I reviewed order management flow with a client some time ago, as part of a process capture exercise within a re-engineering program; it was within a business to consumer context. As orders flowed during the day, the local teams would review inventory manually, and then when it reached an appropriate reorder point, they would manually enter an order in their ERP environment. That order would then be printed out, and Faxed to a central warehouse. The central warehouse would re-enter the order in their ERP environment (which, by the way, was the same one), and an interface file would be generated with shipment parameters.
To make a very long story short, the order continued to be ‘touched’ several more times by humans until it was actually received in the destination warehouse to replenish stock. If something went bump, it was a manual process to adjust all the numbers in all the interfaces along the way. Keep in mind, this was essentially an intercompany inventory replenishment signal. It reminded me of a memoir of a child of a dysfunctional family – what seems strange to the reader, to the child of course was completely normal. I was wracking my brain to figure out why – why there were 12 steps or so, why the same process was performed over and over.
Part of it was that the main warehouse considered the consumer sales warehouse team “a customer”. The process for exceptional shipping direct to a customer, and the process for shipping to their own warehouse were one and the same. “You act surprised,” was the refrain I heard, but I wanted to follow through with a structured change to their business, not a leap to conclusions. “I am surprised.” I finally said, as we looked at developing a “to be” process. You see, I’m used to seeing a philosopy of “Zero, One, Many” when it comes to orders. What does that mean?
An order which directs inventory movement should be a zero touch order, an intercompany order. It should be completely automatically generated, scheduled, and received. Why? Because there is no special requirement for quotation, inventory reservation, confirmation, all the processes which surround a revenue-bearing customer order. It’s merely an inventory balancing problem, and should incur the lowest possible cost. Strangely, where you have good connections to customers – like a channel partner – that also becomes a zero-touch order with demand-pull signals driving order creating (if needed) and inventory transfer.
With an imperfect connection to the customer, the philosophy should be one-touch for stocked and even build-to-order materials. The order should be entered by the customer into your order management solution, automatically scheduled, confirmed, shipped, invoiced. At the worst case, if there is some exception, then it can take on an additional “touch” for work, but keep that to a minimum to incur least cost.
Only where there is a “high touch” requirement do you have a “Many touch” order – Engineer to Order, for instance. You may not receive an order with a SKU – it may only be a material specification. You need a quotation, you need to check design, manufacturability. Engineer may involve a complete program – building a skyscraper – where day-to-day adjustments to schedules and requirements must be considered within your own company program. Coincidentally, many-touch products are quite expensive relative to a standard material shipped on a replenishment signal – you have to pay for all that TLC.
My takeaway for you is to consider if you have problems with reliability, cycle time, inventory, cost, oh, and flexibility – that have you sized the complexity of your order management process correctly for the market you are serving – internal, channel: zero; stock, build-to-order: one; custom solution: many. If you haven’t you may be as surprised as I was.