This post has already been read 17184 times!
For decades, retailers have been managing two supply chains: one for when a product is selling at regular price and one for when a product is on promotion.
Back in the day, there was certainly some merit to this approach. Sales volumes (and by extension, order volumes) during a promoted week could be tenfold that of a nonpromoted week. To the extent that a retailer represented a large proportion of a supplier’s production schedule, it was only prudent to place promo orders on a much longer lead time than regular orders.
Some retailers went so far as to have separate logical inventory buckets in their distribution centres – one would hold promotional inventory and the other would hold regular inventory.
On the surface, it seemed to be a neat and clean way to manage things. One system would forecast regular sales and handle the automated replenishment when an item wasn’t being promoted.
A different system would do the ordering and “push” allocations for promotions. Along the way, every inbound DC order would be tagged as being either promotional or nonpromotional. If a store needed to replenish regular stock, a pull order would be triggered to the DC. Promotional push allocations would be used to get product to the store for promotional weeks.
In summary, neat little bows around inventory and in-transit orders made the supply chain easy and seamless to manage.
Or did it?
The ability for this approach to work was predicated on two assumptions:
- That every regular and promotional sales forecast was virtually 100% accurate at item/store level
- That every store and DC had the “right” amount of inventory at all times
What seems like a “neat and tidy” planning approach gets very complicated whenever these assumptions don’t hold water. For example:
- What happens when a store runs out of “regular” inventory on the shelf a week before a promotion, even though they have tons of “promotional” inventory in the back room?
- Do you allocate “promotional” stock to a store that is sitting on 6 months of “regular” supply because they didn’t sell through their allocation from the last promotion?
- What happens with “regular” replenishment after the promotion is over and all of the left over “promoted” stock is visible to the system that plans regular replenishment?
The underlying problem with this approach is that the underlying nature of a product doesn’t change when it is promoted. Nor does the underlying nature of inventory or orders. In short, the notion that we have 2 separate supply chains to manage is a fallacy.
What’s really different during a promoted week as compared to a regular week? Doesn’t it really just boil down to the fact that:
- Sales are (hopefully) expected to be higher?
- Minimum display requirements may be larger?
- A pre-distribution of promotional setup stock may be required?
Instead of trying to “tag” orders and inventory, can’t we just manage these quantities and calculate a continuum of planned replenishments over time (some larger than others during promotional periods)?
By sharing a forward looking stream of orders to the downstream supply chain (including suppliers), large promoted quantities are automatically reflected in the plan, so additional lead time for promotional ordering is no longer necessary.
There is one supply chain and one process to manage it. No more manual work (and re-work). No separate ordering systems for regular and promotional replenishment. No more complexity in the execution systems to separate stock.
No more neat little bows.
- Small Data, Big Insights - July 14, 2016
- Is Flowcasting the Supply Chain Only for the Few? - June 3, 2016
- Philosophy - February 12, 2016