There is no debate that Demand Forecasting and Replenishment is at the core of the Customer Experience and Supply Chain of any company. With so many recent developments in providing convenience and great experiences to the customer, forecasting demand is becoming increasingly difficult. Due to this shift in customer expectations and retailer responses, we see three main challenges that stand out.
- Identifying whereto attribute customer demand and forecast future demand in an Omni-channel environment
- Deciding how muchinventory to hold to meet your customer demand based on the desired customer experience
- Defining the impact of various events and determining how these events will impact your future forecast
Let’s dive into each one in more detail.
Identifying where to attribute customer demand and forecast future demand
The concept of “Omni-channel”, or “Unified Commerce”, or “Multi-channel”, or (insert retail buzzword here), is not a new one to anyone in the Supply Chain industry. There are more options today than there ever were for customers to receive their products. In a traditional retail supply chain, there is a simple issue of forecasting demand at brick-and-mortar stores. Now, retailers must not only forecast brick-and-mortar demand, but they must forecast eCommerce, Buy-Online-Pickup-In-Store (BOPIS), and many other channels.
Let’s take a simple example of a BOPIS scenario to illustrate the complexity. You are at work and decide that you need to have a certain pair of shoes for an event tomorrow. You go to the company’s website and see there is one pair left in your size at a store that is on your way home. Since you don’t want to wait, you select a BOPIS option for the store on your way home, not a store location you usually visit.
Now, when the company is trying to attribute that demand to drive their future forecast, where do you think they attribute it? The store you picked up the order in but rarely visit? An umbrella bucket for all eCommerce orders? What if they had to ship this from another nearby store to get to the location you selected for BOPIS?
As you can see, there are a lot of different flavors of just this one scenario that can affect how you attribute customer demand in an Omni-channel environment. Deciding where to attribute the customer demand will ultimately be a driving factor in your replenishment plan and fulfillment strategy.
Deciding how much inventory to hold to meet your customer demand based on the desired customer experience
After tackling the challenge of how to attribute customer demand, retailers must find an inventory balance based on their overall customer experience strategy. Service levels should be looked at as a strategic decision that are based on the wants and needs of the customer.
Most retailers look at key metrics that include sales throughput along with margin to determine which SKUs (product/location combination) are the most profitable to hold, but sometimes these metrics don’t fully capture the customer experience. Certain products or product groups that are important to the customer should be important to the company as well. If they aren’t a priority and the retailer is out of stock on these SKUs, then the customer could ultimately take their business elsewhere.
Another challenge that directly correlates to service levels is safety stock. In short, service levels dictate what your safety stock positions need to be in order to meet service level goals. However, the structure of the overall supply chain and fulfillment network can be set up to help reduce the total safety stock needs while still meeting the goals.
Many retailers are moving towards more of an urban fulfillment-type model in which there are urban fulfillment locations that have short lead time to service their retail locations and end customers. These urban fulfillment locations allow retailers to aggregate the demand across a group of stores or geographic areas and reduce the overall safety stock they are holding at each retail location to account for variation.
This safety stock concept can also apply to combining retail and eCommerce channels into a single pool of inventory. Instead of maintaining multiple inventory pools with individual safety stock requirements, retailers can combine/aggregate these inventory pools to reduce the variation and overall safety stock requirements.
Defining the impact of various events and determining how these events will impact your future forecast
Finally, there is a plethora of outside factors that can affect the forecast of a SKU and deciding on how to handle these factors is crucial to forecast accuracy. The first step is figuring out if an event is more likely to be a recurring event or a non-recurring event/outlier.
If Beyoncé decides on a whim to post something on Instagram about a new nail polish she found (not a planned marketing post) and the “Beyhive” rushes out to buy every last one of those nail polishes off the shelf, that may be a one-time event unlikely to occur again. However, if every President’s Day a retailer has a sale that results in increased sales, this would be something you would want to consider when forecasting this time period in the future. These two examples are clear cut, but there are times where the line isn’t so black and white, and the decision making can be more difficult.
The next step is determining how this event will affect the forecast. For a non-recurring event, the most common practice is to cleanse this from the demand history so that it protects the future forecast. For recurring events, the next year could have an implied uplift or decreased forecast, the dates of the event could shift (take Easter for example), or the promotion/event could be discontinued. All of these factors need to be considered when dictating how a recurring event will affect the future forecast. It becomes more challenging when the retailer is one that has a jam-packed Marketing calendar with overlapping promotions that affect various products and categories.
The retail landscape has changed and demand forecasting and replenishment is more complex than ever. Forecasting the Omni-channel customer is no easy task, but it is crucial to creating that customer experience every retailer strives for. I hope that some of the above concepts and ideas can get you thinking about the different aspects that lead to the success of forecasting this new age customer. Drop an email at firstname.lastname@example.org to learn and discuss more.
Contributor: Kevin Quigley