The Case for Data Analytics
Data analytics is a critical skill for anyone that is entering the professional world. Understanding how to trim down large datasets to yield pointed insights, how to scrape together bits of data to string together a hidden narrative, and how flexible data allows analysts to drill down for actionable insights as well as powerful summaries for leadership understanding are all key when finding
Working across multiple supply chain functions early in my career, I’ve found three critical uses for data that have allowed me to:
Make informed decisions based on historical data and/or predictive forecasts;
Provide communication for key partners and summaries for leadership; and
Introduce automation in my role to increase the scope of my work.
In the last few years working in retail supply chain, I have worked as an Analyst/Senior Analyst and am currently an Operations Manager working in our Fulfillment Center (FC)/Import Warehouse (IW). In each one of these roles, I have utilized different types of data to create tools, make decisions, and inform conversations with my leadership teams.
While my experiences may be in retail supply chain, these learnings are applicable to any industry. In today’s world, every click, every purchase, and every scan yields an instance of data that can be spliced and diced to reveal larger trends in your industry.
Pt. 1: Making informed Decisions
As the Inventory Analyst (IA) for the water category, I was purchasing and allocating inventory for each item in my assortment. This meant end to end supply chain troubleshooting from vendor order points, transportation, and distribution centers (DCs), to the stores and guests making the transactions. The bulk of my role was troubleshooting out-of-stocks (OOS), making sales forecast variance updates, introducing new items into the supply chain, and - in my case - responding to hurricanes and other natural disasters to provide crisis response during the critical hurricane stock up period. The responsibility of managing a billion dollar category as an analyst can be daunting , but the data helped me understand my business, manage dynamic growth, and navigate pandemic-level sales patterns and variability.
I could use week-to-date OOS data to make decisions on where to prioritize my efforts to get items back instock. Which DCs have highest OOS, which are the highest sales volume to prioritize over others? Is there vendor context that is necessary to understanding the root cause of this OOS?
Historical sales data for the previous 4-8 weeks helped me make sales forecast variance updates - where were our sales actualizing compared to the sales forecast we had in the system? Do we make updates at a store level? A DC level? Are certain parts of the country seeing faster growth/decline than others that we would want to bring in more inventory before making a true sales forecast update? Are high OOS impacting a strong negative forecast variance, meaning I should investigate a true root cause? Is it a new item that typically see more volatility than existing items?
When it came to crisis recovery, how could I identify crisis stores in the storm’s path and route deliveries of water to stores where we see our guest stockpiling on product? Sending inventory based on predicted relief need is a part of how we make smart moves to help people who are preparing for a natural disaster without inundating the store with countless pallets that they may not have long term space for in the store.
Data helped me understand how to take simple, strategic action by correctly understanding all of the information available to the problems in front of me.
Pt. 2: Provide Communication and Summaries Similar to the sales pattern we see during storms and disasters, COVID brought a sustained sales lift nation-wide for over a year. Which items should we prioritize as our suppliers started imposing inventory allocations and maximums? Where should we allocate the inventory in our internal network? As we’re increasing our safety stock levels, what DCs have the capacity to hold increased inventory levels? How do you make a case for your category to take priority over other categories?
To start, understand historical sales volumes and account for recent sales trends. From there I could understand how many weeks of inventory supply we wanted to keep in the supply chain? Translating between different measures becomes key here. I communicated in terms of dollars to my finance partners, in cubic units to my DC partners, in terms of trailers for transportation, and weeks of supply (WOS) for my inventory leadership.
Each of the partners was given specific communication on timing and volumes by week, and key leadership partners were given recap information for actions taken and impacts it had across the organization. On top of the strategic action, all of the partners were able to understand total ownership cost and predict any capacity concerns before we encountered them along each step of the way in the supply chain. Field sites were able to communicate any concerns with strategy and execution, while leadership was willing to remove obstacles that stood in our way to effectively support our efforts.
Pt. 3: Increase Your Scope of Work
As operations managers in the field, we are tasked with creating daily plans on how much freight we can process based on the trailers available for inbound, current inventory levels for warehousing, and the number of outgoing orders we have to fulfill that day for outbound. We look at the available work and the staffing we have to do it. From there, we need to speak to how we are progressing to the total day plan and explain any major variances.
To do this, we need to know what the available work is - how many trailers are in the yard? How many cartons are on there? How many different items? What kind of freight is it? What are the downstream impacts to warehousing and outbound? And then I need to understand my staffing - how am I going to work each part of the process from receiving to inventory moves in the warehouse to outbound?
To do all of this effectively and be a leader of people there needs to be a degree of automation that you introduce into your job. Utilizing spreadsheet tools and/or programming to automate repetitive tasks can expand how much time you have to run your business effectively and manage your people. Instead of manually calculating our throughput, you can create a tool that calculates hourly throughput by job function so you can understand where and when overperformance or underperformance is located.
Similarly, I can pull historical team member performance to understand over the last 2 weeks to answer the question: how should I staff my team? Who is a top performer in each job function and how does that impact how much backlog I’m able to clear every day? By understanding team member performance by job function, I now have something rooted in data to make my plan for the day. The more realistic my plan is and the more I can reduce variability to what my plan, the better understanding I have of my business.
Increasing the scope of your job means removing tasks or obstacles that you might encounter daily. By automating these repetitive daily/weekly/routinely tasks, you can elevate the scope of your work rather than getting bogged down in executing mundane tasks.
By standardizing how we problem solve and communicate information, we can become more consistent and dependable partners. And when we make missteps, explaining our assumptions based on data becomes critical. Analysts and executives alike need data for the next iteration of problem solving since rarely is the first attempt the perfect solution. But by leveraging the information and overlaying key business context, you can successfully drive results and stand out from the crowd as you grow your professional career in data analytics.
About Jai Ugra
A young professional with industry experience in Supply Chain Management, Crisis Response, and Data Driven Decision Making.
Working for a leading Fortune 500 retail corporation, he has managed supply chain and logistics for a USD $850M portfolio. Alumni of the McCombs School of Business, The University of Texas at Austin with experience in data analytics and visualization.