Data-driven digital services are everywhere. From health and fitness, to agriculture and education, digital technologies and intelligent apps continue to reshape our world.
Digital has transformed industry, too. Companies are increasingly reliant on the wealth of data available to them (and the insights contained therein) to provide their customers with more personalized services. It's also helped to level the playing field. Today, start-ups use data obtained from digital channels to capture significant market share from industry giants.
But longstanding product companies have a crucial advantage: unparalleled knowledge of how their products perform in the real world.
Leveraging digital R&D to offer unique insights
Established product companies have years of specialized digital R&D under their belts. Leveraging this data to create new digital services, they can provide customers with unique insights, greater personalization, and improved value.
Imagine a new skin care subscription service, for example. A start-up can develop a model to track preferences, optimize delivery, and suggest similar products. If they're smart, they can even refine recommendations by broad customer groups using publicly available data on skin types or even weather conditions. But this data is easily accessible – essentially anyone can do this.
Established skin care brands, on the other hand, have exclusive access to vast amounts of digital R&D data. This contains detailed secrets on absorption, scent, and other key factors. They can also see the relationship between these factors and a wide variety of skin types and environments. These specialized models and data can then be used to offer highly personalized skin care recommendations that consider the confluence of skin tone, dryness, age, diet, local air conditions, time of year, and so on.
To do this right and create a truly intelligent app or digital service, they rely on sophisticated models. These are based on the underlying science of how different skin types respond to different products in different environments. This is not measuring clicks and spotting correlations. It's modeling fundamental chemistry and biology, using complex digital R&D data combined with real-world environmental data.
Correlations from customer preference data will always be statistical approximations of what's actually happening. But with scientific data, we can model reality itself and use this as a foundation to build digital services that deliver much deeper insights.
Examples of digital services and intelligent apps built using digital R&D data
The approach described above is already happening. Let's take a quick look at some examples of intelligent apps and services built around digital R&D.
AkzoNobel (makers of paints and coatings) has an app that allows customers to visualize color in a highly accurate way. They can then make accurate, informed decisions, without the need for product samples. Underpinning this are models built around the science of color and light, which in turn enables the app to visualize the right tone for specific situations. It takes into account everything from lighting conditions to the resolution of the user's phone screen, to show customers exactly what the final color will look like.
Mars’s Pet Care division has developed a DNA profiling tool for pets, built around models of genetics, bioinformatics of gut biome, and food chemistry. This allows them to make personalized recommendations of diets aligned to optimal pet health, which helps them sell more pet food. More importantly, it enables them to move from selling tins of food to selling pet health plans that leverage custom diets.
3) AkzoNobel (marine coatings division)
AkzoNobel’s marine coatings division developed Intertrac Vision, a tool which shows how different hull coatings affect fuel efficiency. The underlying models were based on biofouling (accumulation of microorganisms) risk data they'd collected from oceans around the world, as well as experimental data about the performance of their products.
Why build data-driven services on scientific models?
Most companies have embraced the physical-digital convergence. Associating intelligent apps with products make them more desirable, while opening an important channel of communication that brings companies closer to their customers.
As apps improve, differentiating the digital aspect of your product may become as significant as the physical aspect. Models that use scientific data is the logical next step for data-driven services. It enables product companies to provide more personalized recommendations and is an area where R&D-focussed companies are in a unique position to benefit.
There are longer-term strategic advantages to consider as well. As people upload photos of their skin, their pet’s DNA etc., this information can be fed into future R&D initiatives. This helps shape both physical and digital product design.
The growing importance of digital R&D
The mere existence of data does not provide all the answers – companies must also use it effectively. The better organized your data and models are, the easier it is to identify valuable research that you can exploit in digital services. This underlines the increasing importance of digital R&D.
Harnessing digital R&D data to build digital services means setting up data stores, lakes, and warehouses so that data is available to anyone who needs it. This includes selecting tools and building integrators that pipe data directly to data science teams. It means agreeing standard processes for capturing and labeling data that makes it easier to understand.
Think like a start-up
Unlike new products (which require lab research, mass manufacturing, and stock management), digital services can be brought to market quickly and iteratively. For product companies, this may mean embracing new ways of working.
To develop these services, product companies need to think like a lean start-up. Focus on the value you're delivering. Try (and fail) quickly by building the smallest service that will help you figure out what to do next.
Our RAPIDE framework provides a good starting point for de-risking and rapidly progressing data science projects.
This requires a different skillset to existing models for digital services. It relies on people who understand the science and know how to handle complex scientific data – vital prerequisites for anyone attempting to model real-world conditions. This data is not as vast or simple as that found in consumer settings. It's a smaller, multifaceted representation of complex systems, such as physical or chemical properties that cannot be reduced to simple formats.
At the same time, this way of working needs people who understand the customer, the business, and legal requirements if it's to be successful. It needs people who can build the systems, software, and infrastructure required to deliver the service effectively. Most of all, these people must understand each other and work together effectively.
New opportunities in digital R&D
Digital services present a new wave of opportunities for companies. It enables them to offer added value to customers through hyper-personalization, using models built on digital R&D data.
They must create interdisciplinary teams that bring together all the expertise needed to launch the service. This includes data curation, scientific modelling, software development, and marketing. Without these skills, you'll struggle to bring any digital R&D project to market.
Tessella’s unique blend of scientific understanding, advanced data science, and software services make us ideal partners in these teams.
About the author
James Downing has 15 years’ experience leading the design and implementation of data-driven digital services at some of the world’s largest product organizations.