Digitalization changes organizations. Data-driven companies not only consider how to improve operational and management processes using data (analysis), but they also think about data products that are delivered as a service to customers. In this blog post, we will explain what data products and data as a service are, give some examples of different kinds of data products, and also highlight key challenges to tackle when designing and offering data products.
What is a data product?
Data analysis aims to retrieve information from data to support or enable decisions. Relevant pieces of information are identified and delivered via business intelligence tools or applications on computers and mobile devices. Which pieces of information are relevant depends on the business process that is being supported. That is where domain knowledge comes in.
Data analysis can be carried out using business intelligence tools for ad-hoc reporting or OLAP analysis, or advanced analytical methods can be employed. Providing for reporting and data analysis services on a defined set of data in a software products, we get a simple data product. Generally speaking, data products are a new category of services powered by analytics, business domain knowledge, and scalable IT architectures.
Figure 1: Examples for data products and required skills, adapted from Harris
There are several categories of data products. A simple one is reports, where data is descriptively displayed and made transparent to the user. A typical example would be reports for telephone or credit card bills that are provided to customers and create a value-add by providing more timely information and additional analysis possibilities. A good example is benchmarking, another type of data product. Here, a particular behavior or KPI is compared across industries or users. Social networking websites such as ResearchGate and LinkedIn create engagement with users by displaying information about how they score compared to other users in terms of visibility and popularity.
Data products get even more interesting when advanced analytics comes into play, which is the case with recommendations and forecasts. Examples of recommendations would be a social network suggesting you connect with other people or an e-commerce site making product suggestions. At Airbnb, hosts can even use an intelligent price-setting feature that adjusts the price of their property according to current market demand. This provides customers with a more efficient way of using the Airbnb platform, and Airbnb benefits from more business through better pricing. Also, forecasts provide valuable information to act upon, enhancing existing (physical and data) products. Predictive maintenance is a good example of how data powers a value-added service for service organizations, e.g. automotive or aerospace companies. Another example is AmTrust, a business insurance company, which embeds weather forecasts in its products to improve the crop insurance it provides.
Consulting and managed services can also be offered around data products. A good example of additional services would be data management services, such as data integration, data quality management, and data storage to help companies integrate and manage the data they need for analysis.
Challenges in implementing data products
Data products allow organizations to capitalize on data in new ways. However, not many companies have established business models based on data products yet. There are some important topics to consider when setting up data products:
1. Business model
Data products require a business model to determine how users will benefit from the service provided and how the value from data products and services will be appropriated. There are many models to capitalize on the value of data and the services based on data. Which one is best depends on the type of service provided, whether it is related to a platform or a product, and how the customer benefits. Examples are the freemium model, where users are offered part of a service for free but are charged for upgrading to the full service, or charging a premium for additional data services with an existing product.
2. Marketing and sales
There is also a marketing and sales aspect to consider since new – and maybe even disruptive – services need to be promoted, positioned, priced, and sold. Challenges to be tackled arise from new competitors and also perhaps a lack of internal know-how and experience about how to market and sell such solutions and services.
3. Delivery of data products
Delivering data products as a service means to provide them on demand, permanently, scalably, and securely. A user interface is often realized via an app or a Web interface, and the whole service is often based on a cloud-based infrastructure, platforms, and applications. Organizations that are typically not in the software industry need to act like software companies when delivering data products. The software development lifecycle, support, and operationalization of data products need to be managed and provided to the customers of the service.
4. Data management
A key capability when providing data products is handling data consistently and securely. This requires data governance concepts with specific attention to data security and privacy. Complying with constantly changing and heterogeneous data privacy legislation in different countries is one of the biggest concerns for companies that are successfully using data services and products. Special care needs to be taken when handling customers’ personal data (e.g., by anonymizing it).
Establishing data products is certainly a very innovative way to use and monetize data, but also a complex one. It requires organizations to become truly digital – in their products and their business models.
- BARC Survey “Big Data Use Cases. Getting real on data monetization”
- BARC Research white paper “The Benefits of an Integrated Approach to Business Intelligence and Planning,” detailing findings from its recent study.
This post was co-authored by Dr. Sebastian Derwisch, Data Scientist at the Business Application Research Center (BARC). Sebastian holds a PhD in Economics and has extensive experience advising companies in the areas of use case identification for data analytics, tool selection for advanced analytics and the organization of data science teams. He can be reached at firstname.lastname@example.org
Image: Steve Wilson via Flickr