PowerPoint Presentation

[Audio] Welcome to module 4 "Business ecosystems and business models in digital agriculture ".

[Audio] In Module 1 we met the business ecosystem as a concept. Now we use it as an analytical tool for business strategy, so let us sharpen it. The idea goes back to James Moore in 1993, and it has been developed by strategy researchers since. Ron Adner, writing in the Journal of Management in 2017, proposed looking at an ecosystem as a structure. The anchoring point is not a single company but the system of innovations that lets the customer actually use the end product. In other words, the ecosystem links a core product, its components, and the complementary products and services that together add value for the customer. Another influential contribution, by Michael Jacobides, Carmelo Cennamo and Annabelle Gawer in the Strategic Management Journal in 2018, asks what makes ecosystems different from markets, alliances or hierarchically managed supply chains. They describe ecosystems as groups of interacting organisations, enabled by modularity, that are not hierarchically managed, and that are bound together by the fact that their collective investments cannot easily be redeployed elsewhere. So participants stay in an ecosystem because what they have built only makes full sense together. Let us apply this to a concrete service: precision irrigation advice. To deliver it, you need soil moisture sensors from one company, connectivity from a network operator, a platform that stores and analyses the data, agronomic knowledge to turn data into a schedule, a farmer who is willing and equipped to act on the recommendation, and often a water authority or an energy supplier whose rules shape what is possible. No single actor delivers the whole thing. Each depends on the others, and each has made investments, in devices, software, training and relationships, that are tied to this particular combination. This view has two practical consequences. First, the weakest link determines the value delivered. Excellent sensors are worthless if the farmer cannot get a connection or does not trust the advice. Strategists call such a weak link a bottleneck, and finding it is the first task in designing a service. Second, roles matter. An orchestrator sets the architecture and the rules of participation. Complementors add components. Users and customers provide demand and feedback. Enablers, such as universities, standards bodies and public authorities, supply knowledge, trust and infrastructure. In agriculture, the orchestrator could be a technology company, a machinery manufacturer, a cooperative or a public research institute, and as we will see, these choices lead to very different outcomes for farmers. For reflection: choose one digital service you know. Who is the orchestrator? Who are the complementors? And where is the bottleneck?.

[Audio] Before we look at business models in digital agriculture, let us make sure we share the same basic tool kit. A business model is a description of how an organisation creates value for customers, delivers it, and earns a return. David Teece, in a well-known 2010 article in Long Range Planning, explores why business models matter and how they connect to strategy and innovation, and he frames the question as how an enterprise can organise itself to meet customers' needs, get paid for doing so, and make a profit. A simple way to structure this is with four questions. The first is the value proposition: what do we offer, and why would a customer want it? The second is value creation: what activities, resources and partners do we need to produce it? The third is value delivery: through which channels and relationships does it reach the customer? The fourth is value capture: how do we earn revenue and what is our cost structure? In the literature on food chains, these four aspects appear repeatedly as the basis for analysing business model innovation. A very popular practical tool is the Business Model Canvas, introduced by Alexander Osterwalder and Yves Pigneur in their 2010 book Business Model Generation. It divides the business model into nine building blocks: customer segments, value propositions, channels, customer relationships, revenue streams, key resources, key activities, key partnerships, and cost structure. It is widely used in teaching and by start-ups, and I recommend that you try it in your own courses and projects. What do we know about business models in digital agriculture specifically? A review of the social science literature on digital agriculture, by Klerkx, Jakku and Labarthe, noted that empirical research on business models in digital agriculture remains rare, and that existing typologies are often limited, for example to new direct marketing solutions between farmers and consumers. That has changed somewhat since 2019, as we will see in the next slides, but it is a reminder that this is a young field, and that practitioners and students can contribute. There is one more point that is crucial for our ecosystem perspective. A business model never stands alone. A sensor company's value proposition only works if a platform can receive its data, and a platform's revenue model depends on there being users willing to pay. So when you analyse a digital service, always ask two questions: does the business model work for each actor, and do the business models of the different actors fit together? Take a moment now: pick a digital service and fill in the "value capture" box. Who pays, how much, and for what?.

