World Class Data & Analytics

What Are the Necessary and Sufficient
Conditions for Digital Transformation?

Back in 2017, The Economist published a story titled, “The world’s most valuable resource is no longer oil, but data.” If that is the case, the digital transformation is no longer just important, but urgent.

The most essential condition for organizational transformation is the commitment of the top leadership of the organization. This cannot be perfunctory or performative and must include the CEO and, ideally, the Board. The failure to appreciate the importance of driving the culture change from the top is the Achilles’ heel of cultural transformations, no matter how well designed and executed. When top leadership withdraws their commitment (say through a change of leadership personnel), transformation can be undone. It is essential, therefore, that the commitment to the new culture pervade all of the organization including being a prime factor in the selection and on-boarding of new leaders.

Of equal importance is the engagement of middle management. If top leadership is the engine that provides the energy for transforming the organization, middle management are the drive train that transmits that energy from the engine to “where the rubber meets the road,” i.e. the rank and file of the organization. Middle managers must be seen authentically living the new culture and must effect the transformation at the interpersonal level, hence Gomis’ emphasis on the need for “hybrid” data analytics professional.

Executing the new, designed culture at the organizational level will in almost every case require that all aspects of the culture as it affects relationships change, including business and social interactions, acknowledgement of achievements, appreciation both tangible and intangible, communication, collaboration, and even compensation.

The final condition for digital transformation is a shift in the relationship to failure on the part of everyone in the organization. No one likes to fail, and the traditional culture of organizations celebrates success and punishes failure. Unfortunately, success is a terrible teacher. More accurately, human beings, while celebrating success, rarely look deeply into what there is to learn from it. Steinbeck called success the danger which in the past has been most destructive to the human.

Failure, on the other hand, is a spectacular teacher. If we remove blame, shame, guilt, and recrimination from the conversation, failure guides us directly to what didn’t work or was missing from the attempt to achieve the goal. In addition to Gomis’ list of business acumen, technical expertise, and communication skills, we would add a growth mindset. Research has shown that people who believe their talents can be developed through learning from a variety of sources including from failure achieve more than those with a more fixed mindset (the belief that talents are innate and unchangeable gifts).

Business Acumen: in the Digital Transformation, data analytics professionals need to know much more than technology. They need to understand the whole business of the organization and to cultivate an enterprise mindset – not just “what am I developing,” but “how does what I’m developing create the future of the organization and create a foundation for ongoing product development, customer service and satisfaction, and enabling advances in technology that will enhance the organization in a future I can barely imagine?”

Technical Expertise: in this case, technical expertise is not restricted to know-how, but also to asking what will be needed in the long-term and what can be developed to further the organization’s aims in the present.

Communication Skills: It is not enough for Data Analysis professionals to collect data or even to organize it. Socialization of analytics is a critical skill in the new world of the digital transformation. Socialization refers to the sharing of data and data analytics tools with all members of an organization. The key idea behind data socialization is to make data-driven insights available to everyone in a self-service fashion.

Another way of defining data socialization is to say that it involves the “democratization” of data. Whereas the typical business has traditionally assigned data analytics tasks to only a handful of employees who specialize in data management, the data socialization concept aims to involve everyone in the organization in collecting, managing, analyzing, and reacting to data.

Data socialization is innovative because it helps businesses to double down on their ability to leverage data. Most businesses collect huge amounts of information, ranging from machine data (like Web server logs) to manually entered customer reports and everything in between. Yet traditionally, the extent to which businesses have leveraged that data has been limited.

As noted above, the ability to access and analyze business-critical data has typically been available only to a small team of data specialists. Other parts of the organization management rely on data analytics to collect and analyze the business’s data and then provide recommendations based on it.

From a business standpoint, this approach is not ideal, for two main reasons:

It is difficult for a small group of specialists to process an entire business’s data single-handedly and deliver relevant insights and recommendations to every business unit. This is especially true today, when the amount of data that organizations collect is larger than ever.

Unless data specialists have a broad understanding of the business, their ability to leverage data in ways that benefit other business units limited.

Data socialization aims to solve these challenges by placing data and data analytics tools directly in the hands of the people who can use them as part of their jobs. This requires that data specialists have that broad understanding of the business and its strategy, as well as the communication abilities to engage non-technology personnel in learning how to read analytics and draw conclusions and learnings that allow them to make successful, forward-looking decisions for their part of the organization.

Using marketing as an example, data socialization means that data can be collected and analyzed for marketing campaigns, rather than depending on data specialists to perform that task. Because marketing executives will know the business’ marketing needs better than anyone who does not specialize in marketing, they are better positioned than the rest of the organization to derive relevant insights from that data.

The success of data socialization stands or falls on the ability of data specialists to create relationships that allow them to educate others on the importance of a data-centric orientation. They are the experts, and they oversee the tools and processes that enable other parts of the organization to perform data self-service.

This means that data analytics professionals will need to approach data management somewhat differently than they did when only data specialists were involved in the process. Most obviously, they will need to acquire or develop data management tools that enable self-service without requiring a great deal of expertise. This might seem difficult to achieve, but data integration and analytics are simpler today than they once were, and even non-technical personnel are used to using sophisticated programs on their computers and tablets, and so will likely be able to work with data much more effectively using modern data management tools than you might expect.

That said, the ability to design and deliver a data analytics process that is free of complex technical kinks. For example, non-data-specialist employees should not be expected to be able to perform complex data transformation or data integration tasks. Nor should they be expected to clean up low quality data sets. Instead, they need to be provided with data that is readily usable and tools that enable them to visualize data easily

Growth Mindset: The place for the introverted, isolated “data nerd” in today’s data-centric business organization is rapidly fading. The data analysis professional of today, whatever their personality traits, must be willing and able to form productive business relationships, converse fluently in the language of other parts of the business, have a strong grasp on the organization’s vision, mission, and strategy, and be able to present analytics in ways that others can comprehend and use.

In sum, companies are looking for hybrid candidates that can do it all – people who are able to understand the technical side of the job, inclusive of coding languages, architecture, infrastructure and more.  At the same time, companies want people who can speak openly with stakeholders in a way that is digestible. In the Fortune article, Guy Gomis argues that in order to develop people like this, both early education and higher education institutions need to invest more in getting students interested in STEM-related fields earlier to start developing skills earlier on.