One effective method to mitigate panic buying during crises is by using predictive analytics to foresee potential demand surges. This approach was highlighted in an HBR article about how Kraft Heinz used COVID response data from Johns Hopkins University, which allowed them to anticipate areas of high demand and route their products accordingly (think more Mac and Cheese on the shelf). However, this approach requires sophisticated analysis and data integration into demand planning and forecasting processes, which might not be feasible for all retailers. While rationing can help distribute resources more equitably and prevent rapid stock depletion, it can also backfire. When customers become aware of rationing, it may heighten their sense of urgency and scarcity, prompting them to buy up to the limit, even if they don't need the maximum quantity. This applies to both consumers and retail customers, and can contribute to increased demand and exacerbate the very stock shortages that rationing was intended to prevent. Therefore, retailers should implement rationing with caution, and in conjunction with clear communication with both their consumers and retail customers about supply levels. This communication can help alleviate concerns of scarcity. Providing regular updates about when more stock is expected can help reduce panic and the impulse to overbuy. Retailers can also work closely with suppliers to try and increase stock levels during a crisis to meet the surge in demand.

I saw Simon deliver this presentation at the Supply Chain Symposium last week. The answer to this question cannot be applied in a blanket way across the retail industry. H said what is most important will vary based on questions every retailer or manfuctuer must answer for itself. Those 8 options were grouped into 3 categories, essentially investments to place bets on (pioneer), existing investments to scale, and investments to improve or optimize. Picking one from each most applicable for retail, I'd say Actionable AI (Pioneer), Industry Cloud Platforms (Optimize), and Supply Chain Integration Services (Scale). The unification of these three needs to be done with precision given the state of the world today.

One way this plays out for Ahold-Delhaize is with a cloud-based data platform they built called Fiona that helps them deal with the massive scale of data involved and the challenges of extracting the right insights from it. The agile process Mr. Kosla describes helped them build it, and organizationally it unlocks value by having data science workers embedded in the business and within IT collaborating to meet each business area's requirements. It's all part of the connected customer strategy. I was curious what technology was used to build Fiona so I asked ChatGPT: “Some of the key technologies used in Fiona include Apache Hadoop, Apache Spark, Apache Hive, Apache Cassandra, and Apache Airflow, among others. Additionally, the platform also makes use of cloud solutions such as Amazon Web Services (AWS) and Google Cloud Platform (GCP).”

The only thing certain about supply chains is that they will continue to suffer from poorly supported decisions that lack the timeliness or quality needed amid ongoing disruption and now, recessionary conditions. With few exceptions (leaders) most organizations are making calls on forecasting and procurement based on what can be done, not what is best. What is best is having access to the right data given your market, customers, and suppliers, and applying the right form of analytics to support the best decision. There is no easy button, but the fact is that few have mastered analytics value at scale in a manner like this. So, examining processes that suffer a lack of timeliness or are less than ideal because you can’t get the right data at the right level of detail, is a great first step to securing a future for your business in this new reality.

I would say this quote from Lora Cecere's research Feb. 2022 describes the problem quite well: "Unfortunately, excel spreadsheets drive 93% of supply chain decisions. The reason? The users trust the approach, and it is easier for the finance team. The problem? It is impossible to model the details of variability, constraints, and cross-functional trade-offs in a spreadsheet." So it's a skills issue for certain given the complexities of operating a competitive supply chain in today's world. To resolve this -- and I say so as a vendor -- any really meaningful decisions across supply chains will benefit from low or no code analytic software tools intended to pick up where spreadsheets run out of steam. The variety of systems, and internal and external data sources, to say nothing of the necessity to explore predictions and prescriptions where they create value, demands a more concerted -- and exec-led effort -- to upskill managers and analysts.

Great comments from the BrainTrust panel. I wanted to add some more context for the post. When I worked at Salesforce, I saw technology leaders and even CEOs demanding an inventory of tech dependencies on AWS with the intent of migrating away. As I mentioned in the post, I think the answer depends on the strength and defenseability of the retail business, but I would also add also it’s going to depend on the skills and resources available to the retailer. The reason Industry Clouds are in demand is they bring packaged business and industry domain expertise to help companies make a successful migration to the Cloud. Most retailers struggle with technical and analytical resource scarcity, which also drives demand for low or no code tools that expand the pool of talent who can apply advanced analytics like machine learning to priority problems. In the end, I think services organizations and third party solution providers with technical and business expertise are going to be a key bridge to bring any Cloud use case to fruition and improve executive confidence in whatever Cloud provider they choose.

I suspect if Macy's has turned a corner and can remain a competitive integrated omnichannel retailer, it has a lot to do with this reported in the NYT yesterday: “The big difference at Macy’s now versus previous times we dealt with inflation is we have this data analytics and pricing science at a much more granular level that is helping us make the right choices,” Jeff Gennette, the chief executive, said in an interview.

It's interesting that most BrainTrust comments don't mention the potential valuations of the separate businesses. That I suspect is as much the driver here, as opposed to forcing any change from the outside. This story reminds me of others predicting e-comm spin offs among legacy retailers to capture greater value, grow faster, etc.

