The 5C's for Building Data Products
Chris Andrassy
Supply Chain AI Enablement | CEO at Astral Insights | Keynote Speaker | Early-stage Investor
As data & analytics leaders look for new avenues to monetize data, there are several key elements to consider when building data products for internal and external distribution. In this article, written by Astral Insights senior advisor Prashanth H Southekal, PhD, MBA , we discuss a common sense framework to maximize the return on investment from your next data product.
Today many companies are leveraging data for improved business results. Companies that primarily deal in data—Google, Amazon, Uber and more—are among the most valuable in the world, both in terms of market capitalization and innovation. One of the key reasons for their success is solutions fueled by data and analytics to solve pressing business problems. Typical examples of data products include Google search, Bloomberg terminal, Netflix recommendations and more.
But what exactly is a data product? Fundamentally, a data product is an outcome of the data and analytics activity to generate new revenue sources, enhance customer service, improve business efficiency and offer new solutions to problems that span across the industry.
Traditionally, organizations have used data internally for three main purposes: operations, compliance and decision-making. But given the enormous amount of data organizations have historically captured and are generating, enterprises today are beginning to explore the opportunity to leverage their data to generate new revenue streams and improve internal productivity using data products. In 2021, data was a?$231 billion ?industry, and?Gartner ?says data products hold the key to leveraging data for improved business results.
Fundamentally, data products are valuable as they solve a specific problem for the business and the industry. But how can an organization build data products? Where does one start? What are the building blocks? While every company is different, there are still some fundamental and common components required to build a data product. In this backdrop, building data products involves managing the 5Cs: Collection, Consumption, Channels, Compliance and Commercialization. The following sections look at the salient capabilities organizations need within the 5Cs to successfully build a data product.
1. Collection
In a typical enterprise, data is captured and stored in various formats, types, systems and so on. Over 80% of the data in business is in?unstructured ?formats like text, images, audio and video. Unstructured data comes without a standard data model and data type, thereby affecting the capabilities to query and process it efficiently. Hence, data in data products should be transformed to have a standard data model and data type (nominal, ordinal and continuous). In simple words, data in the data products have to be collected or reformatted consistently so that it is of high quality so that the data can be easily queried and processed for deriving the insights. The insights could come from descriptive, predictive and prescriptive analytics algorithms where the output is presented in reports, visuals and dashboards.
2. Consumption
Data products are valuable only when they are consumed to enhance business performance (i.e., data products should facilitate the creation of transactional documents, which are considered to be the first-party data by the enterprise). Transactional data is important as it?holds the key ?to improved business performance in operations, regulations and decision-making. While this transactional data is often considered as second-party or third-party data in data products, it should be in the most granular form to create optimal value in a data product. Overall, the more granular or detailed the data, the more precise and accurate analytics can be performed.
领英推荐
3. Channels
Middleware or channels deliver data and insights to the consumers—either synchronously or asynchronously. If the data and insights need to be delivered synchronously, then APIs (application programming interface) can be leveraged. However, if the need is to deliver data asynchronously, other data integration mechanisms such as ETL (extract transfer and load), EAI (enterprise application integration) and more could be considered. The selection of the integration method in channel or middleware rests on three key factors:
A) Pull vs. push; that is, whether it is the sender (data provider/data product) or receiver (data consumer) who takes the initiative for data consumption.
B) The volume and velocity of data to be integrated.
C) The sequence of data transfer, transpose and orchestration.
4. Compliance
Given that data is a valuable business asset, the sharing of data in a data product needs to comply with the organization’s data protection and data ethics rules, especially on privacy, security and so on. In other words, building a data product should not adversely affect the competitive advantage of the firm. The data compliance issues in a data product can be addressed using data-manipulating techniques such as encryption, anonymization, scrambling, tokenizing and masking, which will potentially address the sensitivities associated with the data. Overall, if data has to be monetized in a data product, striking a balance between revenue generation opportunities and the organization’s data compliance mandates is essential.
5. Commercialization
The data product has practically no utility if it cannot be taken to the market to improve business performance. This involves building the eco-system with the right pricing—aka willingness-to-pay (WTP). The ecosystem is important if data products have to collaborate with other products in the value chain to offer solutions holistically. This encompasses creating a strong value proposition by looking at the entire value chain holistically and identifying value leakages. Value leakages typically happen when there is a hand-over or transition from one value stream element. The value created is determined by the market’s WTP (i.e., the highest point the market goes to buy the data product).
Today, every company intends to leverage data for improved business performance with data products, focusing on data monetization. At the core, data products and data monetization are not just making money from data, but an innovative way of enhancing business productivity. The 5Cs discussed here will help organizations build data products for creating a sustainable competitive advantage. The benefit of a data product include new revenue streams and achieving a quicker time-to-market, reducing the cost of business operations and minimizing business risk.
About the Author
Prashanth H Southekal, PhD, MBA is the Managing Principal of DBP Institute and senior advisor at Astral Insights . He is a Consultant, Author, and Professor. He has consulted for over 75 organizations including P&G, GE, Shell, Apple, and SAP. Dr. Southekal is the author of two books — “Data for Business Performance” and "Analytics Best Practices” — and writes regularly on data, analytics, and machine learning in Forbes.com, FP&A Trends, and CFO.University. ANALYTICS BEST PRACTICES is in the top 100 analytics books of all time and in May 2022 was ranked #1 by BookAuthority. Apart from his consulting pursuits, he has trained over 3,000 professionals worldwide in Data and Analytics. Dr. Southekal is also an Adjunct Professor of Data and Analytics at IE Business School (Madrid, Spain). CDO Magazine included him in the top 75 global academic data leaders of 2022. He holds a Ph.D. from ESC Lille (FR) and an MBA from Kellogg School of Management (U.S.). He lives in Calgary, Canada with his wife, two children, and a high-energy Goldendoodle dog. Outside work, he loves juggling and cricket.
Lecturer & financial consultant
1 年Excellent article. P&L: Data in unstructured format affect the ROI. Balance sheet: The non used data become a liability rather than an asset. Cash flow: the 5 C's stop the financial bleeding.