The second key theme is one of parsimony. Parsimony is a philosophy of intentionally expending the minimum amount of energy required in an activity so as to maintain overall efficiency. This is a key part of modeling; it is about keeping things as simple as possible, but no simpler. Similarly, and in the vein of recursion, teams themselves must be parsimonious. For analytics to mature as a practice while still delivering an accelerated time-to-value, teams need to be scrappy and lean.
At a foundational level, the objective of this book is to provide clear insights into how to structure and lead a successful analytics team. This is a deceptively challenging objective since there are no generalized templates from which to work. Establishing a project management office, information services, or human resources department is an understood process and does not vary materially between organizations. Establishing an analytics team, by contrast, requires a significant up-front investment in understanding and contextualizing the initiative. Many organizations have attempted to use operating models and templates from other functions—often IT and operations research. This fundamental misunderstanding of where analytics fits within an organization has led to visible failures and has set back the analytical maturity of many organizations. Business leaders need to hire or develop value-centric talent who can step back from analysis and project management to view their work as existing within a network of individuals and teams with competing priorities and motivations.
Corporations, without a template or a default methodology to benchmark against, have made expensive missteps in building these teams by installing the wrong leaders, copying other functions, positioning the function under the wrong executives, incentivizing destructive behaviors, hiring the wrong people, and committing to the wrong projects. They have contracted expensive consultants to provide roadmaps that do not consider the unique culture or competencies of their organization. They have embedded disparate data science specialists throughout the organization and incurred enormous technical debt.
These issues are not insurmountable, however. Whether an organization is beginning its first foray into the analytics space or it is rebooting a failed team for the third time, the key is the creation of a carefully considered strategy, the establishment of realistic goals, and the full commitment of executive leadership. The advantages associated with analytics are too great to overlook, and the long-term cumulative impact of interrelated and interdependent models provides a powerful incentive for aggressive adoption.
This book was written for anybody who aspires to lead or be part of an effective analytics team, regardless of managerial experience. Every analytics leader, from a first-time team lead to a seasoned VP, has unique challenges to overcome.
For the Leader from the Business
Every new role is a challenge, regardless of ability, disposition, or motivation. This is particularly the case with a unique subculture of academic technocrats with whom it is difficult to establish credibility without enough time being “hands on keyboard.” Without the respect of your team, it is impossible to get the buy-in required to establish best practices and ensure that the output of the team is not simply self-satisfying experimentation but can bring real value to the organization. As a corollary, every practitioner has experienced a manager who is out of their depth and who has compensated for their lack of self-confidence with authoritarianism and distrust, shifting the focus of the team toward an end they are more confident with, such as reporting.
For all but the most analytically committed organizations, there is a point along the chain of command where a practitioner reports to a non-practitioner. This can be a challenging junction for both parties without clear expectations, transparent communication, and mutual respect. Catching up from a technical perspective isn't feasible or advisable, but by leveraging your business understanding and domain knowledge to become an intermediary, translating business needs into projects and analytical outputs into operationalizable processes, you can unlock the power of your new team and give them the opportunity to develop into more business-oriented individuals.
For the business leader, I hope that this book helps you to reframe and refine your current leadership abilities, and to use them in an analytics context in order to engage your team and find success together.
For the Career Transitioner
Those who transition to data and analytics mid-career have a key differentiator from those who have entered the field directly from university—breadth and context. The ability to leverage your multidisciplinary background from engineering, finance, sciences, and so on is valuable both to your career and the organization you join.
Though it would be almost impossible to compete with trained data scientists on a technical basis, it is the disciplinary and sector diversity of the team that drives innovation, and those who have worked in multiple industries and functional areas bring a unique perspective to the teams with whom they work. Rather than starting a new career as a new hire and individual contributor, with personal study and intentional self-reflection mid-career transitioners are often able to seamlessly make a lateral move. Having familiarity with the different AutoML and analytics-as-a-service offerings, combined with transferable managerial skills, can make for a powerful combination.
For the career transitioner, I hope that this book helps you to prepare for lateral movement into an analytics role and to use your transferable skills to add value in your new function.
For the Motivated Practitioner
It is an unfortunate truth (and perhaps an unfair generalization) that the skillset that makes a practitioner a competent data scientist is rarely the skillset that makes them a competent manager. Though there are certainly analytically minded people who have the natural inclinations toward leadership and bigger picture thinking, it is rare that in practice those people would have the technical depth to stand out as a candidate for management. Often, those with the natural capabilities required to enter management can appear to be less effective as individual contributors on a purely technical basis.
To make the leap to management is to leave an objective and predictable role with performance metrics such as p-values and ROC curves and exchange them for stakeholder management, workshop facilitation, and inherent subjectivity. Those able to successfully make this transition while maintaining the ability to downshift to provide analytical support establish themselves as leaders in the practice and are in high demand. Exceptional managers who have legitimate technical credentials are the unicorns of data and analytics.
For the practitioner, I hope this book helps you to understand what is required to move up the value chain and to prepare for leadership opportunities.
For the Student
When a student pursues an applied field such as business or engineering, the curriculum is generally developed in a way that seeks to balance between foundational academic elements and applied profession-specific education. The curriculum is updated and maintained such that it remains aligned with the changing needs of the field. For several professions such as accounting, law, and engineering, this takes place within a partnership between the administrative body of the professional practice and the educational institute, and through accreditation it's ensured that graduates of these programs are broadly educated and prepared to work in the field they have studied. This is unfortunately not the case with data and analytics.
Most North American universities have data science or analytics offerings, but having no natural home they are generally provided through multiple faculties such as business, mathematics, engineering, finance, or computer science. These programs provide instruction in highly simulated and well-defined problem solving, focusing on the improvement of a statistical metric. The data is often perfectly presented and accompanied by a well-articulated data