Sales Forecast vs. Demand Forecast — What is the Big Deal?

Sales Forecast vs. Demand Forecast — What is the Big Deal?

Retailers confused about sales forecast vs demand forecast

Despite the fact that forecasting is a backbone of most retail organizations, there is still much confusion about the definition of a sales vs. demand forecast (as well as their respective benefits and drawbacks).

All retailers use some form of forecast to anticipate the future, and the reason is simple. Without having an idea of how many items you will sell tomorrow, building a successful and lasting retail business is impossible. You’ll end up buying too much of the wrong inventory and not enough of the right inventory, making your customers angry and losing massive sales.

This is why retailers are becoming more and more data-obsessed, especially in regards to forecasting. A good forecast stands between being profitable and going bankrupt.

And yet, Forrester Research found that while 74% of retailers want to be data-driven, only 29% are able to connect analytics to action. In other words, most retailers are still guessing with their data. They don’t know how to use data to build accurate forecasts, nor how to leverage forecasting to build solid strategies.

How to choose the right forecasting software

Uncover how top retailers research, shortlist, and buy forecasting software to ensure ROI.


A major reason for this difficulty, as will be explained below, is that most retailers have still not grasped the difference between sales forecasting and demand forecasting.

Why is this distinction so important?

As we will explain, these two methods produce forecasts with a very different level of accuracy. This disparity in accuracy results in a massive performance gap between retailers who leverage advanced predictive analytics (like demand forecasting), and those that do not. The former are flourishing in the digital age, and the latter are rapidly sliding into the retail apocalypse.

Let’s examine why.

Definition of Sales Forecasting

Sales forecasting is an approach retailers use to anticipate future sales by analyzing past sales, identifying trends, and projecting data into the future. The simplest version of a sales forecast will look at sales in Store A during last year, assume a continuation of some multi-year trend for Store A (e.g. some percentage of growth or decline in sales), and project forward to predict sales in Store A this year.

Sales forecasting has been the standard approach to retailing from the beginning of the industry itself.

Although modern retailers have more data than ever (as well as new tools and business intelligence dashboards), and can easily leverage more advanced and accurate forms of forecasting, sales-based forecasting is still the backbone of most retail organizations.

Sales Forecast Pros and Cons

The Pros of using a Sales-based forecast:

This is a fairly simple approach that is easy to implement. Most retailers use a combination of their ERP, BI tools, and (in many cases) good ol’ Excel to generate a forecast based on their sales history. This means virtually every retailer, small or large, can build basic sales forecasts without investing into additional new tools or processes.

The Cons of using a Sales-based forecast:

The problem with this approach is that if you’re a modern retailer that sells through multiple channels (with multiple locations and distribution centers), a sales-based forecast will result in far too many exceptions.

What does this mean?

In simple terms, if you use a standard sales forecast approach for a complex organization, you will soon find a lot of inaccuracies in its predictions. Dozens of factors (that a sales forecast does not take into account) will need to be manually reconciled by teams of analysts to get a more accurate picture. Some of these factors include:

  • New product launches: If your retail store experienced a unique product launch last year that will not be repeating again this year — its effect needs to be removed from your forecast.
  • Seasonal peaks: If last winter was unseasonably cold with a longer peak, extending the lifetime of your winter jacket product line — there’s no guarantee the same will happen this year.
  • Price changes: If product pricing has changed on your best-selling products since last year, there’s very little chance that they will sell in the same quantities
  • Promotions: If you had a series of wildly successful promotions last year, your baseline sales numbers are going to be severely inflated.
  • Increased or decreased competition: If a new player opened up shop across from you, or a major competitor went bankrupt — basing your sales forecast on last year’s numbers will be completely wrong.
  • Moving holidays: If you forecast sales on a daily, weekly, and monthly basis at location level — you will need to manually adjust days and weeks of sales uplift due to holidays and days off
  • Etc.

