Entrepreneurship, Complexity, and Data Science: Part II

Business orchestration is a model that arises from the interaction between peers by connecting those who offer a product or service with consumers. Take, for example, businesses like Uber and Airbnb, which do not own material goods, but rather provide a secure platform where users - both sellers and buyers - interact with each other. It differs from traditional models because these companies offer not the product itself but the interaction.

How does an orchestration business model generate its first thousand clients?

This type of business begins by filling the space of the offer on the platform to attract the first users. From here on, the interesting thing begins, since the value proposition of the orchestration business is based on the interactions, so it adapts to the clients' needs. Of course, the examples we have mentioned are large-scale transnational corporations, but they were not always so.

The so-called "early adopters" are the most adventurous. They are willing to try a new service and therefore forgive a longer wait time for a bottle of water, security, and a clean car. They will even recommend the platform to their friends and family. The early adopters market is how an orchestrator gets its first 1,000 clients. From then on, there are several options to grow.

One of them is traditional marketing campaigns, which certainly help, but the most lucrative business is listening to consumers and adapting.

In the case of Uber, in addition to increasing the number of users, sales increased with the diversification of the services that were already offered. For example, with carpooling. In addition, a new market was captured with UberEats, and higher sales were achieved from active users. Even with the pandemic generated by the spread of the SARS-CoV-2 virus, it was possible to evolve the business model by introducing home delivery and grocery stores within the platform.

What makes a business model evolve?

According to economic theories (for example, that of Joseph Schumpeter) innovation is key to economic growth. In this sense, understanding the way in which business models evolve becomes more critical. In the orchestration business, analyzing the massive amount of generated data is crucial for the scalability of the model.

Today, big data provides an emerging environment of innovation, tools and methods that allows interaction between companies and users. Massive interaction can optimize the innovation process and increase the degree of digitization of user behaviour, but not only that. Data analysis allows companies such as Uber and Airbnb to build an environment of continuous improvement and guide and maintain the ecology of innovation (Garousi & Mäntylä, 2016). According to Y. Zhang et al. (2017), the development and innovation of products based on data have generated qualitative changes in the iteration of business models. Product feedback, which used to be weekly and monthly, can now be done much faster. Fast feedback enables better satisfaction of customer demand with innovative solutions (H. Zhang et al., 2017).

How fast can a business model based on big data evolve?

From the perspective of adaptability, Ghasemaghaei et al. (2017) point out that combining big data, research and product development (R&D) and innovation through data mining allows adjusting the direction, structure, process and strategy of a business at any time. Likewise, using big data strategies, companies can promote the iterative evolution of the innovation system and improve its efficiency (Yuxi & Mengjie, 2020).

In terms of the process innovation of the enterprise business model based on big data, it is believed that big data has become as important a production tool as oil. According to estimates, the reallocation of resources through the application of big data can double the liquidity of a business. Against this backdrop, it is no surprise that orchestrators are among the most lucrative business models today.

How does a business model die?

We have already said that orchestration businesses transform data into continuous feedback that fuels adaptability. Bacterial pathogens, for example, apply innovative adaptive strategies to evade and counter host defenses. Such is the case with rapid genome evolution, which allows bacteria to rapidly alter their antigenic epitopes on short time scales to evade immunological recognition and thus avoid expulsion from the infected organism. Thus, its evolution is different from that observed in the organic world.

However, how can we predict the way a business model evolves? Is it possible to replicate its growth? We can think of time series, adaptability, and neural network models to solve this problem. Time-series methods are generally used to model predictions when there is not much information about the underlying variable generation process and when other capabilities do not explain the studied variable clearly (Z. Zhang & Trafalis, 2013). In addition, forecasting from time series models is used to predict the future based on historical observations (Makridakis et al., 1998).

Recently, models based on artificial neural networks have been proposed as an alternative to time series forecasting. Models based on neural networks can be a valuable tool for modeling and predicting time series since an artificial neural network is a universal function approximator capable of mapping any linear or non-linear function ( Cybenko, 1989; Funahashi, 1989).

Neural networks are a data-driven method with few high-priority assumptions about the underlying models. Instead, they let the data speak for itself and can identify the underlying functional relationship between them. Furthermore, the artificial neural network can tolerate the presence of chaotic components and is, therefore, better than most methods (Masters, 1995). This ability is vital as business orchestrators derive data from social interactions, and social interactions, like many time series, have significant chaotic components.

Conclusion: the adaptive evolution of the physical to the social

The mathematical study of business models, especially orchestrators, can lead us to know the answer to questions still foreign to us today. How does a business model die? Is there a scale on which it is impossible to maintain the scale of the model? Why are they so difficult to reproduce?

We will undoubtedly see progress in this regard in the coming years.

References

• Cybenko, G. (1989). Approximation by superpositions of a sigmoidal function. Mathematics of Control, Signals, and Systems, 2(4), 303–314. https://doi.org/10.1007/bf02551274
• Funahashi, K. I. (1989). On the approximate realization of continuous mappings by neural networks. Neural Networks, 2(3), 183–192. https://doi.org/10.1016/0893-6080(89)90003-8
• Garousi, V., & Mäntylä, M. V. (2016). When and what to automate in software testing? A multi-vocal literature review. Information and Software Technology, 76, 92–117. https://doi.org/10.1016/j.infsof.2016.04.015
• Ghasemaghaei, M., Hassanein, K., & Turel, O. (2017). Increasing firm agility through the use of data analytics: The role of fit. Decision Support Systems, 101, 95–105. https://doi.org/10.1016/j.dss.2017.06.004
• Makridakis, S. G., Wheelwright, S. C., & Hyndman, R. J. (1998). Forecasting. Wiley.
• Masters, T. (1995). Advanced Algorithms for Neural Networks. Wiley.
• Yuxi, Z., & Mengjie, S. (2020). Research on Data-Driven Business Model Innovation. Proceedings of the 2020 5th International Conference on Modern Management and Education Technology (MMET 2020). Published. https://doi.org/10.2991/assehr.k.201023.043
• Zhang, H., Ou, A. Y., Tsui, A. S., & Wang, H. (2017). CEO humility, narcissism and firm innovation: A paradox perspective on CEO traits. The Leadership Quarterly, 28(5), 585–604. https://doi.org/10.1016/j.leaqua.2017.01.003
• Zhang, Y., Moe, W. W., & Schweidel, D. A. (2017). Modeling the role of message content and influencers in social media rebroadcasting. International Journal of Research in Marketing, 34(1), 100–119. https://doi.org/10.1016/j.ijresmar.2016.07.003
• Zhang, Z., & Trafalis, T. B. (2013). Time-series Analysis for Detecting Structure Changes and Suspicious Accounting Activities in Public Software Companies. Procedia Computer Science, 20, 466–471. https://doi.org/10.1016/j.procs.2013.09.304

Adapted from:

Valdés-Porras, E. A. (2021). Negocios, complejidad y ciencia de datos: Parte II. En El Analista Económico-Financiero, 26/08/2021. Recuperado de https://elanalistaeconomicofinanciero.blogspot.com/2021/08/negocios-complejidad-y-ciencia-de-datos.html