Organizational context and consideration

One of the primary considerations to keep in mind is that the context in which the advanced analytics and AI team operates is different than operational teams. Operational and functional teams, like finance and manufacturing, work in their defined domains and rarely deviate from their cyclical processes. Analytics teams work across the entire organization, interfacing with all levels of the company with a focus on discovery and innovation.

The interface between the advanced analytics and AI team and the other groups and teams is one that needs and deserves more attention than it currently receives. The advanced analytics and AI team is focused on disruption, creativity, and innovation. This is an activity that is marked by success and discovery, followed by setbacks and retrenchment. Functional areas are characterized by smooth, predictable process flows, and well-established processes that move forward along an established path at an established pace. When interfacing the work of the advanced analytics group with that of a functional group, the communication and planning of how these two teams will interface and collaborate is critical. Any time there is a creative, iterative, innovative, and unpredictable process feeding into a smooth, well planned, and reasonably static process, there are opportunities for miscommunication, misunderstandings, missed deadlines, and unplanned results.

In Chapter 8, Operationalizing Analytics – How to Move from Projects to Production, we will look more deeply into this topic and describe how best to optimize the collaboration and communication between the analytics and operational teams.

A high functioning analytics team acts more like a management consulting group or an internal consultancy than a functional team. This is where teams can and will experience natural friction and sources of contention. In addition to the mismatch in the cycle times, process types, and the kind of work executed, analytics teams are typically not known for being the most tactful and deft in their handling of sensitive topics and subjects. And when the analytics team is discovering new ways of executing processes or interpreting models and results, there are real and substantial areas for misunderstandings to arise. These are challenges that are not widely known or understood, but they will occur, and they can and should be anticipated and considered in the team-building phase of the process.

Another consideration is that while the advanced analytics and AI team utilizes data and technology to accomplish its goals and objectives, the analytics team is not a technology team. The functional teams may perceive the advanced analytics and AI team as an extension of the information technology team and that is a limiting view and a source of future problems. Typically, those issues and problems are in the perceptions of the functional teams and their leadership. The problems manifest themselves in inappropriate requests and the delegation of tasks that are not a good use of the analytical team's time and resources. Many of these requests are to build simple reports and dashboards or to acquire or pull data from internal and external systems. The leadership of the analytics team needs to be clear that the analytics team is not an adjunct or extension of the business intelligence team or reporting group and these types of requests are better served by other inpiduals and teams.

The leadership of the analytics team needs to direct the functional team members and managers to the appropriate contacts in the company to service these requests.

Advanced analytics and AI teams are drivers of deep examination and change. Analytics is a transformational process. Analytics is typically the first step in a transformational process. Problems typically arise when the advanced analytics team finds new and novel ways of doing business from a process perspective or finds data proving that the existing operational processes and activities are not in alignment with the most productive or efficient way to operate.

One of the results of advanced analytics teams engaging with functional teams is requirements for change in the functional organization. The projects executed by analytics teams in conjunction with functional teams are containers for longer-term digital transformation and change. The immediate projects kick off a requirement for process change or, at the very least, an examination of the need for change.

The functional teams and their management leaders may see and be enthusiastic about the immediate projects, but they may not be ready for the larger transformation that is discovered, and possibly required, at the conclusion of the project at hand. The challenge in this situation is that the advanced analytics team needs to be ready to communicate and manage not only the immediate project but the communication of the possible paths of change for the functional team. Technical skills, building PowerPoint slides, and discussing the validity of the analytical methods will not enable the advanced analytics team to manage these larger and more complicated organizational change processes.

The advanced analytics and AI team members you are seeking are more than experts in advanced analytics; you are seeking staff members who are agents of change and can act as guides to the broader organization about how to undertake changes at a scale and scope that the organization can understand and undertake without experiencing traumatic disruption.

These are broader considerations to keep in mind as we begin to consider the type of advanced analytics and AI teams we can afford and want to hire. There are many team structures and constructs that we can use to begin our journey and understanding some of the most pertinent organizational considerations will help us get off on the right foot. Now that we have set some of the organizational context, let's move back to our discussion of building the team.

