Цифровую трансформацию промышленности принято обсуждать через призму отдельных технологий, при этом вопрос о том, в какой экосистеме и по каким стандартам работает предприятие, остается в тени. В то же время именно он определяет трансакционные издержки интеграции, степень зависимости от поставщика и стоимость его возможной смены. Для российской промышленности, оказавшейся после 2022 года вне западных промышленных экосистем, этот вопрос стал сугубо практическим. В рамках статьи автор рассматривает роль экосистем и стандартов с позиций экономики трансакционных издержек и теории технологической блокировки, анализирует положение российской промышленности в 2022–2026 гг., сравнивает три стратегии — собственную экосистему, присоединение к китайским экосистемам и гибридную модель на открытых стандартах — и предлагает матрицу выбора стратегии для отдельных производственных контуров. Отдельно обосновывается необходимость учитывать стоимость выхода из экосистемы при принятии инвестиционных решений.

Ключевые слова: цифровая трансформация, промышленные экосистемы, технические стандарты, технологическая блокировка, трансакционные издержки, индустриальные центры компетенций, промышленное программное обеспечение, импортозамещение.

Introduction

Today, when the digital transformation of industry is discussed, attention goes first to technologies: industrial robots, the Internet of Things, digital twins and artificial intelligence. Yet implementation experience shows that an enterprise rarely runs up against the absence of the technology itself. Far more often, problems arise at the interfaces, where one manufacturer’s equipment does not “understand” another’s control system and shop-floor data cannot be passed to the planning system without manual processing. Many companies still have whole groups of employees who spend much of their working time manually transferring data from one format to another. In other words, the bottleneck is not the technology but the compatibility of systems with one another. Systematic literature reviews support this conclusion, treating digital transformation as a change in the company’s entire system of relationships with suppliers, customers and partners [1].

From an economist’s point of view, a considerable part of the costs and uncertainty of digital transformation is transactional in nature. These are the costs of aligning data formats and integrating heterogeneous systems, as well as the losses that stem from dependence on a particular supplier. Standards are the tool for reducing them, and in today’s economy the vehicles for standards are increasingly industry “ecosystems” that bring together equipment manufacturers, software developers and industrial customers.

For Russian industry, this question turned from a theoretical into a purely practical one in 2022. That year the leading Western suppliers of industrial equipment and software left the market, and enterprises found themselves outside the ecosystems in which they had spent years accumulating data, competences and integration solutions. The choice of a new ecosystem became a strategic decision on which the competitiveness of enterprises will depend for years to come.

In this article we examine how ecosystems and standards affect the economics of digital transformation, assess the position of Russian industry in 2022–2026 and propose a tool for choosing an ecosystem strategy at the enterprise level. The methodological basis comprises transaction cost economics, the theory of network effects and path dependence, and the concepts of business ecosystems and industrial platforms. The empirical base consists of data on the industrial software market, robotisation, the work of industrial competence centres and the development of standardisation in Russia and worldwide.

1. Why a standard matters more than an individual technology

Let us first consider the nature of these costs. In O. Williamson’s classic formulation, the costs of interaction between firms are determined primarily by asset specificity: the more strongly investments are “tied” to a particular partner, the higher the risk of dependence and the more expensive it is to change counterparties [2]. In digital industry, such specific assets include data formats, equipment interfaces, application software and accumulated models. An enterprise that has spent 10–15 years developing its production system on one supplier’s platform has invested not only in licences but also in integration, staff training and a data archive. Outside that platform, a significant part of these investments loses its value. Moreover, the longer the enterprise works on the platform, the more it stands to lose when it leaves.

The second important aspect of the problem is network effects. M. Katz and C. Shapiro showed that the value of a compatible product grows with the number of its users, so compatibility becomes decisive in such markets [3]. P. David, using the example of the QWERTY keyboard layout [4], and W. B. Arthur, using models of competing technologies [5], demonstrated that under increasing returns a market can become locked in to a solution that won not because of its quality but because it spread early. Industrial enterprises know this lock-in effect well. Over time the cost of leaving an ecosystem grows faster than the benefits of staying in it, and at some point changing suppliers becomes practically impossible, even if the supplier has long ceased to satisfy the customer.

Standards affect this problem in two ways. On the one hand, a common standard reduces integration costs and uncertainty. For example, K. Blind et al. showed empirically that under high market uncertainty standards stimulate innovation, since they reduce the risk of incompatibility and make market entry easier [6].

