The phrase, “Data is the New Oil” is widely used but how are companies actually using data to create new and innovative business models? This post explores some of the ways data is being commercially exploited and explains what companies should do which are considering developing data-driven business models.
Google and Facebook have shown how the data which flows across the internet can be organised and repackaged to create billions of dollars of value. While these giants are in a league of their own, there are a growing number of much smaller startups as well as older, established companies which are using data in innovative and, potentially, very profitable ways.
While developing a better search engine or social network may not be the best business strategy at this stage of the internet’s evolution, creating new sources of data, developing services and technologies to organise and analyse it as well as repackaging existing data sources all have the potential to base a successful business model on. Recent research from the University of Cambridge and the University of Oxford has demonstrated a range of strategies which companies are adopting when building data-driven business models. One of the key findings from these research projects was the importance of having data tightly integrated into the core business operations rather than thinking of commercializing data as an afterthought.
New Sources of Data
Many of the devices which make up the Internet of Things (IoT) and the apps we use on our phones are throwing off data at increasing speed and volume. There are also the systems being deployed to streamline industrial processes which often involves fitting sensors to machinery and plant. The data collected from these technologies is not only leading to a better understanding of ways to improve operating efficiencies but also to create new services and revenue streams. Being able to add value to physical products through the collection and analysis of operating data is helping companies like Rolls Royce, Caterpillar, GE and Bosch evolve into service providers. In the last 2 years Bosch has invested more than $300 million in developing artificial intelligence (AI) services to complement its service and product offerings to customers. As a business model development strategy, owning unique, proprietary data sets provides a significant competitive advantage and makes it more difficult for competitors to offer rival services.
New Ways to Manage Data
Just as Capitol One used a highly successful and innovative “Information-Based Strategy” in the 1990s to differentiate their credit card offerings from larger competitors so insurers in 2018 are developing new ways to better understand consumer data. This is leading to insurance product offerings being developed for very precise groups rather than the more traditional “one size fits all” approach. Goji in the United States is using machine learning techniques and proprietary analytics to better target customers but also to offer more efficient sales and service processes. According to financial analysts CB Insights, Goji raised $110 million of funding in 2017 to develop its new service offerings. A business model based on exploiting software developed in-house and which allows unique insights into large customer datasets can provide a highly defensive strategy for profitable revenue growth.
Evolving Business Models for Established Companies & Data-Driven Startups
Business models based solely on the efficient production and sale of physical products are becoming increasingly difficult to profitably sustain. The automobile industry discovered this decades ago and turned to also offering financial services to help customers purchase their cars as well as provide profits from interest payments and service charges. As technology becomes increasingly embedded in domestic and industrial products, new sources of value and, as a result, business models are emerging. A growing trend for companies exploiting data from their established product and service offerings is to build platforms that link up with customers and third parties to offer value-added services and which are able to exploit network effects to drive further growth and competitive strength. GE is doing this with its Predix software platform for the collection and analysis of data from industrial equipment and Airbnb has done that with its room booking service linking property owners with renters. Earlier this year, Airbnb was valued at $31 billion, $7 billion more than Hilton Hotels Corp. Key to the company’s success is the platform it has built that not only provides a 2-sided market to match buyers and sellers but also offers unique information about the ratings of renters and property owners. As eBay and Uber have discovered, such platforms make it extremely difficult for competitors to develop rival products. This is not to say, it can’t be done (of course it can) but it would require a business model which substantially reconfigures the value proposition for at least one side of the market.
Lessons in Best Practice for Business Model Development
My academic research over the previous several years has a number of lessons for anyone thinking of developing a data-driven business model (obviously these are not all relevant to all companies):
-
build a defensive strategy into the business model. This might be through the development of proprietary software which is hard to replicate or the exploitation of network effects;
-
if possible develop or have access to unique data which would be difficult for others to access or create;
-
make sure your company has the skills to manipulate and organise the data;
-
establish what the core value proposition(s) is in relation to data-driven services;
-
work with partners and customers to add value to data, perhaps through the development of a platform for linking complementary assets. The recently launched data exchange, Terbine, is a good example of this.
Where It Is All Going
Predicting how technology will shape the business environment is generally a fools errand but there are some broad trends which seem likely to persist and which anyone building a data-driven business model should bear in mind:
-
AI and machine learning will continue to alter the economics of product and service development for the immediate future. Being able to exploit software to better understand customer needs and market conditions will provide a significant competitive advantage over those still doing business based on previous trends and gut instinct;
-
Working with customers and suppliers to provide added value to all parties will be important as low friction platforms reduce the barriers to communications and information sharing;
-
Just as the application of software has transformed business processes and value creation and capture over the last 20 years, so the combination of this with data collection and manipulation will do the same over the coming years. Companies with business models that do not acknowledge and this and evolve accordingly will go the way of Kodak, Encyclopedia Brittannica and Blockbuster.
