Companies can now make data useful to elevate decision making and to optimise products and processes. But what organizational capabilities are necessary and how to get started?
It's currently easy to acquire data strategically. First, consider that smartphones function like questionnaires that customers are frequently filling out in a passive or active manner [1]. The average smartphone owner uses around 10 apps per day and 30 apps each month [2]. Since 2017, more than 1.5 billion smartphones are being sold to end-users worldwide every year [3]. Then, add to this the ubiquitous usage of customer IDs to uniquely identify users' information, which is acquired across multiple channels [4].
Ultimately, companies are able to systematically analyse examples of customer journeys and internal processes. Such analysis is then used to mitigate decision risks and elevate the customer experience [5].
This is the first piece of a series of three articles in which you'll learn:
the natural linkage between digital transformation and data capabilities (the WHY);
the data capabilities that allow the economic value of data to be extracted (the WHAT);
an iterative process to develop data maturity (the HOW).
If you ever have to explain to friends or colleagues why data capabilities are crucial to navigating the future of work and innovation, try this storytelling tactic. Briefly narrate the modern history of digital technology in these few easy steps.
✅Tell the difference between Digitization and Digitalization
It all began when organizations started taking analogue information and making it digital. The fax machine is my favourite device example of this epoch. But of course, digitization still happens intensively nowadays. For instance, it occurs when a restaurant creates a digital version of a printed menu, so customers can scan a QR code and access it via a browser [6].
Going further, when a restaurant creates a digital channel for its customers to order food online, it is not only digitizing information. It is making use of digital technology to extend the business model to a different audience and way of working.
This is digitalization in the making [7].
Historically, I like to link the digitalization wave with the period when households and workplaces started to have personal computers. Initially, users didn't know much about what to do with those machines. But humans quickly started to figure out how to print documents, play games, exchange emails, listen to music, read the news, buy products and much more. It was mainly a "product first, customers second" mentality of building products and services.
✅Explain the link between Digital Transformation and Product Development
With this wave of digitalization came the need to build proper information technology teams and promoting digital literacy, mainly in the workplace. It was the "Cambrian explosion" of the usage of relational databases, spreadsheets, and slide decks. This phase also mediated the development of business intelligence and the implementation of descriptive analytics [8] to monitor business metrics.
At some point, innovative businesses commenced reversing the process of product development. My favourite examples from this phase are both from the 2000s: the iPod shuffle by Apple [9] and the personalised movie recommendations by Netflix [10]. The iPod allowed humans to carry a huge setlist of songs in their pockets to listen to while commuting or working out.
Netflix solved the long tail problem of offering niche contents in a channel that had no shelf space boundaries but limited attention span from users in the website. It did that by implementing a recommender system based on machine learning. In essence, a body of companies started to build digitally native solutions to solve old and new customer needs and goals.
... a body of companies started to build digitally native solutions to solve old and new customer needs and goals.
Reengineering the process, by starting with the customer needs and wants in mind and working backwards [11], was then possible because it became easier for companies to observe customer feedback not only explicitly but also implicitly. Here, I refer back to where this article started, i.e. the smartphones and customer IDs. They constitute the major vehicles in which customer digital footprints [12] are collected in the form of structured and unstructured data [13].
Add to this the contribution of two other major catalysts of change in the late 2000s [14]. These are the sharp increase in available computing power at affordable prices and the advent of new algorithms that could be used to make predictions and uncover patterns in data [15,16].
✅Demystify what Advanced Analytics really means
Going beyond simple analytics means that companies advance their techniques to incorporate methods that deal with risky choices in a more efficient way than using just business intelligence and human judgement [17]. The first of these techniques is good-old statistical modelling but then powered by much bigger sample sizes and the intense usage of experimentation via A/B testing [18]. The second is machine learning, which then provides decision-makers with much cheaper support in a human-in-the-loop approach or the possibility of automation of repetitive predictable tasks [19].
The ability to develop these proof of concepts is what typically defines the job of a data scientist [20]. The end goal of the discipline of data science is to make data useful in business decisions [21].
But, bottom line, what means to make data useful in business? I like to use the first and second Schmarzo's Theorems of the Economic Value of Data [22]. They briefly represent the intent of using data and analytics in business decisions. They state that:
#1. Data, by itself, provides little value. It's the trends, patterns, and relationships gleaned from the data about your customers, products, and operations that are valuable to monetize.
#2. Predictions, not data, drive value. It is the value of these predictions in support of the top-priority business use cases that ultimately determines the economic value of your data.
While the decisions have not changed over the years, what has changed - courtesy of advanced analytics - are the answers. - Bill Schmarzo.
Obviously, this value is extracted by the existing data capabilities within an organization. These capabilities are formed by the investment in teams, use cases, governance and operationalization. Therefore, this discussion of the 'What' and 'How' of building data capabilities will be made in the coming articles.
This article does not try to define what comes next after a digital transformation. Many practitioners have already extensively discussed this topic [e.g. 22,23,24,25] . It should help the reader to make sense of the reasons to develop data capabilities.
Takeaways
When you acquire data strategically, you can develop data capabilities to make sense of it. Under normal conditions, this investigation contributes to product development and process improvement. But remember:
Data acquisition per se does not generate value. It needs to be combined with the proper capabilities that transform it into actionable support to decision making.
This support happens in the form of higher-quality business questions, predictions, and prescriptions.
This is the first piece of a series of three articles in which you learn about:
the natural linkage between digital transformation and data capabilities (the WHY);
the data capabilities that allow the economic value of data to be extracted (the WHAT);
an iterative process to develop data capabilities and increase data maturity (the HOW).
Thanks for reading. If you find this type of article useful, then follow me on LinkedIn and Twitter. You'll make sense of the applications of data science, behavioural science, and digital solutions to predict consumer behaviour. Until next time!