Much more than a market hype

Part 2: The challenges of multi-period markdowns and the use of AI

26 April 2024

Oded Omer

Following our last month (April) publication which depicted “the basics of multi-period markdowns”, we will expand this month with some of the challenges in the core-existence of multiperiod markdowns. Then, next month (June), we expect to cover the last topic in this series which is “The holy grail”.

So what are the main challenges of markdowns optimizations? And what are the challenges in multi-period ones?

1. Dynamic Pricing Complexity

• Multiple Interdependent Variables: Factors like remaining shelf life, product-specific decay, cross-competition, inter-competition, shelf-capacity, timing, buting prices, original “catalog” price and other factors are some examples that are all heavily influence the right markdown.

• Data Overload: Effectively managing the massive amounts of data required for true dynamic pricing is staggering, especially in a rapidly-changing environment that continually changing demand due to seasonality, and again, many factors; as well as store-specific factors.

  
  

2. Short Shelf Life and Perishability

• Predicting the Sell-Through Date: One of the biggest challenges is accurately determining when an item will reach a point where it's no longer desirable to consumers. Slight variations in temperature, handling, or initial quality mean even a single batch of produce can have vastly different lifespans.

• Balancing Waste vs. Lost Sales: Setting too steep markdown too early can erode profit margins. But, wait too long, and the product might spoil before it sells even at a discount. This delicate balance is constantly shifting for each item. Of course, multi-period markdowns fixes that, but then the question how should it look like. We will discuss that in our next month issue.

  
  

3. Operational Difficulties

• Inconsistent Implementation: Manual markdown processes rely on individual judgment calls. This leads to varying discount levels and timing amongst staff and across stores, hindering the benefits of optimization.

• Timing Mismatches: Staff might not be able to identify and markdown items the moment they become optimal candidates, particularly during peak shopping times.

• Increased Workload: Changing prices incorrectly on fresh items can create surges in printing/replacing labels, communicating with point-of-sale systems, and addressing potential customer inquiries. This is however something we, at wasteless have multiple fixes for.

  
  

4. Consumer Perception and Fairness

• Trust: Dynamic markdowns across stores, with irregular patterns should follow logics and reasoning. They must be done correctly, then they can really win further consumers loyalty. (Opportunity)

• Associations: In the past, during 1950’ customers negatively associate markdowns with lower-quality. This is now less relevant, but a correct association of pricing to product is still needed. For example; if a discount is not optimized and too deep “there may be something wrong with this product”.

  
  

5. Technological Limitations

• Legacy Systems: Supermarkets rely on systems that weren't built with real-time dynamic pricing in mind. These often have trouble handling the huge data input and speed required for fresh food markdown optimization.

• Integration Challenges: Even with newer software, integrating it with existing point-of-sale systems, inventory management, and other platforms can be a hurdle. Knowing that, Wasteless has developed, an independent solution that narrows down the initial integrations to none.

• Perception Barriers: Sophisticated markdown optimization, especially those harnessing the power of AI (Wasteless), can be a significant change for “old school” retailers.