Sales forecasting is the process of using past performance data to inform future performance. It’s become an increasingly common element of the site selection process and is often the focus of real estate modeling projects.
In this blog, we’ll discuss the benefits of sales forecasting models, whether sales forecasting models are right for you, and also the different approaches and challenges that are common with sales forecasting.
Why Sales Forecasting Models Matter
Reliable sales forecasting models are a valuable tool in the site selection process. Why? They provide businesses with a clear understanding of a location’s potential performance, guiding strategic decisions and reducing risks.
Understanding a site's revenue potential before committing to the deal informs breakeven calculations and ultimately the final investment decision. Additionally, many sales forecasting models estimate the expected impact of new sites on existing ones, which companies use to minimize cannibalization risks. Ultimately, sales forecasting models help brands to maximize their ROI.
Common Approaches to Sales Forecasting in Real Estate
Sales forecasting involves using past performance to inform future performance. In real estate, there are two primary approaches for forecasting sales. Let’s dive into each sales forecasting approach, what it looks at, and the scenario it’s used in.
Time-Series Forecasting
In a time-series forecasting approach, the analyst uses the existing sales history of a location to extrapolate likely sales performance at that location in future quarters. The goal is to identify trends in store performance and to set goals. This approach is commonly used for existing locations but cannot easily be used for new locations since it requires prior sales history.
Forecasting Sales Without Prior Data
Forecasting sales for a location with no prior sales history is more challenging. The analyst first compares the performance of existing locations to measurable characteristics of the trade area to quantify relationships between variables and performance. Those quantifiable relationships are integrated into a sales forecasting model that can be applied on a go-forward basis to new sites when there is no sales history to leverage. The model can also be applied to existing locations to forecast what sales “should” be given the underlying factors in the trade area/site.
How to Choose the Right Model
Choosing the right site selection model requires a tailored approach. No single model fits every business scenario. If your goal is to use the sales forecasting model to inform real estate decisions for future locations, then a forecasting model designed to provide answers without prior sales history is going to be the best fit. On the other hand, if you are using the model simply to forecast expected revenue at existing sites, then a time series model can work.
Deciding between the two forecast model types isn’t the only decision you’ll need to make. If you’re a brand with multiple revenue streams, you may need to consider investing in more than one model. Buxton can create multiple models focusing on different performance metrics. For example, a convenience store chain might need separate models for projecting store sales and gasoline volume. This flexibility ensures that each model aligns with specific business goals, helping brands make more precise and informed decisions.
Common Challenges in Sales Forecasting Models
Sales forecasting models are important for both planning and decision-making, yet they come with inherent challenges – from understanding the maturity of new sites to managing forecasting errors. By understanding these challenges, businesses can interpret the models more accurately and make more informed decisions despite uncertainties.
The Maturity Factor
In any sales forecast, one of the factors that must always be considered is the time that it takes for sales at a new location to stabilize – or reach maturity. Some brands reach sales maturity quickly, while others take much longer. By studying the sales stabilization patterns in prior locations, the analyst can determine the average length of time it takes to reach maturity and whether the sales forecast should be based on anticipated sales in the first year of the new location’s operations or on annual sales at the date when the location has reached maturity.
Sales Forecasting Error
No matter how solid the sales forecasting model is, there is always a margin of error. Some errors are due to the fact that it is impossible to perfectly measure all the factors that influence sales performance. Since forecasting models rely on variables that can be quantified, a certain level of error is inherent. Knowing this, it’s important to avoid overfitting the model to the factors that you are able to measure because those factors ignore what you aren’t able to measure.
How can you make decisions with confidence given the fact that all models will have a certain level of error? Consider this:
- While imperfect, the sales forecast allows you to narrow the list of possible outcomes and reduce risk. If you open just one store based on the model, the performance may or may not be an outlier, but over time the performance of stores opened based on the model should even out. Consider how your sales average over time when evaluating the effectiveness of a sales forecasting model.
- Manage the margin of error by factoring it into your evaluation. Don’t put yourself in a position where the forecast must be perfect in order to break even on the investment. Incorporate some margin for error into your decision making.
The Bottom Line
While no sales forecasting method is perfect, sales forecasting models are helpful tools in the site selection process and can help real estate professionals to reduce risk.
Explore our blog to learn more about common types of site selection models.
Want to Streamline Your Site Selection Process?
Buxton’s site selection analysis provides data-driven insights to help you identify optimal locations faster and with greater confidence. Discover how our tools can transform your site selection strategy—learn more here.