On the other hand, whoever shapes a standard gains a long-term advantage. D. Teece, using the example of wireless communications, shows that the owners of technologies included in a standard earn income through licensing and set the development trajectory of the entire industry [7]. C. Shapiro and H. Varian went further and described the rivalry over standards as a “standards war”, whose outcome depends not only on the quality of the technology but also on the ability to assemble a coalition of allies [8]. Coalitions of this kind are usually called ecosystems, but there is no agreed definition of the term in the literature. R. Adner defines an ecosystem as the structure of aligned relationships among the participants needed to deliver a common value proposition [9]. M. Jacobides et al. emphasise complementarity: an ecosystem arises where the participants’ products complement each other but cannot be fully integrated within a single firm [10]. A. Gawer and M. Cusumano place an industrial platform at the centre of the ecosystem: the platform owner sets the interfaces, and the other participants create complements around it [11]. In Russian scholarship, a systemic approach to ecosystems is developed by G. B. Kleiner, who treats them as a special type of socio-economic system combining competition and cooperation among participants [12]. In recent years, Russian researchers have also been developing methods for assessing the maturity of industrial ecosystems [13].

If we compare these approaches, we see that they describe the same object from different angles: some authors emphasise the common goal, others the complementarity of products, and still others the role of the platform. For the purposes of this article, we adopt the following definition: an industrial ecosystem is an association of equipment manufacturers, software developers and industrial customers built around a common set of interfaces and standards that determines the conditions under which their solutions are compatible.

From this we can draw a conclusion that matters for the management of digital transformation. When choosing a technology, an enterprise in effect chooses an ecosystem and a set of standards, and with them its future transaction costs, its degree of dependence and its cost of exit.

2. Ecosystems as a source of standards: international experience

The most telling example of recent years is the German initiative Catena-X, which has brought together carmakers, their suppliers and IT companies to exchange data along the entire value chain. Its participants include BMW, Mercedes-Benz, Volkswagen Group, Porsche, Renault and Volvo, as well as Microsoft, AWS and Huawei. In July 2026, Catena-X announced the launch of the first cross-border automotive data ecosystem between Europe and China, with the participation of the China Association of Automobile Manufacturers and the China Automotive Technology and Research Center. Operations are scheduled to start in November 2026 [14]. In other words, it is no longer individual companies that are converging but entire systems of rules for data exchange.

At the equipment level, a similar role is played by the concept of the Asset Administration Shell, enshrined in the international standard IEC 63278–1:2023 [15]. The standard describes how any production object, be it a sensor, a machine tool or an entire line, provides information about itself in a uniform machine-readable form. For an enterprise, this means being able to connect equipment from different manufacturers without integrating each unit individually.

China, for its part, is building its own standardisation system. The National Standardization Development Outline adopted in October 2021 set two targets for 2025. The share of international standards converted into national ones was to exceed 85 %, and the average time needed to develop a national standard was to be cut to no more than 18 months [16]. The strategy combines rapid adoption of international norms with active participation in their development, that is, a move from the position of a “user” of standards to that of their co-author.

It can be argued that ecosystems that have become a source of standards gain the ability to set the rules for new industries, while participants not involved in developing these rules have to bear the costs of adapting to them. The convergence of European and Chinese ecosystems that began to take shape in 2026 raises the price of isolation for those who remain outside both. As we will see below, Russian industry is in exactly this position.

3. Russian industry after 2022: leaving foreign ecosystems

For Russia, this scenario played out in its most extreme form. In May 2022, Siemens, one of the largest suppliers of automation systems and industrial software, announced that it was winding down its business in Russia [17], and other Western developers followed it out of the market. Enterprises lost updates, technical support and access to cloud services and, most importantly, the ability to develop their production systems within a familiar ecosystem. The dependence turned out to be deep, and it persists to this day. According to the National Computer Corporation (NKK) Mashinostroenie, in 2025 Russian products held less than a third of the computer-aided design (CAD) market, estimated at 29 billion roubles, while about 70 % was accounted for by unlicensed foreign software. For a number of tasks, such as the design of composite parts in aviation and shipbuilding, there are still no domestic equivalents [18]. In effect, a large part of the industry continues to work in someone else’s ecosystem, only now without licences, updates or support.

In robotisation, one external ecosystem has been replaced by another rather than abandoned. About 80 % of the industrial robots used in Russia are still imported, and China has become their main supplier [19]. The stock of industrial robots in manufacturing grew by 62 % in 2024 [20]. The growth looks impressive, but given that four out of five robots are imported, what we see is a change of supplier rather than the emergence of a domestic industry. At the same time, the state began to build its own ecosystem infrastructure. In July 2022, 35 industrial competence centres (ICCs) were formed to replace foreign industry-specific digital products. They brought together industrial customers, experts and developers in 16 industry committees, and the budget of the mechanism for the period up to 2024 amounted to 37.1 billion roubles [21]. By April 2026, the ICC portfolio comprised 175 particularly significant projects to create domestic industry software, 120 of which had been completed, and state grants for them had reached 23 billion roubles [22]. The figures look solid, but the completion of a project does not yet mean that the product has been deployed at dozens of enterprises and has become a common standard for the industry. We will be able to judge how well these solutions take root only later, when statistics on their wider roll-out become available.

