May 3, 2017
Hidden Causes of Excess Inventory (3 of 5)
Is customer forecasting creating excess inventory?
When we discuss forecasting with prospective clients, one common topic is customer forecasts. Many planners are adamant that they should create forecasts per customer, per product, per warehouse. The technology is available to make something like that work, but is it the best approach to planning?
Ultimately, this customer-level forecasting is a tool companies use to hold the sales team accountable or to get customers to commit. Those objectives are completely understandable, and sometimes may even feel necessary, but what if it affects the buying process negatively? What if the result is excess inventory from overly-enthusiastic sales projections?
Let’s break this down a little more. Say we have 100 customers we want to get information on. We have five warehouses and an average of roughly 200 products involved here.
That makes for simple math: 100 x 5 x 200 = 100,000.
That’s 100,000 product level forecasts to be collected. If that task is spread across 10 sales people, each salesperson would be responsible for 1,000 forecasts from customers every month. Even if only took five minutes per forecast, that would take nearly 85 hours per salesperson. That’s half the month right there. Is that really what the sales staff should be doing? And what is the reliability of each forecast?
Some businesses rely on customers to submit their own forecasts, usually on a spreadsheet. What is the quality of that forecast? If you were the customer and you want to ensure there will be enough stock when you order, would you maybe round up a little on your forecast, just as a buffer? And what if, say, half of the other customers filed similarly inflated forecasts?
It’s pretty easy to see how this process lends itself to inaccurate forecasting and how those inaccuracies can quickly become costly excess in the supplier’s warehouse.
Another option is using a forecasting engine to crunch the numbers, which is certainly better than asking for customers’ best guesses. With any statistical engine, the results are only as good as the inputs. Smoother data creates more predictability.
There are a lot of articles out there that suggest starting at a highly aggregated level to produce a better result. Think about this for a moment. If a customer is buying product and the rate of sales are in the single numbers per month, they could have an ordering pattern of 10 every second or third month, give or take. That is going to create a horrible forecast unless you have some causal based model that proved to be predictable. Now multiply that by all customers and you end up with a bunch of bad forecasts.
So what’s the takeaway on customer-based inventory forecasting? If it’s tool to keep the sales team striving for their monthly goals or to keep customers engaged, that’s fine. In fact, those figures could be helpful when building forecasts. But to use those numbers in place of a forecast is a recipe for excess inventory, month after month. Using a statistical forecast is a far more reliable method of planning, and the customer forecasts can help provide extra analysis and context to keep everything running smoothly.
Remember, the key is a balanced warehouse, where both excess inventory and stock-outs are kept to a minimum. If customer forecasting isn’t doing the trick, it’s time to consider other inventory planning tools.
See the next lesson: Hidden Causes of Excess Inventory (4 of 5)