[Audio] Let us look at what business models are emerging. I will combine evidence from the research literature with a simple framework that I propose for teaching. First, the evidence. A 2025 systematic review in the journal Systems, by Sun and colleagues, analysed ninety-five peer-reviewed studies on digital innovation and business model transformation in agricultural small and medium enterprises. The review highlights five main clusters of digital technologies that these companies use to reconfigure their business models: mobile platforms, artificial intelligence, the Internet of Things, blockchain and traceability systems, and cloud-based data infrastructure. A notable finding is that, in thirty-seven of the studies, firms moved from volume-driven, single-stream models to transaction-based, multi-stream and performance-centric ones. The review also reports that value propositions were reshaped through bundling and personalisation, and that platform-based models created new interdependencies, with challenges around governance, data ownership and pricing transparency. In some cases the platformisation of a small company positioned it as an orchestrator connecting groups such as farmers, buyers and logistics providers. Now my teaching framework, which is a synthesis for discussion rather than a research classification. I suggest six archetypes. One: the product with a digital add-on, such as a tractor, sprayer or milking robot sold with sensors and software. Two: software as a subscription, such as farm management systems paid per year or per hectare. Three: advisory and analytics services, where the value is a recommendation, a map or a forecast, delivered by an advisor or company. Four: the marketplace or platform, which connects buyers and sellers of inputs, produce, services or data, and typically earns a commission or fee. Five: outcome-based or performance models, where payment is linked to results, such as savings, yield protection or verified sustainability outcomes. Six: data intermediation, where a neutral party facilitates data sharing under agreed rules. Real companies often combine several of these. A machinery manufacturer may sell the machine, charge a subscription for the software, and open a platform for third-party apps. There is also an important note from the Klerkx review: a study in the same journal issue, on Western Canadian agriculture, contrasts top-down corporate approaches with open platforms and identifies several models along that spectrum, which reminds us that the degree of openness is itself a strategic choice. As you listen to these examples, think about your own region. Which of these six archetypes do you see already, and which are missing?.

[Audio] Value capture is where many digital agriculture initiatives struggle, so let us look at it carefully. Let me list the typical revenue models. Subscriptions: a fixed fee per year or per month. Usage-based pricing: per hectare, per animal, per analysis or per transaction. Commissions, typical for marketplaces. Licences or access fees, for example charging for access to a data set or an application programming interface; the French platform Agdatahub, for example, ran a marketplace where data sets and interfaces could be offered free or against payment under a licensing framework. Freemium models, where a basic service is free and advanced features are paid, a pattern that Teece already discussed in 2010. Public funding, where a public body pays for the service because it serves a public interest. And membership fees in cooperatives, where a service is financed collectively. Behind this list lies a structural problem that I call the "who pays?" problem. In digital agriculture, the actor who generates the data, usually the farmer, is often not the actor who gains most from it. Consider a farmer who records sustainability practices. The farmer incurs the effort. The benefits may accrue to a processor who needs to prove compliance, a retailer who wants to market a sustainable product, an insurer who can price risk better, or a bank that can lend more safely. If only the farmer is asked to pay, adoption stalls. If the beneficiaries pay, or share the cost, the model can work, but then we need agreements and trust between the actors, and we are back to ecosystem governance. This also connects to what Rotz and colleagues observed: many farmers struggle to see the value of the data their own technology produces. A business model that does not make the benefit visible to the farmer, in euros saved, risks avoided or premiums gained, will have difficulty. The DIVINE project that we saw in Module 3 makes a related point: because a broad cost-benefit analysis of agricultural data sharing had not been published, private data holders were hesitant to share with others. Evidence of benefits is therefore itself a commercial asset. A final warning concerns reliance on grants. Many digital services in agriculture start as projects financed by public money, which is fine for experimentation. But a project ends, and a service needs a revenue model that survives it. We will see in a moment a very concrete example of what happens when sustainable financing is not found in time. For discussion: for a service you know, list the actors who benefit from the data, and next to each, write whether they pay today..

