Data-driven business


Data-driven business means making money through collecting, aggregating, analysing and (re)selling data.


Why consider it?

Data are omnipresent nowadays and are an invaluable asset for any company to exploit. More and more people and machines generate ever more data, and sophisticated technologies for storing, analysing and extracting knowledge from these data are freely available and rapidly maturing. While there are many ways that companies can use data analytics internally to gain insights and improve processes, or add value to products and services, companies should ask themselves whether the data they own have value to other organisations within or outside the industry. 

Data are a key resource for data-driven business models. These can be (a combination of) data already owned by the company that may not be fully exploited, data specifically collected for this purpose, data purchased from other sources or freely available data (open data, web-crawled data).

Companies selling data, such as financial and economic data, existed in the past but collecting and structuring these data often entailed a lot of human effort. New business models can now be devised by aggregating, analysing, visualising and delivering multiple data sources, thus creating new services for existing or new customer segments. This can be done by collecting and combining data in one domain (e.g. financial, mobility, health) and offering insights based on the data, or by using data to make a match between client and provider (e.g. travel, dating sites). Value can be captured by means of a subscription or marketplace model, classical licences and so forth.    

The main reason for developing data-driven business models is to unlock extra value in existing data assets or data assets that can be generated by the company. These data-intensive models are harder to copy and their value grows over time by adding more data and gaining new insights.

What does it involve?

To create data-driven business models, companies need to overcome various challenges and master new skills and capabilities:

  • Firstly, being able to assess which data is available and whether the quality and volume is high enough to build on, identifying which areas will be the most beneficial, and being able to apply the overall business model patterns or success stories to the specific product/service/domain.
  • Integrating data from many different heterogeneous sources into a single comprehensive whole. This requires dealing with technological challenges, such as different ways to access the data, different data formats that need to be unified, semantics, deduplication, and so forth. Non-technological aspects need to be considered as well, such as the provenance, correctness, trustworthiness and recentness of the data.
  • Enriching the (integrated) data with additional insights and knowledge by applying intelligent data-mining algorithms. This requires in-depth expertise in data science to gain an understanding of the data's potential, derive discriminative features, select a suitable algorithm, correctly interpret the results and objectively validate the performance.
  • Facilitating the (semi-)automation of the entire process as much as possible, from data gathering to data enrichment through to data distribution, so that a representative data set can easily be kept up to date and so that not only the most recent but also the most relevant data for a given context is always available.
  • Making the data available through methods aligned with the target business model, such as one-off download without updates, rate-limited API, and so on.
  • Creating a legal framework that includes diverse aspects such as 1) privacy, e.g. conforming to the GDPR, 2) IP, e.g. appropriately identifying and taking care of ownership, 3) liability, e.g. what if (parts of) the data are incorrect, and so on.

Further information
Agoria can assist you with questions on legal aspects and standardisation


Contact person: