Bob Trebilcock: Welcome to The Rebound, where we'll explore the issues facing supply chain managers as our industry gets back up and running in a post-COVID world. This podcast is hosted by Abe Eshkenazi, CEO of the Association for Supply Chain Management, and Bob Trebilcock, editorial director of Supply Chain Management Review. Remember that Abe and Bob welcome your comments. Now to today's episode.
Bob: Welcome to today's episode of The Rebound, the roadmap to future-ready your workforce. I'm Bob Trebilcock.
Abe Eshkenazi: I'm Abe Eshkenazi.
Bob: Joining us today is Ashok Viswanathan. Ashok is the Director of Supply Chain Analytics at Best Buy, and an adjunct professor at the Rutgers University, where he teaches supply chain digital transformation. Ashok, welcome.
Ashok Viswanathan: Thanks for this opportunity, Bob and Abe. I'm excited to talk supply chain and what the future holds, as all of us pursue our digital transformation goals.
Bob: Those are two of our favorite subjects. This is going to be a great episode. I don't think there's a supply chain executive alive who isn't in the midst of or about to launch a digital transformation, however they define that. Digitalizing the business, including supply chain, is just no longer a nice to have. It's an imperative. The part that gets far less attention is the impact of that digitalization on the workforce. I think there was once this thought if that if we automate everything, well, the supply chain can go on autopilot, and we'll need far fewer people.
I think, if the pandemic taught us anything, it's that people are still essential to running a supply chain. Instead, the trick to a sustainable digital transformation is upskilling workforce in areas like digitalization, automation, artificial intelligence and analytics. Indeed, industry leaders like P&G have launched citizen developer programs to put digital tools in the hands of everyday employees from the associate level up to the VP level.
The question today is, well, what if you're not P&G? Where do you start? How do you get there? What's the roadmap? That's what we're going to talk about with Ashok. Ashok, let's get started. First question, easy one. Just tell us a little about yourself, your role at Best Buy, and how you got interested in this topic.
Ashok: Thanks, Bob. Supply chain is at the heart of Best Buy's ability to connect people to the technology we all love, and I have the privilege of leading analytics for supply chain at Best Buy. I also fulfill my passion for teaching, and sharing my knowledge and passion for supply chain by teaching a course on supply chain digital transformation at Rutgers University.
In my 15-year career in supply chain and analytics, working for 3P’s and large shippers, I found democratization of data, and the role of technology in the transformation journey to be unique for every organization, and also a very complex process.
While software engineers and software and data scientists are adept at delivering really good digital solutions, the key to success lies in the ability of the entire workforce to participate in the journey as an accelerator, not a bystander. This realization of the role of people and culture and adoption led me to focus the journey more in making my journey more inclusive of the workforce from all parts of the organization, more as a partner than an internal customer.
Abe: Ashok, let's start with some of the basics here you're describing, obviously, a significant transformation for a lot of supply chains here. Give us a sense of what digital upskilling means to you at Best Buy, what does it involve and more importantly, who are we targeting in terms of upskilling? Are these coders, data scientists, digital natives give us a sense of who was it that we're looking at here?
Ashok: Digital upskilling the workforce doesn't really mandate the need to be a coder or a data scientist. Instead, it involves learning how to think, act and thrive in a digital world. What does that mean? If I may borrow from my teaching content to differentiate between three terms--digitization, digitalization, and automation--since the scope of upscaling depends on these.
Digitization means converting from paper or analog to digital format, like data entry into a computer or a system instead of paper, which largely requires basic computer literacy as part of upscaling.
Automation is the technology to improve the overall efficiency of repetitive tasks. The first point that comes to our mind is a robot. To operate robot, one needs advanced skills to periodically instruct the robot, and at times, troubleshoot.
Now, digitalization is the process of improving the efficiency of processes, large processes through technology, like augmented reality for remote tech support, which requires the skills to continuously operate the technology, which might need extensive upskilling. As you can see, digital upskilling comes in various forms even within the domain of supply chain.
Abe, to your question on who's eligible, I believe the entire workforce can elevate and transform the skillset depending on their role in the digital world. If the organization and the employee are mutually invested in the journey, I think we have the recipe for success. It's also true that today's supply chain workforce does interact with technology via an enterprise platform like a TMS, WMS, ERP, or a handheld or a logging device, or an ubiquitous office productivity software.
Long-term use of this technology has given the users a degree of confidence. The aspect of digital upskilling that is newer is the role of data and analytics. Supply chain workforce is at various degrees of maturity in leveraging data to make decisions. A crucial component of digital upskilling involves building the trust in analytics-driven decisions, especially if it runs counter to an operator's experiential gap. Now, I would like to clarify, it's not the goal of analytics to replace human judgment. Instead, the success of upskilling lies in the ability to achieve a healthy balance between experiential gap and analytics insights to take the optimal decision.
Bob: Ashok, in my introduction, I said we're going to walk through a roadmap to try and get you there. I know that you've developed a four-point roadmap on how to get through this process. Through the rest of this, let's walk through each of those points. I'm going to ask you to start with number one, which is document operational processes and decisions at every step, what's involved here?
Ashok: When we think about an end-to-end supply chain, it is what we can possibly call a very complex system. Again, if I can borrow from my academic side, the complex system is one that has a lot of interconnected and interdependent components, but whose collective behavior cannot be predicted from the behavior of an individual company. Doesn't it resemble a typical supply chain, where the flow of product involves numerous events, handoffs, modes, decisions, systems, people, and exceptions?
These events are handled by a myriad of parties, from DPS to carriers to freight forwarders to internal employees. All these events are also recorded in disparate systems and occasionally in a spreadsheet. The outcome of all this is an absence of an end-to-end view, which works for, let's call it localized decision-making, whether load planner is striving to complete that task without visibility to upstream events, or a comprehensive understanding of the downstream events.
The first step to holistically optimize the supply chain through analytics and technology involves process mapping every step of these operational activities. This means mapping all events, actions and decisions. Special attention has to be paid to ensure that the exception management processes are also documented, not just the happy path. In addition to the physical flow, the bidirectional information flow also has to be documented. I would recommend this to be a cross-functional initiative orchestrated by the presence of what's called a business process effectiveness team in a lot of organizations to be performed where the actual operation takes place, and not virtually.
Finally, there's a heavy catch-22 debate on whether platforms like a supply chain control tower enable a process mapping exercise through visibility or in order to leverage the benefits of a control tower, do we need to perform an operational process map?
As you can see, a comprehensive process map enables visualization and analysis of a connected product journey in which it's very critical in a supply chain, similar to how the customer journey is critical for customer experience because that helps the analytics team to track and interpret the current state of flow, nodes, lead times, and all the decision points.
Abe: Ashok, very interesting and obviously a significant foundation for the rest of the steps. Let's get into step number two, and that's incentivizing data governance at the source. Give us a sense what it means to Best Buy and how do we do that.
Ashok: We are all familiar with the saying "garbage in, garbage out". The credibility of data influences the credibility of insights, the quality of decisions, and the effectiveness of resulting actions. In a lot of scenarios, this is termed as a data problem. In reality, I believe it's lack of process adherence that manifests itself as a data problem. In order to overcome this data problem, there is usually an army of these data engineers who undertake tremendous effort in data cleansing. They can only cleanse what has been captured.
If certain critical elements of the supply chain like the exact time of departure of a truck is not captured, then the ability or the effectiveness of the cleansing process is quite limited. One would say, doesn't it make sense to fix all the data source so that we generate all the right data? Not so easy. Typically, operational processes are designed to offer state complexity and navigate the product through a pre-configured network. An operational analyst, like a load planner or a warehouse worker, looks to a TMS or a WMS to simplify the task, and maximize their productivity, because that's how they are measured.
An unintended consequence of this simplification is the possibility of certain data elements not being captured due to the lack of visibility to the power of data. While this does not impair their ability to perform the task, it severely limits the ability of data and analytics to visualize, mine or model the data. To realize an acceptable level of maturity or data maturity, a continuous review of data quality and completeness should be conducted at regular intervals. Completeness refers to capturing data. It does not mean you capture all available data elements.
It means collect data that is relevant to the business goals. Not just immediate business goals, but long-term business goals also. Quality refers to the credibility of the data being captured. There exists a slew of data governance and data quality tools in the market to automate these tasks and provide right visibility so that all the anomalies can be sensed and corrected early. The key lies in communication of these benefits to the workforce so that they capture all the events that take place in the supply chain, and importantly, incentivizing so that there is all-around buy-in.
As mentioned earlier, good data quality at source saves a lot of effort by the data teams to cleanse the data. It is true that even for world-class data teams, effectiveness is directly proportional to the quality of the data they get to model.
Bob: In your last answer, you used the term measurements and metrics a couple of times. That's a great segue to this step three, which is identify metrics that matter and align with your business goals. I actually have two questions about this. I hope you can keep two thoughts going at once. The first is just explain the differences between the different levels of metrics that you've identified and why this is important. Second, I was wondering as you were speaking, as we're going through a digital transformation, do whatever were our traditional metrics change as a result of digitalization?