[Audio] Let us map the main actors in the ecosystem and the roles they can play. I will group them, and for each I will point out an opportunity and a tension. Technology and machinery providers supply the devices and software at the core. Their opportunity is to offer integrated solutions. The tension is between closed systems that lock customers in, and open interfaces that enlarge the ecosystem. Input suppliers, as we saw in Module 2, are moving from products to solutions, and often have the closest relationship with farmers. Software start-ups bring agility and new ideas, but are often vulnerable financially and dependent on larger partners. Cooperatives, producer organisations and advisors are the trusted intermediaries closest to farmers. They can aggregate demand, negotiate terms, translate between technology and practice, and represent farmers' interests in data discussions. Processors, retailers, banks and insurers are the downstream and financial actors who increasingly need data, and who may be the ones able to pay for it, as we discussed. Research organisations and higher education institutions have a dual role. They create and validate knowledge, and they educate the people who will work in the ecosystem. They can also act as neutral partners and testbeds, which brings us to the concept of living labs and innovation hubs. Public authorities set the rules, run subsidy and advisory systems, hold public data, and can finance shared infrastructure. A particularly relevant type of intermediary is the Digital Innovation Hub. Under the European approach, these are support organisations that aim to make businesses more competitive by speeding up the development and uptake of digital innovations, offering a one-stop shop where companies, especially small and medium enterprises and start-ups, can get technology testing, financing advice, market intelligence and networking. The SmartAgriHubs project, for instance, built a network of such hubs for the European agri-food sector, with twenty-eight flagship innovation experiments across nine regional clusters. For farmers and cooperatives, a good hub reduces the risk of trying a new technology. I would like to draw your attention to how this links to TALLHEDA. The project is building a long-term alliance for digital agriculture among higher education institutions in Greece and Serbia with partners in Belgium, using a quadruple-helix approach that brings together academia, industry, government and society. It also features a Virtual Innovation Hub, which unites stakeholders from education and agriculture, showcases real-world success stories and best practices, and is intended to continue communication and collaboration after the project ends. In ecosystem terms, this is an enabler and connector role. For reflection: in your region, which of these actor groups is strongest, and which is missing or weak? Where are the gaps in the ecosystem?.

[Audio] One response to the power asymmetries we have discussed is collective action. Agriculture has a long tradition of cooperatives, and this tradition is now being applied to data and digital services. A good example is DjustConnect, a data-sharing platform for the Flemish agri-food sector. It was launched by the research institute ILVO together with several cooperatives and companies in the chain, namely AVEVE, Boerenbond, CRV, DGZ and Milcobel. Its design principle is neutrality: the central role of a public research institution is meant to guarantee that the platform stays a neutral service, which in turn allows commercial players to use it. According to ILVO, in 2025, five years after launch, the platform had more than three thousand registered farmers and horticulturists, who can use it for free, with fifty-six data connections and seventeen applications. It started in 2020 with one farmer and the support of five cooperatives. ILVO also reports that DjustConnect received a European award for best data space in the category of user engagement and financial sustainability. Notice the business model logic. The farmers do not pay. The platform is run by a public institution, and value is created by the data connections and applications that companies and cooperatives build on it. So the question "who pays?" is answered by public funding for the core infrastructure and by the benefits that the connected companies derive. Another important step happened in June 2024, when DjustConnect, the French platform Agdatahub and the Finnish platform Tritom signed a memorandum of understanding to link their platforms, forming what they described as the first transnational agrifood data space. The motivation is practical: a Finnish farmer using a harvester from a Belgian machinery manufacturer should be able to use data in a Finnish farm management system without registering on a Flemish platform. It is also a sign of fragmentation: in recent years, around eighty platforms for sharing data in the agricultural and agri-food sector have been set up in Europe, according to Agdatahub. What does the research say? A 2025 systematic review in Smart Agricultural Technology analysed fifty-one studies on community-based business models for agricultural and forestry data ecosystems, and identified fifty-three benefits and twenty-five barriers. The authors conclude that benefits outweigh barriers. The idea of a sector-governed facility is not new. In the Wolfert review we discussed in Module 2, the authors report that Poppe and colleagues proposed, as early as 2015, shared infrastructure for data exchange between businesses, which they called Agricultural Business Collaboration and Data Exchange Facilities. For discussion: would farmers in your region trust a platform run by a research institute, a cooperative, a company, or a public authority? What would change their answer?.

