Data is increasingly becoming an article of trade or commerce - in short, a product. The era of data is about the process of data commoditization, where data is becoming an independently valuable asset that is freely available on the market. A “commodity” is defined as something useful that can be turned to commercial or other advantage.
Examples of traditional commodities are Grains, Gold, Beef, Oil and Natural gas. A commodity is a basic good used in commerce that is interchangeable with other commodities of the same type. Commodities are most often used as inputs in the production of other goods or services.
Open data is the strong manifestation of this new era. The government mandates and open data policies from multiple countries and public entities continue to contribute to the process of data commoditization. Openness has the benefit of increasing the size of the market. The greater the size of the market and the demand for a resource, the greater the competitive pressure on price and, hence, the increase in commoditization of the resource.
Open data is mostly about sharing and access to data. After the hype around open data, monetization of data has emerged as one of the most discussed topics, which is visible also in academic research. Emerging data economy markets are not fully yet here yet. Instead we are still in the phase of learning to utilize and monetize data - both require data products.
5 categories
One way to see data product types is to divide them into 5 categories: static data products, rendered data products, dynamic data products, low-code no-code data products and functional data products.
Example of Static data products is dataset which is the most common sharing format in open data. The datasets are static and are sometimes updated occasionally. Selling datasets has been there also for a long time. Example of commercial dataset is company contact informations. Different kind of reports and documents go to the same category.
Rendered data products offer dynamic often visualized and formatted view to the data product content. Also data products which are compatible with various digital twin solutions and game engines such as Unreal as well as map service compatible data products fall into this category. These products often resemble services.
Dynamic data products are data streams driven and APIs driven. These can be divided to two subcategories. The first one is request driven data products, which expect data product consumer to be active and request content by invoking API calls. The second type data product is subscription, in which data product consumer subscribes to the data product and after that is offered changed information automatically without invoking API calls. Either one of these can have predefined or elastic content. In predefined data product owner has locked the content and in elastic data product data consumer can select subset of the content.
Low-code slash no-code platform data products are out of the box compatible with services like Zapier, If this then that and Microsoft Powerapps to mention a few.
Functional data products are APIs and algorithms. Examples are algorithms for data mining, matching, cleansing, relevance computation and lineage tracing. Developers may upload these algorithms to a data marketplace as a black box user-defined- function, so other participants of the data marketplace may ’buy’ and try out these algorithms. The other subcategory is bidirectional data products which are intended to give control commands or alike to data source system instead of just getting information for application.
Commentaires