On The Level: Choosing and Prioritizing Forecasting Techniques

Words and Figures: Jude Nosek

Autumn is upon us! At Keson, we strive to have a consistent rhythm of our annual activities. Like dealing with the different seasons in the Midwest, there are business challenges we need to deal with at different times annually. At this time, we take up a few tasks that are intimately intertwined: 1) setting a forecast and 2) defining a sales and marketing strategy that can help us achieve it, and 3) determining a budget that can cover the cost of executing that strategy and that covers our other costs. The goals-setting process of forecasting will define a large portion of what we are going to spend our time and energy on as a company. 

Over the last two decades, we have built a few processes to help complete these tasks.  In the last few articles, I addressed many of the techniques we use (and subject others to) to determine the importance of one option in relation to another and how to prioritize and tackle those projects. This article is about how we forecast the near future, namely the sales for the next 12 months. 

I found it interesting that the word “forecast” shares a root with the process of rolling dice and weaving a spell and is exactly what we say when we’re predicting fickle weather. When we are scheming to package the future in this way, it’s good to realize that what we are doing comes with limitations. 

Forecasting

We have hundreds of products and customers. We could evaluate each family of products and each industry and type of customer. Some years or times in a year, we drill down all the way to a specific product or specific distributor. Periodically performing these activities can yield insights and opportunities that might go unnoticed. Deep analysis is particular to each company. They can be fruitful, and they can be time-consuming. These types of deep dives are not typically part of our forecasting process, though they can influence it.  

We take a few approaches to forecasting. We use these techniques because they are simple and because they get us to a decision that all of us can get behind. There are some checks and balances we utilize in our approach. These tools can work for any size business and can be done fairly quickly. 

We often do a self-evaluation prior to the forecasting/budgeting. A SWOT (Strength, Weakness, Opportunity, Threat) analysis is a good tool for this. You can find resources online for this technique, and we will address it in an article in the near future. During this process, we also look at the markets we serve and our competitors, and we take the trends of the economy as a whole into consideration. There is a lot that we cannot change and cannot anticipate. Things change, people want more, people want something different; machines, vehicles and tools need repairs and replacement. For us, there’s not a practical way to try to account for it all. So, we look at what we can control. There is a chart at the end that will give you a sample of what three models can look like. 

Not much of a “forecast.” This is the “nothing changes” model. 

We take a snapshot of our business at midnight on the last day of the year. The sales we have completed at that time for the last 12 months are the same sales that we will get for the year ahead.  Essentially: 0% growth. 

Can we run the business that same way? What’s changing? If sales end up being stagnant, what choices will we make to ensure that we can stay in business? If something were to happen, could we handle it? 

We take a snapshot of our business at midnight on the last day of the year AND the percentage change of each product family compared to its baseline (the year prior sales). We then project this percentage forward. 

I call this “Newton’s First Law Forecasting”: sales in motion tend to stay in motion (we hope). The idea is based on Newton’s first law of motion – “every object will remain at rest or in uniform motion unless compelled to change its state by the action of an external force.”

This is the best real-world/on-the-go analysis of what is happening at this moment with our sales. It includes anomalies that we should strive to identify and account/adjust for. We also weigh and pay more attention to the big movers over the small. For example, for us, measuring products like long tapes, wheels and levels are given more attention than marking products like stake flags and mason twine. 

The final model we use is to select areas of interest and scale them up or down based on the knowledge that we have (e.g., a competitive price change, a supply chain shortage, longer deliveries, etc.) or actions we are planning to take (e.g., a new salesperson with history of success, entering a new market segment, introducing new products, etc.). These are the external forces that will act on the sales already in motion. We are going to leverage these to hopefully change our sales. In this method, we identify the family and the action or reason that will have an effect and what that effect might be. 

Here is an example of a “million-dollar” company