But how should companies decide which ones to use, and how can they get the most out of the technology?
In many cases, as with POS, it depends on the nature of the business and what its sales cycle looks like, according to Simon Ellis, practice director for supply chain strategies at IDC Manufacturing Insights, based in Framingham, Mass.
It also depends on how frequently the data is needed. “Do you need it on an hourly basis or do you need it on a daily basis or weekly basis? I just think it depends on what you’re using it for,” Ellis said.
If, for example, a company expects to sell about 10 units an hour of a particular product and a few hours go by without any sales, it’s likely that’s because the product is out of stock, he said.
“If that’s your application, then you need that frequency of information. You need hourly, or even half-hourly, increments,” Ellis said. On the other hand, if you’re trying to estimate when to make your next delivery to a customer and they need restocking every 100 units but only sell three to four a day, real-time point of sale probably isn’t necessary.
“You probably only need daily point of sale data,” Ellis said. “I think it’s easy to fall into the trap of paying for levels of granularity that you ultimately don’t need.
One of the pieces of advice that I always give companies is, ‘If you’re going to pay for the capability to have hourly point of sale data, make sure you’re really using the point of sale data. Otherwise, get it daily or weekly.’ ”
The key role of demand signal repositories
Demand signal repositories, central databases that provide for the daily tracking of goods down to the store and SKU level, are one of the most important technologies in the demand-driven supply chain, according to Ellis.
“Companies can’t be effective in terms of leveraging point of sale data without them,” he said. “Having all that information in one place, that sort of notion of having one single version of the truth, is very valuable.”
But getting to that single version of the truth requires consistent data governance, he said.
“You have to put in place the right kind of data governance. It’s tedious, it’s not sexy, it’s boring. But it’s critical,” Ellis said. “You have to make sure that every piece of data has an owner, someone who’s accountable for the accuracy of that data, and that you’re following acceptable industry standards.”
Don’t overlook ERP
One of the most common ways to manage a demand-driven supply chain is through an ERP system, according to Cindy Jutras, an independent analyst and former vice president at Aberdeen Group.
In such an arrangement, the demand history resides in the ERP, along with related records of customer orders, shipments and inventory transactions. For manufacturers, that’s also where the bill of materials also resides.
That information is then routed through the demand planning application, which in many cases exists outside of the ERP system, Jutras said. “In days gone by, it was unusual to have that kind of functionality in the ERP. That’s changed.”
Some companies may still opt for a standalone demand planning application because they require functions that the ERP doesn’t provide. In either case, demand information is then routed from the demand planning application through the ERP, where action can be taken on the next steps in the supply chain, such as processing of purchase orders.
One of the advantages for managing the demand-driven supply chain in a company’s ERP is that there are fewer chances for data to get lost, Jutras added.
“There are certain limitations to doing this outside of your ERP because there are some connect points in terms of your inventory,” she said. “There’s also some connect points in terms of the actual demand and whether it’s coming from the sales forecast, which is oftentimes in a sales force automation solution, or it might be transferred to your demand planning in either the supply chain or your ERP.”
In the best-case scenario, Jutras said that a company’s past sales records, current inventory and sales information and forecasts should all reside in the same place within the ERP so that “everything is funneled together,” she said.
“Very seldom are they fully, transparently integrated. That’s the best of all worlds.”