Free Demo menu close

Reducing Excess Inventory

May 3, 2017


Hidden Causes of Excess Inventory (1 of 5)

Bad forecasts lead to excess

Let’s start with the leading cause of excess inventory: bad forecasts. Lots of management types skip right past this one, thinking they don’t really use forecasts in the first place. But further investigation almost always shows that they’re using a simple sales average or static ordering model. Sure, those are basic calculations, but they qualify as forecasts — and they’re rather crude forecasts at that.

Since everyone forecasts in some fashion or another, let’s look at how a forecast gets made and implemented. We like to break it down into two basic sides: first, the forecast engine that does all the grunt work, and second, the process and tools used to quickly identify forecasts that are off the mark. One without the other is simply not possible.

One important point before we dive in further is that forecast engines reach the point of diminishing returns. We’ve seen companies spend huge amounts of money adding incredible layers of complexity to their engine to account for every possible scenario. The end result is always an engine that no one on the team understands, which inevitably leads to an engine that no one on the team trusts. If you can’t comprehend or trust the forecasts you’re generating, what’s the point?

That’s why the tools and processes are so important. A forecast engine will only be so effective—no matter its complexity—so it’s far more efficient to employ a solid engine with an effective series of checks and balances to keep everything tidy.

So with that said, let’s start on tools. Earlier, we discussed a basic sales average as a rudimentary type of forecast. Such an average does not detect or address seasonality or trends, which are key components to many forecasts. Using a spreadsheet to suss out that kind of data can be quite complicated. It’s also important to address the concept of a “tournament” or a “best-fit test,” which compare the various formulae that an engine can use. That’s what allows an engine to pick the best algorithm and range of parameters needed for a specific item in a specific month. Be careful in editing the formulae or changing the parameters, as slight changes could have major ramifications. Some people will even edit the historical data to game the engine, but if the engine worked correctly, that shouldn’t be necessary.

The process required here means getting the right people involved at the right time. The sales team plays a key role to provide updates to the inventory planner. A word of warning, though: please do not use sales projections as a forecast for purchasing stock! The same holds true for budgets, which is a similar type of input that’s usually on the optimistic side. Both are quite likely to be out of date quickly. They can provide useful context, but they shouldn’t be blindly followed.

We suggest starting each month with an overall review. Here are some aspects to look at:

  • Compare sales against past forecasts overall and by group
  • Look at over- and under-forecast percentages for recent months
  • Identify any products or groups of products that seem to be behaving unusually (and loop in more senior team members to gather more intel here)
  • Compare the top over- and top under-forecasts from the last three months with the corresponding sales figures
  • Examine new products and intervene manually if data is lacking

It’s important to get these steps out of the way before placing any major orders for the month. During the month, it’s important to check in on items where the run rate is ahead or behind the expectations. For example: if there’s an item with a forecast of 100 units for the month but there’ve been only a few sales after a couple weeks, something may need to be adjusted. Check in on these exceptions in week two and three of the month to help align the forecasts.

Obviously, forecasting comes with a set of challenges and headaches, but a smart forecasting engine aided by some basic processes for monitoring goes a long, long way to a more balanced inventory.

Without the right balance of an inventory forecast engine and an on-going process to check the forecasts, the warehouse is doomed to stock-out at inopportune times. Most planners will just go with extra buffer inventory so they don’t have to worry about stock-outs, but that’s where the expense of excess becomes a cashflow killer. For businesses that balance the tools and process, the result will be a balanced inventory.

Does proper forecasting sound too complicated?

NETSTOCK simplifies inventory management with built-in inventory best practices.

Request a FREE demo