- Sales forecasts are known as the most biased forecasts.
- They usually degrade the quality of the statistical forecast, yet they continue to be used because sales have the power to force supply chain organization to use this lowest of quality inputs.
- We offer specific actions for reducing forecasting that is also related to how sales pipelines and forecasts are set.
This salesman is crushing his quota and providing inaccurate positively biased forecasts to supply chain. He, like almost all salespeople, is not measured on his forecast accuracy, only quota attainment. The amount of inventory carried by his company is not his problem. He does get highly irate if his product is not in stock at a 99% service level.
Introduction to Sales Forecasting
Sales forecasts are forecast developed by a company’s sales department. Even at this late date (2014), sales forecasting is often not differentiated from sales quotas.
The Bias of Sales Forecasts
Sales forecasting has the highest bias of any forecast generated in any department within a company. The following quotation is instructive on this topic:
“Political pressures within organizations affect sales force composite forecasts because of a lack of understanding of the difference between sales forecasting and setting sales quotas. A sales forecast should be a realistic prediction of product sales for a certain interval of time given a set of assumptions regarding the environment. A sales quota is a motivational tool that assigns a portion of required revenues to each sales region and salesperson. Sales quotas and sales forecasts developed by the sales function should be developed with these differences in mind.” – Sales Forecasting Management
The Lacking Mathematical Support
Salespeople and brand managers are often not provided with the training to efficiently convert their market intelligence into useful adjustments to the forecast. Simply being a salesperson is not proper training in producing a quality forecast. I am quite amazed that salespeople are not provided with a quantitative analyst to help them in the translation of their market intelligence to numbers.
Forecast Bias Definition
Let us make sure that we are on the same page on the term forecast bias. Forecast bias is when a forecast is consistently inaccurate in one direction, either higher or lower than the actual forecast. A forecast can have zero bias but also may still be highly inaccurate, yet the number of times and the degree to which it is higher than the actual outcome is more or less balanced with the times when the forecast is lower than the actual result. Therefore bias is the degree of the direction of forecast error and not an overall measurement of forecast error.
Forecast bias is one of the most important topics of forecasting, but also one of the least discussed, and this is nowhere more true than from the perspective of purposeful bias.
The Layers of Sales Forecasting at Companies
Sales forecasts are performed both by salespeople, as well by people that manage other salespeople. In other articles on sales forecast bias, there is often a line of thinking presented that while the salespeople may submit a biased sales forecast, the sales Directors and VP’s are motivated to reduce this forecast bias. In fact, this happens to be the overall reasoning for the application created by Right90. This is one of the few applications that allows the buyer to address how to identify and adjust forecast bias and it is entirely focused on sales forecasting.
A Correct Assumption?
Years ago I was enthusiastic about the Right90 application, until I began to become more experienced in sales forecasting, and realized that the use of this application contains an assumption. And this assumption is that Directors and VPs of sales want to decrease forecast bias.
Is this assumption actually in many cases not true. The reason is easy to understand, the managers of salespeople are as subject to having their careers negatively impacted by not presenting an appropriately rosy picture to their higher-ups, in the same way, that the salespeople are.
If a salesperson or sales director generates a realistic forecast, someone who can provide a forecast that is more to the liking of the higher ups may replace them.
And this is the problem with the assumption that those that manage salespeople necessarily have an incentive to neutralize the bias that is contained in their sales forecast. To understand why, let us first layout the naïve hypothesis, one that one might make if one had not seen how sales processes tend to work.
The Naive Hypothesis
Under this hypothesis, sales management would have to have an incentive to reduce sales forecast bias, because if they do not, then the salespeople more often than not, will end up not meeting their quota for the quarter which sales management would then be held accountable.
The reality of how sales forecasting works is quite a bit different from this, and while the naive hypothesis “seems to make sense,” in fact, it does not account for the reality of sales teams. And the reason the naive hypothesis is untrue has to do with some of the following factors:
- Turnover: A company’s sales force, as well as the sales management, is filled with people that are in various stages of their career within the enterprise. The turnover at most businesses in sales is high when compared to other professions. Therefore it is very common for a Director of VP of sales to be working with a new Salesforce (if the Director of VP of Sales is new to the company) or with new individuals within the sales force (when new salespeople join the sales force). Some of the salespeople will not be able to make their quota. Therefore they will have the incentive to continue to forecast that sales that have almost no probability to close are genuine opportunities to keep their pipeline looking healthy.
- The Lifecycle of the Sales Person: The sales management may know that some salespeople have a small likelihood of closing much of their pipeline, but if they remove the salesperson, they still have to find another one, and the new salesperson will take the time to bring up. In fact, it is estimated that it takes 7 to 10 months for a new salesperson to become fully productive (on average). Therefore, sales management must decide if they are willing to replace an underperforming salesperson – with a new one, with the knowledge that they probably will more likely than not sell little for at least the first two-quarters.
- New Products or New Services: Any company will have sales in different product lines. As the company often has significant investments in the new product lines and very much wants the new product lines to be successful, as they see this as their growth. In fact, sales management will often have promoted the new products to their management. What all of this means that is an incentive to be more accepting of having the product line underperform for a while.
