3 Common Forecasting Software Issues and How to Fix

You Are Probably Experiencing these Forecasting Problems

Demand Forecasting should be a forecast of future needs for your customers. Most people need to notice the little things wrong in their forecasting software that result in expensive overstocks and out-of-stocks. Three Common Forecasting Software Issues are:

1. Integer-based Forecasting (whole number forecast)
2. Companies Not Using Big Data in their Forecasting Math
3. Demand Planning Software used for Demand Forecasting

Forecast is a whole number, not a Decimal. What?!?

1. Every Forecast Period is Wrong to some degree. Those ‘degrees’ of error simultaneously result in too much safety stock and outs.
2. Somebody cannot effectively use whole-number forecast for slow and intermittent demand products. Worse, forecast systems using time series or time series with regression analysis will never correctly calculate slow and intermittent demand forecasts. Forecast errors require large, costly human interventions; some call it job security. Also, the poor forecast creates costly overstocks and out-of-stocks.

Run the math for one example: a product needs 1 a month in a four-week month.  Run your safety stock calculation using .25 for a weekly forecast for 4 weeks and select one week to have sales 1, the other weeks’ sales 0. Then, calculate your safety stock using three weeks of forecast 0 and one week of forecast 1 against the same sales pattern.  Now multiply that result across the thousands of products with Slow and Intermittent Demand.

Forecasting Software is Expensive to Fix, Don’t Expect a Fix

Does your wholesale or retail forecasting software use a whole-number forecast? Research the origin of the software you use. Someone probably built the software for manufacturing. Most people fail to realize that when a software program architecture considers a forecast as a whole number, forecast errors are expensive. When a software company designs around a forecast as an integer (whole number), converting the software and current customers to a decimal-based forecast is often financially impossible.

Integration of Forecast into Customer Other Data Creates a Larger Block to Software Updates

Hundreds of calculations and reports use that whole number forecast; to change to a decimal forecast is cost-prohibitive. The lines of code through the software program that expects a whole number will inflate the cost of updating the code.  Add to that complexity the fact everyone using that software creates reports and other data with the whole number forecast. Those reports and analysis tools need updates to use integer-based forecasts. The complexity and cost are immense; getting marketing to write materials that turn attention away from the integer-based forecast error costs is far cheaper.

Misses in Big Data, Your Forecasting Algorithms Don’t Know Type of Sales?

Does your forecast algorithm use sales data broken out by sales type? The standard sales types are regular, lost, promo, and closeout.  If your Forecast System only uses total aggregate sales, it’s a sales forecast model, not Demand.  To be called a Demand Forecast model, the algorithms must be able to accept data on the entire sales picture, not one type of sale.

Expensive Lost Sales issues in some Forecast Systems

1. One month’s out-of-stock will be added to next month’s Demand. While this might have worked long ago, in the digital age, people alternate sources. When you are out, they get products from another place, and customers don’t wait until next month.
2. Lost Sales data is only a KPI – meaning the user must adjust the forecast manually after a reforecast to fix the lost sales issue. A KPI tells you to fix something that an actual demand forecast solution would have fixed already.

Demand Planning is not Demand Forecasting

Know Before You Buy any forecasting software the answers to these three questions.  Is your forecast integer-based, a whole number, or no decimal?  Many retail and wholesale companies bought forecasting software designed for manufacturing, Yikes! Add the issue where the type of sale is not considered in the forecast algorithm, and the re-forecast can drastically impact business. A Sales Forecast is generally higher than Demand. Why ever talk to a salesperson who said I plan to sell less than last year? A sales forecast will be higher than Demand most of the time.  Demand Plans are designed around a store, floor space, and – the big point – unlimited inventory.  Have you ever seen a plan that includes lost sales and closeouts?

The Digital Age, the Internet of Things and Demand Forecasting

The digital age and the Internet of Things continually change how people buy goods.  Reviewing your forecast tools and finding the right tool for your business makes sense.  At Data Profits, we talk with many people about Demand and how Demand Forecasting should correctly impact replenishment.  Call us for ideas and or ask for a demo and see how our iKIS solution can improve your sales and inventory in less than 90 days.

• About
• Latest Posts

Stuart D

Stuart Dunkin is a strategic visionary who leads by calling on his broad range of experience as a former retail executive as well as a consultant for top retailers and for the E3/JDA software company. He founded Data Profits to solve a real-world problem plaguing retailers today – the disjointed relationship between supply chain data, people and business goals. He has experienced first-hand the challenges and frustrations that occur when decisions are made based on wrong data, which often leads to the wrong actions being taken with crippling results like overstock, out-of-stock, and increased working capital.
Stuart graduated from Auburn University, with bachelor of science in business administration.

Latest posts by Stuart D (see all)

• Demand Forecasting: The Ultimate Secret for Your Organization’s Success - August 7, 2024
• How to Avoid Carrying Cost Mistakes in Inventory Optimization - June 10, 2024
• 3 Common Forecasting Software Issues and How to Fix - May 20, 2024