[Audio] One of the most dynamic areas for new business models in agriculture links digital data with sustainability. The logic is straightforward. More and more buyers, from food companies to financial institutions, want evidence of environmental performance, such as reduced emissions, better soil health or biodiversity measures. Farmers who can document good practices may gain access to premium markets, contracts or payments. And digital technologies, from sensors and satellites to farm records, can generate that evidence at a lower cost than manual inspection. A concrete European policy example is Regulation (EU) 2024/3012, which establishes a Union certification framework for permanent carbon removals, carbon farming and carbon storage in products. It is voluntary, which means that certification is not a precondition for carrying out these activities. The framework relies on independent third-party verification by accredited certification bodies, and on certification schemes recognised by the Commission and a Union registry for certified units. For carbon farming activities, the regulation specifies a minimum duration of five years and requires a biodiversity co-benefit. It also promotes synergies with existing information systems such as the land parcel information system under the Common Agricultural Policy. For our topic, the interesting point is that the framework creates a potential market for data-based services: monitoring, reporting and verification, often shortened to MRV. A company, cooperative or advisory service can combine satellite data, field records and measurements to prepare and manage the evidence that a farm needs. That is a business model in itself. The framework is still being filled in. The Commission needs to adopt certification methodologies for different types of activity through delegated acts. As of early 2026, a first delegated act for permanent carbon removals had been adopted, while draft methodologies for carbon farming were published for public feedback until 19 February 2026. Anyone using these slides should check the latest status. And there are open questions that you should be ready to discuss with students. Who pays for verification, which can be costly relative to the income of a small farm? Who are the buyers, and how stable is the demand? Who owns and controls the farm data used as evidence, and can it be reused for other purposes? How can we avoid claims that overstate benefits? These are not technical questions. They are ecosystem design questions, which require cooperation between farmers, certifiers, buyers, technology providers and public authorities. The same logic applies to other sustainability claims, such as water use or antibiotic reduction, so the concept extends beyond carbon..

[Audio] Let us draw lessons from experience. What helps digital ecosystems to work, and what makes them fail? Start with a large-scale experiment. The Internet of Food and Farm 2020 project, IoF2020, was a Horizon 2020 initiative led by Wageningen University and Research with a consortium of seventy-three partners. It organised thirty-three use cases across five trials, in arable crops, dairy, fruit, meat and vegetables, to demonstrate Internet of Things solutions in real operating conditions. A paper by Verdouw and colleagues describes how the project aimed to foster business and software ecosystems for large-scale uptake, and a later analysis highlights its lean, multi-actor approach, with use case teams made of end users, technology providers and others. The lesson is that ecosystems can be designed, but they require deliberate work on partnerships and business support, not only on technology. The DESIRA project, a Horizon 2020 initiative coordinated by the University of Pisa with twenty living labs, offers a social perspective. Its coordinator summarised that the structural challenges of rural areas are made worse by the unequal results of a market-oriented digitalisation process. The project identified three key drivers of impact: design, access and complexity. In plain words: how technologies and services are designed, who can get access, and how complex they are to use. This is a useful checklist for any new service. Now the cautionary case. Agdatahub, the French agricultural data intermediation platform that signed the transnational agreement with Flanders and Finland in 2024, applied for judicial liquidation proceedings in November 2024. A Paris court placed the company in liquidation on 3 December 2024, with the date of payment cessation set at 31 October 2024. In its own statement, the company explained that its economic model had not been sustainable without public support during the period needed for the technology to be widely used, and that a government decision in June 2024 to support it, subject to a transition to public management, could not be put in place in time. In February 2025 another company, Netframe, took over the technical assets with the aim of building a technical operator for the agricultural profession. I mention this not to criticise anyone, but because it is exactly the lesson of the "who pays?" question: even a technically advanced platform with strong sector backing can fail if the financing model does not cover the long period before adoption reaches scale. Putting it together, success factors that appear repeatedly are: a trusted orchestrator, whether neutral or well governed; a visible benefit for each participant; a critical mass of users; interoperability with other systems; skills among users and advisors; and a financing model that lasts beyond the project phase. Failure usually has more to do with economics, trust and adoption than with technology..