And these examples are just scratching the surface, as there are hundreds of factors that have a powerful impact on your bottom-line sales (e.g. product cannibalization, inventory constraints, assortment diversity and depth, vendor constraints, etc.).

So, simply put, a sales-based forecast requires constant “massaging of data”, human interventions, generalization, estimation, and customization — defeating the purpose of what a forecast aims to achieve (an accurate prediction you can base your strategy on).

Another major con of using a sales forecast is that it forces you to repeat the same mistakes over and over again.

For example, if you sold less merch last year because you ran out of stock, a sales forecast will suggest the same inventory level (or slightly more with assumed nominal growth) for you in the future, repeating last year’s costly mistake again and causing you to miss out on sales.

Get the free case study

Discover how a top North American fashion retailer accurately forecasts 250,000 style / color combinations.

Demand Forecast Definition

Demand forecasting is the process of predicting consumer demand for products. Simply put, demand is the willingness of customers to buy a product at a specific price point. In other words, if you know how much demand a product will have at a specific price point, you will know exactly how many units to stock to maximize sales and minimize cost.

While this would have sounded like magic a few decades ago, retail demand forecasting is already being used by the most sophisticated retailers. Advances in mathematics and artificial intelligence make it possible to calculate the effect of influencing factors (e.g. price, seasonality, cannibalization, etc.) and generate prescriptive forecasts that not only accurately predict your future sales, but will also tell you how to maximize them.

Under the hood, this is a fairly complex process. In order to remove the effect of factors like seasonality, inventory availability, and price changes, retailers require the ability to identify every major influencing factor, accurately weigh the impact of each, and map out how they influence one another.

For the purpose of running a business, building a plan, and buying effectively however, a retail-specific AI and predictive analytics. make demand forecasting a cinch to use for retailers.

Demand Forecast Pros and Cons

The Pros of using a Demand-based forecast:

The benefit of using this approach is a highly accurate, dynamic, and reliable forecast. Accurate demand forecasts allow retailers to confidently execute merchandise assortment plans, bring the right quantity of inventory to the right location to match promotional uplifts, and avoid unnecessary markdowns.

This also means if you sold less merch last year because you ran out of stock, the system will identify those lost sales and account for this when suggesting inventory levels for you in the future — allowing you to avoid past mistakes and increase your sales.

The proof is in the pudding. According to a 2019 McKinsey report, retailers that use advanced analytics (like demand forecasting) are outperforming their competition by over 68% — with the number growing exponentially.

The Cons of using a Demand-based forecast:

Most retail ERPs and business intelligence don’t offer true demand-based forecasts powered by advanced retail AI. This means retailers will need to invest in new technology. On the bright side, this investment will pay off in spades, and companies like Retalon make it easier than ever to implement Retail AI into any platform.

The digital age requires a more focused, agile, and proactive approach. Not being able to distill data into specific, tangible, and actionable recommendations creates a ripple effect though the entire business that dramatically inflates costs and decreases gross margins.

It’s a must have for any merchant that wants to experience success with their retail business in the digital age.

Sales vs. Demand – What Are You Planning With?

As Gartner’s VP of Research, Robert Hetu, recently put it, “Retailers that cannot successfully leverage prescriptive product recommendations will find themselves outside of the path to purchase. Not only will they lose existing customers, they won’t be able to attract new customers due to their lack of relevancy and ability to influence.”

The good news is that if you’ve managed to read up to this point, and so you’re becoming more aware about the power of advanced analytics for demand forecasting in the digital era. But don’t wait too long.

One of the biggest mistakes retailers make is overthinking their approach with new technology. Don’t hesitate to pilot new ideas. Run a pilot or proof-of-concept on your own data to see the difference in forecast accuracy and the impact on your business. You’ll be pleasantly surprised and it will be hard to argue the results. 

How do top retailers leverage forecasting?

Uncover how forecasting helped a major North American leader increased sales while decreasing inventory.