Presidential Decree No. 309 of 7 May 2024 set a benchmark: by 2030, at least 80 % of organisations in key sectors should use Russian system and application software [23]. If we compare this benchmark with the situation in the CAD segment, where Russian products account for less than a third of the market, the task looks extremely ambitious.

A national body of standards for digital manufacturing is also taking shape. Rosstandart has approved GOST R 57700.37–2021 “Digital twins of products. General provisions” [24], and Technical Committee for Standardisation TC 194 “Cyber-Physical Systems” is developing the “Smart Manufacturing” series of documents, including documents on catalogues of equipment behaviour for a virtual production system. The first parts of the series came into force in 2023–2024 [25]. These steps are important, but they have an obvious limitation: national standards are being developed in parallel with international ones and are not always aligned with them. As a result, Russian enterprises risk a double incompatibility: with Western systems, from which they have already been cut off, and with Chinese systems, to which they are in effect moving where equipment is concerned. This is a particular concern for exporters and for manufacturers embedded in international supply chains.

Summing up the interim results, we can say that over four years Russian industry has not escaped its dependence on external ecosystems but has largely exchanged one dependence for another. Its own infrastructure, in the form of the ICCs and national standards, is being built, but it has not yet become a fully fledged alternative for enterprises.

4. Three strategies and their comparison

Under the current conditions, three fundamental strategies can be distinguished, and both individual enterprises and the industry as a whole have to choose among them. They are compared against the main criteria in Table 1.

1. “Domestic ecosystem”: the enterprise relies on a domestic set of software and hardware solutions and on national standards, participating in the work of the ICCs. This strategy ensures independence, but it takes time and carries the risk of replacing an external monopoly with an internal one. If only one domestic supplier of a solution remains in the industry, the enterprise finds itself in the same dependence trap, only within the country.
2. “Joining”: the enterprise switches to the solutions and standards of Chinese suppliers, primarily in equipment and control systems. This path delivers quick results but essentially reproduces the situation that existed before 2022. In effect, only the supplier’s country of origin changes, while the cost of exit will again grow as data and integrations accumulate on someone else’s platform.
3. “Hybrid model”: a domestic core of critical systems is combined with open international standards at the interface level (for example, the OPC UA protocol and the Asset Administration Shell), which makes it possible to connect equipment of different origins. This path appears to be the most robust for most enterprises, since open interfaces reduce the cost of changing suppliers in the future.

Table 1

Comparison of ecosystem strategies of Russian industrial enterprises

Source: compiled by the author.

As Table 1 shows, none of the strategies dominates on all criteria. The “domestic ecosystem” wins on influence over standards and contribution to technological sovereignty, while “joining” wins on speed and start-up costs. The hybrid model wins only on the cost of exit, but it is not the worst on any criterion, and for an enterprise that has already gone through a forced change of supplier this matters a great deal. Hence a managerial conclusion that we consider central to practice: the cost of exiting an ecosystem should become a mandatory criterion when choosing an industrial platform, on a par with licence fees and the payback period. It can be estimated as the sum of the costs of data migration, reconfiguration of integrations, staff retraining and equipment downtime when the supplier is changed.

Before 2022, this parameter was, as a rule, not taken into account in investment decisions. It was precisely its underestimation that made the departure of Western suppliers so painful, and it would be strange to repeat this mistake when choosing new platforms.

5. A matrix for choosing an enterprise’s ecosystem strategy

It makes little sense to choose a strategy once for the entire enterprise, because the answer will differ from one production area to another. For practical use, we propose a matrix built on two criteria (Fig. 1).

The first criterion is the criticality of the process and data: how far a system shutdown or a data leak in a given area threatens the continuity of production, industrial safety or trade secrets. The second is the maturity of domestic solutions, namely the availability on the market of proven Russian products already deployed at comparable enterprises.