[Audio] Let us close this module by bringing the main ideas together. We began by sharpening the ecosystem concept: an ecosystem is made of interdependent actors whose contributions must fit together, it is not hierarchically controlled, and it has roles such as orchestrator and complementors, and bottlenecks that limit the whole. We then reviewed the basics of business models: value proposition, creation, delivery and capture, and the Business Model Canvas as a practical tool. We saw that research on business models in digital agriculture is still developing, but that reviews show a shift from volume-driven models toward transaction-based, multi-stream and performance-centric ones, with platformisation and its governance challenges. I also proposed six archetypes for discussion. We addressed the central economic question: who pays? The actors that generate data are often not those that benefit most, and services that depend on grants can fail when funding ends. We looked at the role of the different actors in the ecosystem, including intermediaries such as Digital Innovation Hubs and the connector role of universities. We examined collective models, such as DjustConnect, and the cross-platform bridges between European data-sharing platforms. We looked at sustainability-linked models, using the EU carbon farming certification framework as an example. And we ended with success and failure factors, from the experience of IoF2020 and DESIRA to the cautionary case of Agdatahub. Now your exercise, which works well as group work in a classroom or workshop. Choose one digital service that exists or could exist in your region, for example a crop monitoring service, a traceability system for a local product, or a data-sharing platform for a cooperative. First, draw the ecosystem map: list the actors and draw arrows for the flows of data, money and services between them. Mark the orchestrator and the bottleneck. Second, fill in the Business Model Canvas from the point of view of the service provider. Pay special attention to the revenue streams and ask: who pays, why, and what happens when the funding stops? Third, check your design against the DESIRA checklist: design, access and complexity. If you are watching live, share your idea in the chat. If you are watching the recording, pause here for at least ten minutes. In the next module, we will step back to policy: the European and national frameworks, regulations and funding programmes that shape these ecosystems, and how projects like TALLHEDA connect with them..