Ashok: Great set of questions, Bob. To start out this, let's start with another thing, which is you can't improve what you don't measure. Typically, supply chain performance has been measured on standard metrics, like cost per mile, cost per unit, units shipped, units per hour, throughput, so on and so forth. While these metrics are very important and are quite pervasive, they do not necessarily signify the competitive performance of the supply chain, nor do they signify an alignment with enterprise goals. Also, a load planner or a warehouse employee doesn't realize or have the tools to impact these metrics.
They don't have the knowledge on what they can do to impact these metrics. To answer your original question, Bob, the key lies in the identification of the metrics, the key lies in alignment with the enterprise goals. That's where the hierarchical set of metrics comes into picture. Every organization has a set of, let's call it, executive enterprise schools. The first step in this journey should be identifying what are those enterprise goals and then roll it down to how supply chain can impact those goals for the enterprise goals from level one metrics.
Once it is identified as to how supply chain can impact enterprise goals, then supply chain metrics that directly impact enterprise goals need to be identified. These metrics, again, high-level metrics, but still supply chain-oriented metrics form the level-two metrics. Level three correspond to the operational metrics that drive the high-level supply chain metrics that then eventually drive the enterprise goals. All the diagnostic analytics to perform root cause analysis take place on these level-three metrics at the operator level, who will then identify corrective action.
The good thing about these hierarchical set of metrics is it empowers the operator to understand the problems and positively influence them and give them a satisfaction that their ability to influence a level-three metric directly impacts the enterprise goals. While one would say this concept of measuring metrics at multiple levels sounds effective, how do companies go about achieving this? One part I would suggest based on my past experience, is to leverage or start this exercise by leveraging industry associations and experts to guide workshops and create these hierarchical set of metrics.
Their frameworks engender open discussions that are very impactful. Once this hierarchy is established, the internal teams can own and drive the measurement, the root cause analysis and corrective actions. This whole exercise ensures that every part of the organization is rowing towards the same goal, which is meeting the enterprise goals.
Abe: Ashok, last question, and let me set it up by some of the research that we've done recently about competencies and capabilities within organizations, and the need for governance, more importantly, the trained workforce. One of the studies indicated that the capabilities of individuals coming into the workforce is that they lack critical thinking and some of the real-world experience that we're looking for and yet most of the organizations as you're describing right now are investing heavily in technology, which is probably much more powerful than we're using right now.
When you marry a very powerful technology system with a staff or a team that can't critically evaluate the data inputs and the outputs, you come to a very dangerous outcome for the decision-making. Walk us through why the governance committees and why data analytics teams are critical to this process.
Ashok: When we think about a supply chain and the decisions being taken, they range from operational, tactical, or strategic depending on the impact. Operational decisions usually are guided by transactional systems like a TMS or a WMS. Tactical decisions usually have a longer horizon and strategic are more longer-term. Decisions usually that are taken at the tactical or at the strategic level, the way these take place is a functional leader makes, let's call it an experiential recommendation, accompanied by a high-level cost-benefit summary for the sign-off by a governance committee.
Now, the committee could be their internal teams or cross-functional group depending on the scope of the decision. While this approach works perfectly, it projects a recommendation heavily based on gut, short analytical rigor, and potentially devoid of a review of all alternate options. This is where partnering with an analytics team can alleviate some of these limitations. It is true that an analytics team has access to a plethora of data sources, be it internal or external, has the skills to integrate these sources, and can leverage its extensive modeling skills to impart what I call an unbiased and comprehensive assessment of the solution options.
I do want to point out the role of the business teams in guiding the analytics team through the options, constraints, and rules is paramount to ensuring the quality of the recommendations. This collaborative exercise where each side exercises its strengths and enables a holistic, data-driven recommendation is what I think is key for decision-makers taking the right decisions and leveraging the plethora of information that is around them and driving their organizations forward in this digital era.
Abe: Ashok, very helpful for a lot of the organizations and individuals that are going through the digital transformation today. I think you've laid out a great foundation for individuals not only to understand how to apply the digital transformation, but at least the teams necessary within the organization. I can't thank you enough.
That is all the time that we have today. A special thanks to our guest, Ashok. Finally, a special thanks to you for joining us on this episode of The Rebound. We hope you'll be back for the next episode. For The Rebound, I'm Abe Eshkenazi.
Bob: I'm Bob Trebilcock.
Abe: All the best everyone. Thanks.
Bob: The Rebound is a joint production of the Association for Supply Chain Management and Supply Chain Management Review. For more information, be sure to visit ascm.org and scmr.com. We hope you'll join us again.