We all know, and should follow, many of the well-known rules, structures, and norms of building a new team. It is clear that we should begin with ethical, sensible, and legal team-building foundations such as pay equity, clear reporting lines, defined roles, a flat hierarchy, and more, but there are also many elements of building a high-performing, highly collaborative analytics team that, in some cases, appear paradoxical. In this chapter, we will delve into the more unique aspects of building an advanced analytics and AI team.

We will examine what we have learned from our career and what we know we will continue to leverage in our team-building efforts. We'll also look at what we should discontinue doing because it does not serve our greater goals in constructing what we hope will be a dream team. There will also be new things introduced in this chapter that you can try in your ongoing search for improvement in building your team.

New innovative minds

Experience and expertise are wonderful attributes and, in many cases, are a requirement for inpidual and team success. Your team needs staff members who have both experience and expertise, but you also need bright, young sparks who do not know what has been done before, and therefore, do not place limitations on their thinking and approaches to challenges that most experienced staff members would consider impossible to solve.

Hiring smart young staff members can be a good check on confirmation bias, which we commonly see in staff that have been in an organization for a considerable number of years.

Early in my career, I was asked to join a team at a client site. It was a good client with interesting challenges in a growing industry. The firm was one of the leading Consumer Packaged Goods (CPG) companies in the world. The team leader from my company was quite poor; she only wanted people on her team who parroted back her thoughts and slavishly followed her directions. For the most part, the team did just that and toed the line she set out each day. She exhibited a strong project manager mindset. She wanted the trains to run on time, no projects were to deviate from their project charters, all projects should be delivered on time, and all projects were to be reviewed and approved by her.

After being at the client site for about a month, I was approached by a leader in the client organization and asked to listen to an idea for a new way to solve a long-standing, vexing challenge. I listened carefully and asked a few questions to ensure that I understood exactly what the client wanted and needed. After the office grew quiet, I spent the night building what had been described and discussed. The next day, I demonstrated the newly built application for the client team. The consensus was that this was a vast improvement over what had been the state of the art up to that point. The application was fast, accurate, and utilized source data and intermediate results from 17 internal and external databases. Previously, no one had considered approaching the challenge in this manner because the brand management and information technology teams involved to date thought it was impossible to do.

This was my first "impossible" project. Some of the primary reasons why the project was a success were:

1. I had no idea of the current state of the art for this type of application in the CPG industry.
2. I had access to all the information and databases needed to build the application.
3. I was given a complete overview of the business problem by a senior functional leader who owned the problem and wanted a solution as soon as possible.
4. I acted without asking for approval to spend my time on the "project."
5. I worked through the night building, what seemed to me, the most efficient application to solve the challenge.

The team leader from our firm was not happy and asked that I be removed from the team. I was surprised, but in hindsight, her reaction could have been predicted. Her metrics for success centered on project delivery dates and adherence to agreed project charters, and control of the team. The innovation that was delivered was spontaneous and unplanned and, while the client was very happy, and continued to employ the application for years, she was very upset that the project did not flow through her.

I went on to be assigned several "impossible" projects across the CPG industry in the US and UK. My managers knew that I enjoyed difficult technical challenges and the clients were impressed with the novel approaches and solutions that were developed and delivered competitive advantage to them.

Young, talented staff members view challenges from new and varied perspectives. They use new thinking, new technologies, and see prospective solutions in new and unique ways. Unleash their creative and innovative spirits and you will see solutions that you had not dreamed of before.

This is another area where leadership from the analytics team is important. You will be asking young, talented, and creative analytics team members to work with functional team members and leaders.

The young staff members need support to ensure that they understand the challenge completely, communicate thoroughly, and do not over commit to delivering more than they can in too short a time frame. Analytics leadership needs to set the expectations with the functional team and leadership about timing, the type of results they can expect, and if this is a project that will produce a singular outcome to derive insights or a set of models that will require implementation by inserting those models into existing processes. One-time insights are easy to understand and work with. Production-based models require processes, workflows, and procedural changes, which may be harder to understand and take more time to plan for and implement.