References. Adner, R. (2017). Ecosystem as structure: an actionable construct for strategy. Journal of Management, 43(1), 39-58. https://doi.org/10.1177/0149206316678451 Agdatahub (2024). Agdatahub applies for bankruptcy proceedings (company statement). https://agdatahub.eu/en/agdatahub-procedure-collective-liquidation/ Agdatahub (2025). Netframe takes over AgDataHub to develop its role as technical digital operator for the agricultural profession. https://agdatahub.eu/en/netframe-reprend-agdatahub-pour-developper-son-role-doperateur-technique-numerique-de-la-profession-agricole/ Agdatahub / ILVO / DataSpace Europe (2024). Agdatahub networks with ILVO (Flanders) and DataSpace Europe (Finland). https://agdatahub.eu/en/agdatahub-reseau-ilvo-dataspace-europe/ CORDIS. SmartAgriHubs: Connecting the dots to unleash the innovation potential for digital transformation of the European agri-food sector (project 818182). https://cordis.europa.eu/project/id/818182/factsheet CORDIS. Internet of Food and Farm 2020 (IoF2020, project 731884). https://cordis.europa.eu/project/id/731884 CORDIS (2024). Impact and potential of digitalisation in rural communities (DESIRA). https://www.cordis.europa.eu/article/id/447331-impact-and-potential-of-digitalisation-in-rural-communities European Parliament and Council (2024). Regulation (EU) 2024/3012 establishing a Union certification framework for permanent carbon removals, carbon farming and carbon storage in products. https://eur-lex.europa.eu/eli/reg/2024/3012/oj/eng European Accreditation. Carbon Removals and Carbon Farming (CRCF) Regulation (EU) 2024/3012. https://european-accreditation.org/carbon-removals-and-carbon-farming-crcf-regulation-eu-2024-3012/ Ghalandarzadeh, S., Kurti, A., Unell, C. et al. (2025). Community-based business models for agricultural and forestry data ecosystems: a systematic literature review. Smart Agricultural Technology, 11, 100958. https://www.sciencedirect.com/science/article/pii/S2772375525001911 ILVO (2025). DjustConnect celebrates its fifth birthday with strong growth. https://ilvo.vlaanderen.be/en/news/djustconnect-celebrates-fifth-birthday-with-solid-growth ILVO (2020). Launch of "DjustConnect": turntable for sharing data in the agri-food chain. https://ilvo.vlaanderen.be/en/news/lancering-djustconnect-draaischijf-voor-het-delen-van-data-in-de-agrovoedingsketen ILVO (2024). ILVO kicks off the first transnational agrifood Data Space. https://ilvo.vlaanderen.be/en/news/first-transnational-agrifood-data-space Jacobides, M.G., Cennamo, C. & Gawer, A. (2018). Towards a theory of ecosystems. Strategic Management Journal, 39(8), 2255-2276. https://doi.org/10.1002/smj.2904 Klerkx, L., Jakku, E. & Labarthe, P. (2019). A review of social science on digital agriculture, smart farming and agriculture 4.0. NJAS: Wageningen Journal of Life Sciences, 90-91, 100315. https://doi.org/10.1016/j.njas.2019.100315 Moore, J.F. (1993). Predators and prey: a new ecology of competition. Harvard Business Review, 71(3), 75-86. Osterwalder, A. & Pigneur, Y. (2010). Business Model Generation. Hoboken, NJ: Wiley. ISBN 978-0-470-87641-1. Poppe, K.J., Wolfert, J., Verdouw, C.N. & Renwick, A. (2015). A European perspective on the economics of Big Data. Farm Policy Journal, 12(1), 11-19. Rotz, S. et al. (2019). The politics of digital agricultural technologies: a preliminary review. Sociologia Ruralis, 59(2), 203-229. https://doi.org/10.1111/soru.12233 Sun, B., Yu, J., Khattak, S.I., Tariq, S. & Zahid, M. (2025). Digital innovation, business models transformations, and agricultural SMEs: a PRISMA-based review of challenges and prospects. Systems, 13(8), 673. https://doi.org/10.3390/systems13080673 TALLHEDA project. https://www.tallheda.eu ; Robs4Crops (2024). Robs4Crops joins forces with TALLHEDA. https://www.robs4crops.eu/robs4crops-joins-forces-with-tallheda/ Teece, D.J. (2010). Business models, business strategy and innovation. Long Range Planning, 43(2-3), 172-194. https://doi.org/10.1016/j.lrp.2009.07.003 Verdouw, C., Wolfert, S., Beers, G., Sundmaeker, H. & Chatzikostas, G. (2017). IoF2020: fostering business and software ecosystems for large-scale uptake of IoT in food and farming. PA17, International Tri-Conference for Precision Agriculture, Hamilton. Wolfert, S., Ge, L., Verdouw, C. & Bogaardt, M.-J. (2017). Big Data in Smart Farming: a review. Agricultural Systems, 153, 69-80. https://doi.org/10.1016/j.agsy.2017.01.023.

[Audio] Thank you!. thank you!. TALLHEDA has received funding from the European Union's Horizon Europe research and innovation programme under Grant Agreement No. 101136578. Funded by the European Union. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or the European Research Executive Agency (REA). Neither the European Union nor the granting authority can be held responsible for them..