
What went wrong?
One of the extensive stress tests for every supply chain is a holiday, and probably none more so than Black Friday. Especially in these times with severe supply chain disruptions, it is harder to get it right than to get it wrong.
But let’s assume for a moment that your supply chain was not disrupted, perhaps because you source locally, manufacture in-house, or your raw materials are not affected by a shortage. (If that’s the case, lucky you!) You were able to order and stock precisely the way you wanted, but the big day came and went, and it all looked completely different than you expected. What now?
You didn’t stock enough and sold out.

Anyone new to stocking and demand planning might think that nothing feels better than sitting on a stack of empty pallets in front of an empty warehouse while watching the last few customers drive off the lot. Selling out is generally a good problem to have, except for that nagging question in the back of your head: “Could we have sold more?”
Which leads to the question: Is it better to have zero units left in stock at the end of the day or to have one unit left in stock at the end of the day? Instinctively “zero” might come to mind but that is not necessarily true:
If you have one unit remaining, you know for a fact that you stocked exactly one more than there was demand for.
Good job. If you have none left, it could be that you were dead-on, but you actually don’t know! It could very well be that you would’ve sold ten more if they would’ve been in stock. Could have shifted another hundred. Shoulda, woulda, coulda - with zero left, you simply don’t know; with one left, you have a precise result.
A precise result is also what you get from Remi AI’s demand forecasting engine. Based on historical demand and other patterns like weather, price changes, and, yes, seasonality, our Artificial Intelligence algorithm will predict what actual sales will be on a given day. Whether you want to cut it close and set your minimum stock to a single unit is up to you, and you can configure that. But everything else is determined by the forecasting engine, Black Friday included.
You stocked way too much, and you’re sitting on inventory now.
Unless you’re selling fresh donuts or similarly perishable goods, having too much on hand is the lesser evil because you can always sell it later. But products that are sitting idle in the warehouse make nobody happy. Enter the fire sale, which in some places is as predictable as a Tuesday following Monday. It is evident that all these “50% OFF” sales eat into your bottom line and could even destroy any gain made on Black Friday.
Here is what to consider if you’re not sitting on empty pallets but on boxes upon boxes of merchandise:
Make sure your inventory costs are correct, not just on a global level but on a per SKU and marginal situation. What does that mean? Most companies place a simple, single figure on stock-keeping cost: 0.8% of the sales price per month, or 50c per kg per month, or another easily determined factor that roughly reflects your reality. And in general, this makes sense, but to make decisions for clearing stock, it can be misleading for many reasons:
Stock-keeping costs are often long-term, and reducing stock will not reduce actual warehouse costs, at least not in a linear manner. (You still pay rent for an empty warehouse.)
The actual savings can vary significantly by SKU. A cheap, bulky product will have a significant impact on cost, while a small, expensive product will not play a big role - you can keep the small, expensive product in stock longer without incurring much additional cost.
Generally, you want to focus on the Working Capital aspect of inventory. If you’re strapped for cash, your excess inventory is an excellent source of liquidity. But don’t pay too much for that cash! With interest rates generally low, there might be better ways to gain that liquidity than by emptying your warehouse.
In the end, it all comes down to the proper discount. Tweak it just right for each product so that you get the maximum benefit from consumers’ price sensitivity. Remi AI’s Price Optimisation Module does just that. You might be vacillating about whether to mark a given SKU as “20% OFF” or “30% OFF”. The Price Optimisation Module will tell you (for example) that 16.89% off is the exact sweet spot where the combination of increased volume and reduced margins results in the highest profits. And with our numerous integrations, you don’t have to maintain any of that manually. Review the suggested discounts in Remi AI, and with the click of a button, they can be directly applied to your ERP or e-commerce system.
You stocked in-store, but everybody shopped online.
This is undoubtedly better than the obverse, and also a much more common occurrence these days. The suggestions for next year might seem obvious, and also depend on the size of your business. One such trick is to include supply chain visibility along the entire chain, including in-store stock level, updated in real-time from your POS system. Most modern POS and inventory systems can handle that. Remi AI's Replenishment Module can optimize stocking and ordering decisions along a multi-echelon supply chain with multiple warehouses, distribution centers, and sales locations (stores). Even if you are not fully integrated and only control the distribution elements, it is often possible to integrate your customers’ systems and gain visibility of their end - the benefit is for both of you.
Lastly, even if you don’t have a large, sophisticated system but are in an earlier stage of building your business, you can benefit from this approach: Instead of frantically refilling your online fulfilment locations, divert some online sales to your stores to fulfil. It might require some cross-training of your store staff, but it can help tremendously to balance online vs. store-based demand.
You sold out of some items and have excess inventory of other items.
If that is the case, chances are that your planning or business intelligence is not granular enough, either on an SKU or location basis. You might have said, “last Black Friday, our volumes were up 15%, so let’s order 15% more of everything. With Remi AI's Demand Forecasting Module, you don’t need any such simplifications, because the platform forecasts at a SKU level. This microscopic attention to detail, as well as integration with your ERP, means the platform acts as your deputy - trusted with the important tasks that need to be done but would be a sub-optimal use of your time. To illustrate: Product A may have sold 18% more, but Product B only 11% more - if that is the case, place your orders accordingly and specifically - not just a rough average. Remi AI can track 10,000s of SKUs across many locations and learn both on the SKU level as well as on the aggregate product level.
You stocked well and sold well, but your shipping costs went through the roof.
Finally, all the frenzy around peaks like Black Friday sometimes leads to rash decisions: “Quick, another truckload of widgets! Overnight!” Do that a couple of times, and the actual cost of a widget may quickly rise above its estimated bookkeeping cost. While it requires some longer-term planning, Remi AI’s Container Optimization module can provide real relief, and represents the next logical link in the supply chain.
Especially for bulkier products like furniture or landscaping equipment, optimally filling your trucks or shipping containers can drive significant cost reduction. Taking advantage of this leading up to Black Friday can keep your costs under control, even if everything else is going crazy.
Summary
While Black Friday presents many unique challenges to a supply chain, today more so than ever, and Remi AI’s suite of Forecasting and Optimization tools helps you to keep control of stock levels and price points, even when you face much uncertainty. Additionally, it does so on a granular level that lets you optimize complex decisions across thousands of SKUs without having to track minute details of every single product.
Want to find out more about Demand Forecasting? Why not have a read through of our case studies, where you’ll find out how we’ve used demand forecasting to help increase stock availability and improve the accuracy of forecasts. Or, check out our blog for the latest AI reads here. Once you’ve sated your hunger for knowledge and you’re ready to take your Demand Forecasting strategy to the next level, drop us a line here.