Order Allocation: Is your orchestrator already old news?

If you’re a retailer who has implemented an Order Management System, then you’ll agree that orchestration is a key component of omnichannel performance in terms of customer promise, cost and delivery time.
With the pandemic having raised customer expectations regarding digital commerce and accelerated the challenges retailers face in meeting them, it is important to take stock of orchestration performance to ensure that you are, and most importantly remain, at the forefront of available technologies to ensure that you are part of the leading retailers who have improved their operations over the past year.

Order allocation 1.0: The Waterfall

OMS have typically unified stock from the different storage points by sending a specific stock image to each sales channel and then orchestrating the resulting orders.  In other words, for each order, the solution will determine which stock point (warehouse, shop, supplier, etc.) is best suited to deliver the ordered items in compliance with a set of business rules defined in advance (stock level, back-up stock, stock point’s days and hours of operation, capacities, etc.).

At first glance, this process seems quite efficient because it takes into account all the stock points’ limitations in order to identify the best candidate to fulfil a delivery promise. However, your conductor may be orchestrating classical music while the audience is waiting for heavy metal.
Traditional OMS solutions work in a waterfall fashion, i.e. they will identify a specific stock point that can fulfil an order (e.g. a shop). This allocation is then sent to the shop, which has the option of accepting, declining or ignoring the order.

If no action is taken by the shop then, at best, the order will be allocated to another stock point within a few hours. However, this is not always the case and sometimes the shop has to manually decline the order so that it is allocated to another stock point…

When the order arrives in a second shop, the process starts again and can be repeated endlessly throughout the network, drastically lengthening processing times and breaking the delivery promise sometimes given to the customer.

Order allocation 2.0: Competitive order allocation

Thanks to our careful consideration of our clients’ problems, OneStock has invented the concept of competitive order allocation.
The principle is simple: allocate an order to ALL the stock points that can fulfil it in compliance with business rules (for example, offer the order first to stock points within 50kms of the delivery address). The first to accept the order removes it from other stock points’ listings. This is the same concept found in the majority of the most successful B2C service applications (Uber, Deliveroo, etc.).

If a store accepts an order for two items but finds during the preparation stage that one of the items cannot be shipped to the customer (e.g. the item is damaged), the store staff has the option of cancelling the preparation process and the order will go back into play to be picked up by another store. If the orchestration rules allow the split, it is also possible that the shop sends the first item and leaves the second one in competitive orchestration for the other eligible stock points in the network.

In this way, the order is constantly in motion, constantly seeking to be delivered in the most efficient way possible. This system is the most efficient one in existence today, but it can present some limitations.
If an eligible shop does not respect the rules of the game and accepts any order without even looking at their capacity to honour said order, the delivery promise will be broken.

In order to address this over-zealousness issue, which could be considered as cheating, OneStock has implemented an algorithm that calculates, for each stock point, the ratio of order capture to order abandonment, as well as the average preparation time. These two indicators are used in the orchestration to prioritise the most efficient and fair-play stock points in order to reduce the processing time of the customer’s order as much as possible.

No orchestration without store staff!

You can implement the best possible orchestration solution and algorithms, if your store associates don’t use it, the strategy falls apart. In addition to dealing with customers in the stores, a new logistical task is added to the salesperson’s day that they didn’t ask for. To change this perception, it is vital to motivate your teams. In order to ensure that orders are processed seriously and quickly, we advise all retailers to reward staff for each order processed.

For many years, e-commerce has been perceived as a competitor to physical store networks. However, with the emergence of unified commerce scenarios such as Click & Collect, Reserve & Collect, Call & Collect, Ship from Store, etc., retailers need to reconsider the models for allocating revenue to stores. Allowing sales staff to see web orders as an additional source of revenue.
If the change is guided and well pitched, transparent allocation rules can be very well received and sometimes even pacifying in highly political retail contexts such as franchise networks.

The reconciliation of the digital and physical world is no longer an option but a necessity to ensure optimal orchestration. The perfect combination of algorithms and people will guarantee the success of an Order Management System project. All the more so as, in the midst of the pandemic, it will take a lot of effort to empty the overflow of stock accumulated during the various lockdowns.