Fig. 1. Matrix for choosing an ecosystem strategy for a production area (compiled by the author)

For areas with high criticality and mature domestic solutions, a “domestic stack” is the rational choice. This means switching to Russian products while participating in the ICCs and technical committees, which gives the enterprise a say in the development of the product and of the standards. With high criticality but immature domestic solutions, a “hybrid with managed exit” is appropriate. Existing systems are kept for a transition period, interfaces are moved to open standards, and the development of a replacement solution is commissioned through the mechanism of particularly significant projects, followed by parallel operation. For non-critical areas where mature Russian products exist, an “off-the-shelf domestic solution” with minimal customisation is sufficient. Finally, for non-critical areas without domestic equivalents, “joining with exit control” is acceptable: using ready-made foreign, primarily Chinese, solutions provided that data formats are specified and documented and the cost of exit is assessed in advance.

The matrix is dynamic: as domestic solutions mature, areas shift to the right, and within a few years some “hybrid” areas may move into the “domestic stack” quadrant. It is therefore advisable to review the ecosystem strategy annually, together with the enterprise’s investment programme.

Our analysis leads to a number of recommendations for enterprises. First, include requirements for open interfaces and for data export in standard formats in technical specifications. Second, calculate the cost of exit when choosing any platform. Third, take part in the work of the ICCs and technical committees, because having a vote while a standard is being developed costs less than adapting to someone else’s rules later. Fourth, keep a single register of equipment and data described according to a standardised model, which makes any future migration easier.

As for the state, it is advisable to treat the development of industrial standards as a public good. The whole industry benefits from a common language for describing equipment and data, while the costs are borne by individual participants, so without support such standards will emerge slowly. It is also important to align national standards with international ones wherever this does not conflict with security objectives, so as not to create additional barriers for exporters. Finally, ICC selection rounds should favour solutions with open interfaces; otherwise, having escaped foreign monopolies, we risk growing our own.

Conclusion

In this article we have shown that the digital transformation of industry involves not only the introduction of technologies but also the choice of an ecosystem and standards, which determines future transaction costs, the degree of dependence and the cost of exit. International experience shows that ecosystems that shape standards set the rules for entire industries, and in 2026 European and Chinese industrial ecosystems began to converge.

After 2022, Russian industry found itself outside Western ecosystems and is now simultaneously building its own infrastructure, in the form of industrial competence centres and national standards, and switching to Chinese solutions for equipment. A comparison of the three strategies shows that for most enterprises the most robust option is the hybrid model, which combines a domestic core of critical systems with open standards at the interface level.

The proposed matrix makes it possible to choose a strategy for individual production areas, taking into account the criticality of the process and the maturity of domestic solutions. In our view, the main condition for success is to account for the cost of exiting an ecosystem in every investment decision. Without this, any new platform risks becoming, ten years from now, the same trap that Western systems turned out to be for many enterprises in 2022. The assessments in Table 1 are expert judgements, and quantifying exit costs with data from real implementations, together with piloting the matrix at manufacturing enterprises, offers ample scope for further research. In a broader sense, the question is whether Russian industry will be able not only to secure its independence but also to take part as an equal in shaping the rules by which the global digital industry will operate.

References:

1. Hanelt, A. A Systematic Review of the Literature on Digital Transformation: Insights and Implications for Strategy and Organizational Change / A. Hanelt, R. Bohnsack, D. Marz, C. Antunes Marante // Journal of Management Studies. — 2021. — Vol. 58, No. 5. — P. 1159–1197.
2. Williamson, O. E. The Economic Institutions of Capitalism / O. E. Williamson. — New York: Free Press, 1985. — 450 p.
3. Katz, M. L. Network Externalities, Competition, and Compatibility / M. L. Katz, C. Shapiro // American Economic Review. — 1985. — Vol. 75, No. 3. — P. 424–440.
4. David, P. A. Clio and the Economics of QWERTY / P. A. David // American Economic Review. — 1985. — Vol. 75, No. 2. — P. 332–337.
5. Arthur, W. B. Competing Technologies, Increasing Returns, and Lock-In by Historical Events / W. B. Arthur // The Economic Journal. — 1989. — Vol. 99, No. 394. — P. 116–131.
6. Blind, K. The impact of standards and regulation on innovation in uncertain markets / K. Blind, S. S. Petersen, C. A. F. Riillo // Research Policy. — 2017. — Vol. 46, No. 1. — P. 249–264.
7. Teece, D. J. Profiting from innovation in the digital economy: Enabling technologies, standards, and licensing models in the wireless world / D. J. Teece // Research Policy. — 2018. — Vol. 47, No. 8. — P. 1367–1387.
8. Shapiro, C. Information Rules: A Strategic Guide to the Network Economy / C. Shapiro, H. R. Varian. — Boston: Harvard Business School Press, 1999. — 352 p.
9. Adner, R. Ecosystem as Structure: An Actionable Construct for Strategy / R. Adner // Journal of Management. — 2017. — Vol. 43, No. 1. — P. 39–58.
10. Jacobides, M. G. Towards a theory of ecosystems / M. G. Jacobides, C. Cennamo, A. Gawer // Strategic Management Journal. — 2018. — Vol. 39, No. 8. — P. 2255–2276.
11. Gawer, A. Industry Platforms and Ecosystem Innovation / A. Gawer, M. A. Cusumano // Journal of Product Innovation Management. — 2014. — Vol. 31, No. 3. — P. 417–433.
12. Kleiner, G. B. Economics of ecosystems: a step into the future / G. B. Kleiner // Ekonomicheskoe vozrozhdenie Rossii [Economic Revival of Russia]. — 2019. — No. 1 (59). — P. 40–45. — (In Russ.)
13. Gileva, T. A. Assessment of the intellectual maturity of industrial ecosystems: a configurational approach / T. A. Gileva, I. A. Babkin, S. V. Zdolnikova // Vestnik Permskogo universiteta. Seriya «Ekonomika» [Perm University Herald. Economy]. — 2025. — Vol. 20, No. 4. — P. 468–490. — (In Russ.)
14. Catena-X Launches World’s First Cross-Border Automotive Industry Data Ecosystem Between China and Europe. — 22.07.2026. — URL: https://catena-x.net/news/catena-x-launches-worlds-first-cross-border-automotive-industry-data-ecosystem-between-china-and-europe/ (accessed: 30.09.2026).
15. IEC 63278–1:2023. Asset Administration Shell for industrial applications — Part 1: Asset Administration Shell structure. — Geneva: International Electrotechnical Commission, 2023.
16. The Chinese Communist Party Central Committee and the State Council Publish the “National Standardization Development Outline” / translated by the Center for Security and Emerging Technology. — 2021. — URL: https://cset.georgetown.edu/publication/the-chinese-communist-party-central-committee-and-the-state-council-publish-the-national-standardization-development-outline/ (accessed: 30.09.2026).
17. Siemens announced the termination of its business in Russia // Interfax. — 12.05.2022. — URL: https://www.interfax.ru/business/840361 (accessed: 30.09.2026). — (In Russ.)
18. The share of Russian products in the CAD market is no more than 30 %: a study // D-Russia.ru. — 13.04.2026. — URL: https://d-russia.ru/dolya-rossiyskih-produktov-na-rynke-sapr-sostavlyaet-ne-bolee-30-issledovanie.html (accessed: 30.09.2026). — (In Russ.)
19. Up to 80 % of industrial robots in Russia remain imported: the main supplier in 2025 is China // RB.RU. — 2026. — URL: https://rb.ru/news/do-80-promyshlennyh-robotov-v-rossii-ostayutsya-importnymi-glavnyj-postavshik-v-2025-godu-kitaj/ (accessed: 30.09.2026). — (In Russ.)
20. Year of accelerated robotisation: express information / Institute for Statistical Studies and Economics of Knowledge, HSE University. — Moscow, 2025. — URL: https://issek.hse.ru/news/1053851356.html (accessed: 30.09.2026). — (In Russ.)
21. 35 industrial competence centres formed to replace foreign “industry digital products” // D-Russia.ru. — 11.07.2022. — URL: https://d-russia.ru/sformirovany-35-industrialnyh-centrov-kompetencij-dlja-zameshhenija-zarubezhnyh-otraslevyh-cifrovyh-produktov-pravitelstvo.html (accessed: 30.09.2026). — (In Russ.)
22. 120 particularly significant projects completed // ComNews. — 28.04.2026. — URL: https://www.comnews.ru/content/245004/2026–04–28/2026-w18/1008/zavershilis-120-osobo-znachimykh-proektov (accessed: 30.09.2026). — (In Russ.)
23. On the national development goals of the Russian Federation for the period up to 2030 and for the future up to 2036: Decree of the President of the Russian Federation of 07.05.2024 No. 309. — URL: http://www.kremlin.ru/acts/bank/50542 (accessed: 30.09.2026). — (In Russ.)
24. GOST R 57700.37–2021. Computer models and simulation. Digital twins of products. General provisions: national standard of the Russian Federation: approved and put into effect by Order of Rosstandart of 16.09.2021 No. 979-st. — URL: https://base.garant.ru/403019790/ (accessed: 30.09.2026). — (In Russ.)
25. Digital factories have received a standard // ComNews. — 15.02.2024. — URL: https://www.comnews.ru/content/231591/2024–02–15/2024-w07/1008/cifrovye-fabriki-poluchili-standart (accessed: 30.09.2026). — (In Russ.)