of these systems (Coyle and Ellison, 2017; Jürkenbeck et al., 2019; Yano et al., 2021). As such, by providing a visual element, the GSS providing firms are attempting to break down barriers by providing further transparency to how food is produced in vertical farming environments and engage with consumers. The firms in our study also pointed to this important aspect of the distributed model, which enables consumers to understand hydroponic growing. Placed in the retailers, the module systems provide a unique experience and educational opportunity. Located in homes, consumers are given the power over the product decisions, harvesting and availability. Such effects expand previous PSS research which have highlighted how consumer awareness of PSS systems challenges conventional product ownership, especially in urban areas, with systems for rental, sharing, and services (Acquier et al., 2017; Zamani et al., 2017; Hollingsworth et al., 2019; Martin et al., 2019b, 2021). In addition, few previous studies have outlined B2C examples of PSS systems, where the module is included at home. While such examples are available for B2C applications in the home, e.g., printing (McIntyre, 2018), robotic vacuums (Electrolux, 2019), no systems have outlined food production systems.

While the study provides some general insights into the business models of GSS solutions, it was difficult to obtain a detailed view of any one firm's business model. This could be due to a number of factors including the relative recent entrance of GSS solutions in the market, a desire for secrecy about this aspect of the farms and also the limitation of using open-ended surveys, where firms may have felt less inclined to write detailed commentary on this aspect. This was especially difficult in the B2B-focused firms. However, the analyzed information did uncover a number of interesting points.

First, although the long-term sustainability of the business model is unknown, almost all of the firms acknowledged the benefits of the systems in helping to grow awareness of hydroponics and build market acceptance. For the B2B focused firms, the modular-based systems also provided an opportunity to further develop relationships with retailers and restaurants by providing a unique experience for their end customers. Thus, by introducing the modular systems, even as the business model may be in flux, the GSS providers are able to explore the market and grow a network; which has been acknowledged as instrumental for technology entrepreneurs and a key function of business models (Doganova and Eyquem-Renault, 2009). This ability to extend the current offering beyond the delivery of plants and build relationships with their customers also aligns with key factors for the success of PSS offerings (Annarelli et al., 2016).

Second, while specifics vary, the customers of these systems are paying for a bundling of both products and services. B2C models require the purchase of the system, which also includes access to a number of software applications. The majority of the B2B systems are either leased or rented, and contracts include a number of services, which may or may not include performance indicators around the number of plants harvested and sold. From our results, it was difficult to suggest which conventional PSS model was employed and there does not seem to be one dominant model at this stage (i.e., product-oriented, use-oriented and results-oriented per Tukker, 2004). The B2C models align with a product-oriented model, as the main offering is still the product, which in this case could be considered both the plants and the physical module. B2B solutions, however, are harder to categorize. Some systems seem to align best with the use-oriented model, especially those found in restaurants or offices, as the systems are rented and largely run by the customer. However, retailer-focused solutions are harder to categorize. Some seem to be use-oriented, but others are also based on the number of plants harvested, aligning more with a results-oriented model. This difficulty in categorization points to the difference in our study vs. past PSS studies, which have generally focused on the manufacturing sector. Many times, in those cases, services were added to a long-term use product, where in GSS systems, a plant is the original product. Thus, the GSS system is introducing both a new product (the module) and services to a product that is consumed and used in a relatively short period of time, making it more difficult to fit into the established categories of PSS models. As highlighted, more information is needed for GSS firms to improve upon their business models in order to achieve economic viability of the offering. As such, further design developments and business model iterations may be necessary. Similar assertions are also highlighted in Kambanou and Lindahl (2016) and Bocken et al. (2018).

Last, our study uncovered some insights into business model innovation in the food industry. Vertical farming systems are constantly improving and expanding. As suggested in Klerkx and Rose (2020), vertical farming innovations are potentially game-changing, affecting the way in which food is produced, processed, traded, and consumed. The visibility and benefits of hydroponic growing enables customers to make decisions based on new characteristics of food, such as environmental effects, or by taking into account intangible benefits such as eating a product closer to harvest. The ability to differentiate products based on intangible and tangible benefits, along with “turning ordinary products into extraordinary experiences” have been identified as key PSS benefits (Annarelli et al., 2016). These developments are also in line with consumer demand for more locally produced food, especially in wake of the Covid-19 pandemics (Toler et al., 2009; Granvik et al., 2017; Pulighe and Lupia, 2020). In particular, the B2C modules are challenging the dominant business model in the food industry, where typically an individual buying a product from a store supports the business model of the food retailer (Kaplan, 2012). B2C modules enable the GSS firms to capture the value directly from the end-consumer. Some firms argue that giving the consumers the control over production is an intangible value consumers are willing to pay for. As all of the firms in the study point to a desirability to improve the environmental performance of the current food system, the experimentation of business models that support sustainable innovation is an important and ongoing endeavor, as it is difficult to simply transplant business models from one economy to another if sustainable development is a goal (Boons and Lüdeke-Freund, 2013).

