Blog: “Demystifying” Digital Transformation
A pragmatic definition to apply Digital Technologies to Businesses
Author: Ricardo Ivison.– Senior Enterprise Architect at Pharma R&D in Bayer Business Services & Digital Transformation Evangelist.
Digital Transformation Definition
Digital Transformation is the topic nowadays. It is sometimes perceived by senior managers as some kind of “magical and mysterious” power that will bring enterprises into a whole new era. Like a “holy grail” that will solve business challenges or create new disruptive business models that will pour in revenue streams while creating efficiencies.
Once a specific industry term like Digital Transformation becomes “hyped”, there will be as a result many service providers, like software vendors, consulting firms, industry publications, etc. trying to hook into the hype by talking about the topic from different angles. I have read books and publications that show how to create new digital business models, or how to automate business processes, or how to get insights out of enterprise data by digitalizing data assets, etc. Although in some cases such diverse views will provide valuable input for organizations to build their digital transformation strategies, in others it’s honestly just a full of buzzwords used to sell software products or services which claim to give organizations that edge into the new digital business era.
But let’s go back to the basics: What does exactly Digital Transformation mean? In order to be pragmatic and clear, I have come up with the following simple definition:
“Digital Transformation is the journey within an organization to transform data from business processes into a machine-readable format so it can be used in value-driven business applications or new business models”
Although the above definition is simple, it doesn’t mean it’s easy to execute or exempt of profound consequences. Therefore, I will break down the key elements of this definition to unveil the implications.
Journey to Transform an Organization
Transformation is a deep word and is not cosmetic. In simple terms, it means going from a state A to a state B. For example, if you want to lose weight because you are overweight, then you want to move from a state when you are above or far above your ideal weight, to a state when you are at or slightly under your ideal weight.
STATE A: Overweight -> STATE B: Ideal Weight
In order to transition to overweight to ideal weight state, it requires a change of eating habits, exercising, etc. which as a result will challenge the existing “status quo”. It can’t be expected any transformation without changing established habits and behaviors.
The same holds true for organizations. When companies want to embark in any kind of transformation initiative, whether is culture transformation (e.g. moving from a hierarchical organization towards a leaner, open one), business model transformation (e.g. how to move from being an innovation company to a premium service one), etc. there has to be a purpose behind it. Of course, digital transformation is no different. In all cases, the main question is: Why as an organization, do we want to go through this transformation? Is it to provide a better customer experience? Is it to be able to remain competitive in markets? Is it to create new business models? Increase shareholder value? What is it?
The purpose of any transformation initiative should not be vague, the clearer the purpose, better the results will have. The purpose should not be fuzzy or unclear, because the implications to the status quo of organizations are not minor.
Any transformation initiative is a journey. Going from one state to another will require a period of transition. Ideally, it should be an enterprise continuum. To use the same example I used above: the more overweight you are, the more time you will need to get to the desired state: to be lean. Along the way, resistance to change will show up and only if the purpose, motivation, and implementation of new habits are strong enough, resistance can be eventually overcome.
Trying to embark in a digital transformation initiative without changing or challenging existing status quo, it’s like trying to lose weight without eating better and not committing to do any kind of exercise.
Because Digital Transformation is a journey, organizations need to identify and prioritize business processes to target, e.g. processing purchase orders, case management, customer up/cross-selling, managing customer relationships, etc. In R&D, examples can be such as getting insights on data generated through experimentation, or how to automate patient recruitment in clinical trials, etc.
Because resources are limited, organizations have to be clear in terms of the business processes they want to focus on.
A good practice is to map the different digitalization value cases to the overall digital transformation purpose and goals. In other words, Why to transform a certain business process as opposed to others in terms of their contribution to the overall purpose? Another good practice is to prioritize those with higher impact on the overall purpose and lower implementation impact.
The following figure shows how to bucket digital transformation value cases in terms of the risk vs reward:
The criteria of what the value is to the business vs. risks are closely related to the purpose of the organization. For example: let’s assume that the purpose of an insurance company is to provide the best Customer Experience. There may be a case where there is a high risk / high reward digital value case, where the technology is not mature or unproven, but on the other hand, it is highly aligned to the overall purpose defined. For instance: the use of IoT devices to help customers be on better health condition to lower premiums in Health and Life insurances. The challenge is not only capturing and measuring customer biometrics in a safe, reliable and affordable way but also profiling the customer to tailor their journey and motivational approach. All these may be highly risky, but if an organization is really committed to the purpose to use Digital Transformation towards the best customer experience, it may worth the risk of the unproven.
Another example: if the purpose of a Digital Transformation Initiative is to create as many efficiencies as possible, think about another insurance company that is focusing on the health insurance claims process. The value case is to use the amount of historical claims data, available in internal/public databases to automate the process outcome. The company has also data scientists available in-house. It may be relatively straight forward to use machine learning algorithms to train and automate the ruling of the health insurance claim (assuming is regulatory allowed) and saving significant costs in the process. In this case, the value case would be bucketed in the “Must Do” quadrant, which refers to “low hanging fruit”.
