Value of data
Transform data as a by-product into data-fueled value streams
Up until recent, data — in most cases — is perceived as by-product. With the potential to deliver (new) insights. This is the case for e.g., IoT sensor data which is mainly used for process automation & preventive maintenance within manufacturing or healthcare. A secondary stream of data usage. Companies with an open mind for opportunities are now realizing that by embedding these IoT sensors across product lines, manufacturing facilities or patient journeys, the generated data — and the services it can deliver — can become a strategic asset.
Any company should ask the following: Does the company generate lots of data? Or is it a data rich company that produces and/or maintains market leading products?
Answering this question requires a real-data driven mindset. There is no doubt that the value of preventive maintenance is valid. Yet, it is not scalable enough to become that data rich company. And if the investment for (IoT) data can be returned in multiples, companies should be creative and look for opportunities beyond just predicting equipment issues and maintenance requirements.
Don’t just act data-driven, be it! Look beyond the initial goal of the IoT sensors and the data that is generated. Get at the research & business chair. And drive new products and services. Data initially generated by IoT* for automation, can provide new insights in e.g. live performance of products, providing customers with valuable services. A good example is the Rolls Royce IntelligentEngine. The IntelligentEngine IoT sensor data — aggregated and analyzed in the cloud — is providing Rolls-Royce with unprecedented insights into the live performance of its machinery. A modern passenger jet generates an average of 500GB of data per flight(!) and several terabytes on long-haul routes. The thousands of sensors in each Rolls-Royce engine track everything from fuel flow, pressure and temperature to the aircraft’s altitude, speed, and air temperature, with data instantly fed back to Rolls-Royce operational centers. The company’s aircraft availability center is continuously monitoring data from 4,500 in-service engines. Rolls-Royce can tap into a — cloud based — ecosystem of small, specialist 3rd parties to analyze different parts of available data. And that data capability is rapidly evolving to providing customers with valuable aftermarket services that range from showing airlines how to optimize their routes to keeping a survey ship in position in heavy seas.
Next to this, the company recently launched another data-centric initiative; R2 Data Labs. This act as an acceleration hub for data innovation. “Using advanced data analytics, industrial artificial intelligence and machine-learning techniques, R2 Data Labs will develop data applications that unlock design, manufacturing and operational efficiencies within Rolls-Royce, and creates new service propositions for customers,”.
Combining data & analytics into potential new value streams is often a first step within companies that are aiming to become more data driven. In practice, this can bog down into analytics MVP without generating a scalable product or services which is understood, carried and supported by business stakeholders, account management and sales. This connection needs to be in place to be successful [ link naar andere artikel MV]. The R2 datalab shows what additional circumstances and behavior is needed for success. At its heart, Data Innovation Cells will comprise experts drawn from multiple disciplines across the company and apply cutting-edge DevOps principles to rapidly explore data, test new ideas, and turn those into new innovation and services. In other words, successful data products & services comprise of a fusion of the following elements: a technological (ecosystem) environment, sufficient internal & external data, understanding the value of data, trusted data & analytics, an understanding existing & new markets and product development.
Rolls Royce had the advantage of having available data, sufficient circumstances as well as a very good market understanding. They were part of a mature market and ‘just had to tap into’ data to deliver relevant new products.
There are multiple examples of data driven value streams within existing and upcoming companies. You should check e.g., Google Spin-off Examples of The Climate Corporation / ClimateEngine, that enable crop insurances based on satellite data or Twiga Foods that introduces smart crates with tags for real-time data collection, thereby enabling food distribution in Kenya. Examples where data is driving new value models. First Access and Tala use data from mobile phones to provide alternative credit scoring services that help financial services providers assess the risk of people at the base of the pyramid. BBOXX has developed the Bboxx Pulse® platform which harnesses remote (real-time payment) monitoring data and IoT data to deliver energy access in a scalable and distributed model.
The above mentioned examples show how data can be monetized via different value models. By comparing monetization models and determining which is best suited data value offerings, companies can increase revenue margins and introduce beneficial new products and features, while expanding customer relationships and delivering specific value that keeps your clients coming back for more.
1. Perpetual model: the traditional model where customer pay for a product once upfront. And then have a perpetual rite to use the product (e.g. the raw data, aggregated data, meta data or insights). The seller has full responsibility for upkeep and updates.
2. Subscription (‘as a service’) model: here; Data-as-a-Service or DaaS. The customer buys service subscriptions to access of the data, right to use the data and updates & support. Advantage is that DaaS provides a predictable revenue stream that can be projected into the future. This revenue predictability has made the subscription model — similar to the software industry — increasingly popular.
3. Usage model: In this leading-edge monetization model, customers pay providers based on specified usage metrics ( e.g., pay per tick, when data is needed real time. Number of tests performed per data set. Or another data-related (technically it is software) example is the data cloud providers metric where customers are charged based on terabytes of data storage) with periodical invoicing. A customer pays for what they use.
4. Outcome model: A quite leading-edge and interesting model. Suppliers are not selling data products or services, they’re selling an outcome. Something like those law firm commercials that say, “We don’t get paid unless you see cash,” this model is about achieving a defined business result rather than delivering individual IoT data. This model is used by Rolls Royce and Climate (now Monsanto).
5. Impact model: Companies, e.g., some the largest agricultural companies in the world are collaborating to help eradicate Malaria by 2040. This is funded by donors. Some business models can create value by measuring social impact and reporting it to relevant bodies, such as government departments, donors and impact investors. This enables companies to think beyond data business models. Additional value streams for this could be generated by payments as grants or in ‘pay by results’ schemes.
*Note, this article has examples of IoT sensor data, which mostly does not include privacy related data. However, for all data monetization efforts it is applicable that they always must be performed within the requirements of the data privacy regulations in force as well as companies’ ethical guidelines.