The results suggest that most firms highlight a number of benefits of modular systems. Owing to their proximity to consumers, the location was highlighted as a beneficial aspect of these systems, where the freshness and nutritional aspects of the products were suggested to be superior in these systems. This is especially important for leafy greens, which can begin to lose nutritional value as soon as they are harvested. Indeed, previous studies have suggested that vertical farms can control the genetics, quality, and sensory experience of different croups through optimized conditions during growth and pre-harvest, (see e.g., Selma et al., 2012; Nicole et al., 2019; Sharathkumar et al., 2020). Furthermore, many firms also suggest location is important for sustainability, e.g., by reducing transportation along the supply chain. However, previous studies have shown that the transportation of foods has a relatively minor impact on the overall impact (Edwards-Jones et al., 2008; Coley et al., 2009), and specifically for urban-vertical farms (Martin and Molin, 2019). Nonetheless, an important benefit also highlighted for vertical farms in close proximity to consumers is also related to variety of crops which can be produced, which can be chosen for flavor and taste, thus providing differentiation, which is not always possible in conventional varieties found in retail which may be optimized for transportation resistance (Bogomolova et al., 2018; Harada and Whitlow, 2020; Renmark, 2021).

Beyond transportation, many of the firms outline the advantages the GSS systems provide for environmental sustainability, primarily relating to resource efficiency improvements and reduced toxicity from the lack of pesticides employed. Such motivations are common amongst urban agricultural systems, see e.g., assertions in Specht et al. (2014), and have been found to be a major driver in consumer acceptance of such systems for different vertical farming systems (Coyle and Ellison, 2017; Jürkenbeck et al., 2019). However, no firms highlighted other sustainability pillars, e.g., social or economic sustainability. There are a limited number of studies reviewing sustainability or specific case studies of urban farms in different scales beyond plant factories and rooftop farms (Kulak et al., 2013; Romeo et al., 2018; Martin et al., 2019b) and thus more research could focus on the implication of GSS systems in comparison to their larger counterparts. Furthermore, while such PSS systems are promoted as sustainable alternatives, their economic, social, and environmental implications are rarely accounted for and are often confined to qualitative reviews of the potential of these systems (Lindahl et al., 2014; Salazar et al., 2015; Kambanou and Lindahl, 2016; Bocken et al., 2018). It is important that further developments, case studies, and assessments are explored and tested to ensure they achieve the desired intentions and provide value to both provider and users of the systems (Kambanou and Lindahl, 2016; Bocken et al., 2018; Martin et al., 2021).

A further benefit outlined by most firms is the potential to control the systems to allow for learning and ease of use by the consumers. This is often included in PSS offerings, allowing for the provider to control the system and maintenance and reduce risks for the user (Tukker and Tischner, 2006; Lingegård, 2020), although its influence on the sustainability of the systems are not well-known (Martin et al., 2021).

Our analysis of GSS could be improved in a number of ways. First, the study sought to understand the nature of the activities and technical functions surrounding GSS solutions but did not evaluate their effectiveness in any one area, e.g., market development, sustainability, innovation management. Future studies of GSS systems could include further information and questions relating to the business models employed. As many of the firms suggested that they are designing the systems with the users and consumers in mind, in the future, research could focus on user and consumer perception and perspectives of these systems. Furthermore, while the questionnaire and interviews did not address the lifetime of the modular units, the lifetime, and design for durability are important for the PSS systems. Our study also highlighted a limited geographical selection of such cases, which has examples worldwide, but has a more European focus. Further work can be done to develop knowledge from a broader set of GSS solutions worldwide, especially as they are becoming increasingly apparent. Finally, as the study is focused on a novel method for vertical farming, a more longitudinal approach could be employed to study the change in these systems over time to study their development. Further studies could also focus on the complexity of business ecosystems for GSS solutions.