So consumers want: More regular updates about shipping status (63 percent); More transparency on inventory (59 percent) and potential supply chain issues (54 percent). They are not the only ones - many retailers themselves are not yet able to solve for this internally, let alone pass the the information to the end-customer. Of course from a CX POV retailers absolutely should provide visibility like this, but they need to first solve for it internally to make better decisions that might avoid the situations to begin with.

As Ron writes to start, there as a struggle "to meet consumer demand for delivery, buy online and pickup in store (BOPIS) and other options last year as the novel coronavirus pandemic hit." Fixing that, even incrementally, in line with investments aimed at omni-channel marketing and promotional improvement, is arguably the only way to reduce the ultimate friction that's going to hurt many companies in the coming months: product availability. It's about time that retail and cpg look much more closely at embracing a CX that proactively and pre-emptively considers the capacity to serve consumer demand requirements at every step of their journey.

Are robotics deployed at Walmart’s regional distribution centers “a game changer”? This must be part of their $14B supply chain transformation and is a great idea to improve efficiency and responsiveness. If it’s not so much a game changer, it certainly will improve competitive position relative to Amazon who has since 2012 been using robots like this and has 350,000 of them today. What will this supply chain development mean for store operations at Walmart and for those companies competing with it? You may have seen the recent Sam’s Club test of “scan in store and ship to home” test news. Versus Amazon, robotics in the distro centers should help Walmart deliver a super-efficient buy anywhere/anytime model. So whether you shop in store or buy online and have product shipped or pick up in store, product availability is almost guaranteed. For competitors, the theme is automation and scaling with as much intelligence as possible the processes connecting demand to supply.

What challenges continue to limit the use of predictive models in supporting forecasting? I think models are already used widely, the question is more about improving their utility and value to the business. To that end, it’s about relying less on internal and historical data, and more on blending those sources with external and real time factors that might influence demand even before customers express intent. Has the pandemic offered any lessons around the reliance on historical data or the capabilities of machine learning? Clearly yes, relying on backward looking data creates a blind spot that hurt many companies and even those with strong analytics capabilities struggled to adapt their operational supply chain ML systems. The assumptions on which they were based didn’t recognize the huge variance in data versus what they were trained on when faced with mass panic buying. The net of all is that disruption is going to continue into the foreseeable future, so you must improve, but this presents an opportunity. Demand forecasting should arguably become a core analytic competency subject to the same test and learn cultural iteration of any transformational AI use case. The enterprise’s brand promise and entire customer experience ultimately hinge on decisions made at this level.

We all experienced this problem at the onset of the pandemic and sometimes there were no substitutions. Of course it still happens but the fact that Walmart has achieved 95 percent acceptance by customers of AI-proposed substitutions seems evidence enough that it works at scale. Low satisfaction of substitutions cited in mid-2020 had way more to do with supply chains caught off guard upstream of stores due to panic buying. Interestingly, data gathered by customers' accepting substitutions provides Walmart with insight into supplier diversification strategies that can leverage AI to predict disruptions in supplier networks to avoid this shopper friction to begin with. That simple example highlights why it's best practice to think about your AI use cases beyond siloed problems and rather across the business.

Even being a data warehouse and platform advocate, I can't help but think it best to respond in this manner, especially in the era of cloud and the ability to scale up or down based on your requirements. Of course there are many elements critical to the needs of meeting customer demand, and having data, insights and operations aligned to anticipating and serving customers is among the most important. Retailers and all industries have been chasing the value behind centralized data for a very long time, but only those able to pair technology investment with greater organizational and process changes have been able to succeed. That’s because new insights and prescribed actions fall down when layered into faulty process models and old norms. Few retailers have appetites for large transformational data projects with uncertain benefits and timelines, so point solutions remain the status quo for many. These offer some value but won’t likely lead to best-in-class results and organizational change takes time and is difficult. Best advice would be to lead with goals for improvement tied to specific changes in processes first, then leverage the best technology to serve the objectives of the use case. Proving this out over the course of a series of projects might show justification for the scale value available from a centralized data effort. If not already, try to make this a pillar of digital transformation strategy.

The pandemic’s disruption on labor markets and impact on consumer behavior have the potential to cause a lot of pain for retailers focused solely on looking at this issue from a product supply perspective. First, retailers need to quickly understand if customers will predominately continue shopping online for the foreseeable future, or excitedly leave home to shop in person. So, immediate consumer and shopper insights. Having been at Macy’s two days ago on a Sunday, I can say the latter at least for my local store. Many retail workers no longer wish to hold these jobs and won’t return to them, meaning competition for scarce workers and higher labor costs that must be managed. This will certainly hit holiday season performance and margin. Second, retailers need to adapt whatever demand planning capabilities are in place to be more predictive, prescriptive and adaptive, to incorporate as many leading indicators as possible of shopper preferences and behaviors. That’s where things broke down in early 2020, causing so many furloughs, layoffs and bankruptcies. It’s now less about crisis adaptation, and more about leveling up to what’s expected of a modern high-performance retailer. A huge challenge is going to be balancing rising labor costs with increasing shipping and fulfillment costs, as expectations for same day or even intra-hour delivery become a real thing. My advice would be to ensure supply chain functions have a seat at the CX table, if not already, and everyone is armed with the best cross organizational customer and business insight possible.