These are two examples to show why a clearer purpose helps to prioritize where investments should go in competing for value cases within Digital Transformation programs.
Secondly, data needs to be in a machine-readable format. That is the key characteristic of digital. This is a necessary step to turn data into information, and information into knowledge.
Regardless of the use case, data derived from business processes need to be in a machine-readable format. For example, if the data derived from the sales closing process is still in paper, that data can’t be used by machines. There may need to use OCR (Optical Character Recognition) technologies and NLP (Natural Language Processing) to extract text out of scanned documents. Another example: If the target process is focused on customer service e.g. case management, but the case is not recorded anywhere, then it will be necessary to either fill in the outcome in a CRM system or to record the audio, then convert the outcome of the case into structured text machines can read it. Let’s take one last more complex example: imagine a use case where scientists want to understand adverse events associated with a particular compound. The compound is part of an active substance of a drug. Today, data is spread across different data sets like chemical compounds databases, clinical trials, and real-world evidence. They are owned by different organizations and use different vocabularies. The correlation of this data can provide great insights about indication expansion of the drug or how to further improve safety in the development of the next generation of the drug. To get access to all necessary data, it will require data to be curated and linked semantically so machines can find them and thus data scientists use it.
There is no digital transformation without making data of business processes machine readable.
Value-driven business applications
Lastly, turning data in targeted business processes into a machine-readable format must be done to add clear value to the business.
Making sense of data is imperative to make it functional from a business value point of view. Creating a massive stream of data out of a process with no business use is just a waste of resources. On the other hand, making data in a machine-readable format may provide incredible business opportunities.
The ultimate goal of digitalizing data out of processes is to create business value from it. However, the effort to make it available in a machine-readable format in many cases is not trivial. Unless the value for digitalizing data is clear, it could easily end up in expensive failed implementations
Business value can be put into three categories:
· Revenue-oriented: how the value case can bring more income into the organization? Is it a new business model, it is supporting existing business models and their processes? Etc.
· Effectiveness: how does the use of machine-readable data can achieve efficiencies/cost reductions? Is machine-readable data available on a hard scale? What can we automate?
· Risk Management: Can risks be controlled/prevented by digitalizing data triggered by business events or processes?
If people, money, technology and time will be used to digitalize data assets, the application of what to do with the outcome in terms of business value should be clearly understood. As mentioned already, if the business value is not clear, it may lead to ineffective and expensive technology investments.
Digital Transformation initiatives should be treated as a journey. Identifying business processes, applying technology to make data out of the machine-readable and getting clarity in business value. Doing all these along this journey may have profound implications. Following are a few examples of this:
Organizational & Cultural Implications
Organizations may want to focus on a large and complex business process first, which in reality goes through many departments, run in functional silos and fragmented systems. Without senior leadership looking across the organization to help break silos so data can be found, accessed & interoperated across functions, digital transformation value cases may end up in the best case scenario, in fragmented implementations with underperforming results. At the end of the day, all comes down to people. And people, especially in large organizations, have different priorities and objectives. This is why many organizations are setting up cross-functional Digital Transformation offices, with strong senior sponsorship and shared objectives.
However, are organizations truly willing to embrace the cultural challenges which come across by making data transparent across the organization? What about security and governance aspects related to sensitive data assets, i.e. remaining compliant? What about the complexity to deal with legacy applications or poor quality data? What about resistance to change in this new mindset of creating digital data or working in collaborative environments? Just to mention a few challenges that will show up along the way.
Unfortunately, sometimes value cases for digital transformation may have very clear business drivers behind them, but they may also have collateral implications on people. The fact that any data out of business processes in scope becomes machine readable, it means that technology such as Machine Learning, Robotic Process Automation, etc. can be put in place to automate tasks that in the past where performed by humans. The implications involved people that were laid off is never easy to manage. Change management is critical and should be always part of any digital transformation initiative.
It is not hard for organizations to be let carried away by the hype, investing in state of the art technologies, or bringing an army of data scientists thinking that will do the job by itself. However, before even starting, the most important question to ask is: what is the business value out of this initiative? How does the investment to turn data out of business process bring more revenue or new business models, create efficiencies or manage risk better? What are the key KPI’s that will define success? What is the applicability to make a digital data asset in practice? If these questions can’t be answered in a simple way, then perhaps there is still work to get done to get more clarity in terms of business value.
Digital Transformation Definition Summary
The Digital Transformation definition explained here is a simple sentence that can be used at different levels in most organizations. However, simple does not prevent complex and profound implications. Nonetheless, avoiding unnecessary buzzwords is always a good start. Furthermore, it can also serve as an input for guidance in terms of what to do and how to embark in Digital Transformation initiatives.
And that is the next step! How to use this definition in practice, in a pragmatic but tangible way? That will be explained in the next blog: The Digital Transformation Canvas