- The Lack of Forecast Accuracy Checking: Companies that sell high volume items quite often do track sales forecast accuracy, but few hold their salespeople accountable. Enterprises that sell low volume items often do not track forecast accuracy. Small volume item forecasting is extremely challenging generally, which is explained in this article. If a salesperson forecasts one customer to sell at 50% and another at 50%, and one sells, and other doesn’t have the forecast accuracy would still be small.
Sales Management Setting the Sales Forecast
According to salespeople with which I have discussed this subject, in many cases, sales management sets or just overrides the forecast values in the CRM system for their sales team. Sales management often gives the impression that the salespeople have added the sales forecasts. This brings up obvious problems on checking the forecast accuracy of the forecasts that are produced by the actual salespeople. This is an enormous faux pas that would not be accepted in other types of forecasting, yet in sales forecasting it is routine.
Often CBF improvement is confused or commingled with bias reduction. This is actually a very significant error which undermines the understanding of CBF. Bias can be and most often is present in both the consensus side as well as the statistical forecasting side. The objective is to remove bias from both sides. However, very few companies do. Why?
Below are listed both business and technology reasons why bias reduction does not tend to be a focus of companies.
Who is Focused on Bias Reduction?
I can say that most forecasting applications I have come across do not have forecast bias adjustment workflow built into them. This workflow is important because the software must provide an easy and systematic way of reducing bias.
Up to this point, the most advanced in this regard that I had seen were the consensus based vendors Inkling Markets and Consensus Point.
The Sales Pipeline
The sales pipeline is the primary sales forecast. The sales pipeline is not set by the salesperson but by sales management, and when compared to other types of forecasting, this is incredibly strange.
The reason it is so weird is that the sales pipeline (or the “initial forecast”) is a goal and part of the objectives that the company would like to meet. The sales pipeline is a forecast that the sales management does not need to meet, it is the sales pipeline that someone else needs to meet. Of course, broadly speaking, sales management must attain some combined sales pipeline for its sale people. This won’t normally be met because only a little over half of salespeople meet their sales pipeline. So some of the salespeople will not meet their sales pipeline without necessarily affecting sales management.
Let consider a VP of sales which works with the CEO to set sales pipelines for the year. Both the CEO and VP of sales want the sales pipelines to set high. And because they don’t have to meet the quotas, there is little to moderate them from fixing the quota at a high level. The salespeople have to meet the quota, as they keep their job if they meet the quota. Therefore, there is an apparent conflict between sales management and salespeople regarding the quota. A forecast should not be a goal. For instance, if someone is asked how much money they would like to make in a year you will get one number, which is different than the number of what they think they will make in a year.
Quota setting blurs the line between what is desired and what is a forecast, and that is the heart of the problem, and why there is a significant bias in the large majority quotas unless they are set through some highly structured process, which actually separates the desire from the forecast.
The Effect of the Sales Pipeline on Forecast Accuracy
Sales quotas are typically set too high, which is demonstrated both by the percentage of salespeople that meet quotas combined with an analysis as to how sales quotas are set.
While sales management will often feel satisfied with their quotas because they give their salespeople “stretch goals.”
Not all stretch goals are motivational. If goals are set to the point where they are unattainable, the goal will often become de-motivational. Experience climbers set attainable goals for themselves regarding the progress that they will make per day.
This also has an unavoidable negative impact on forecast accuracy as each sales person (except for those that are the top sales people and are having good quarters) as it promotes the sales person not to qualify out of opportunities and to keep customers that have a low probability of closing in the forecast. And this not only affects the average or below average salespeople, because when top salespeople exceed their quota, but they also have a tendency to have their quota raised the following year.
Much of the game playing that occurs in the communication between sales and sales management, much of the bad data that exists in CRM systems can be attributed to a lack of reality regarding the first part of the sales management process, that is the setting of the quota.
A wide body of academic research contradicts the often repeated claim that judgement forecasts are more accurate than or add substantial value over the statistical forecast. However, this information is not percolating up to executive decision makers, who appear to have little interest in measuring the accuracy of marketing or sale forecasting inputs into the consensus forecast.
Brightwork Forecast Explorer for Monetized Error Calculation
Improving Your Forecast Error Management
How Functional is the forecast error measurement in your company? Does it help you focus on what products to improve the forecast? What if the forecast accuracy can be improved, by the product is an inexpensive item? We take a new approach in forecast error management. The Brightwork Explorer calculates no MAPE, but instead a monetized forecast error improvement from one forecast to another. We calculate that value for every product location combination and they can be any two forecasts you feed the system:
- The first forecast may be the constant or the naive forecast.
- The first forecast can be statistical forecast and the second the statistical + judgment forecast.
It’s up to you.
The Brightwork Forecast Explorer is free to use in the beginning. See by clicking the image below:
Mentzer, John T. Bienstock, Carol C. Sales Forecasting Management. Sage Publications. 1998.
Sales Forecasting Book
The Problems with Combining Forecasts
In most companies, the statistical and sales forecast are poorly integrated, and in fact, most companies do not know how to combine them. Strange questions are often asked such as “does the final forecast match the sales forecast?” without appropriate consideration to the accuracy of each input.
Effectively combining statistical and sales forecasting requires determining which input to the forecast have the most “right” to be represented – which comes down to those that best improve forecast accuracy.
Is Everyone Focused on Forecast Accuracy?
Statistical forecasts and sales forecasts come from different parts of the company, parts that have very different incentives. Forecast accuracy is not always on the top of the agenda for all parties involved in forecasting.
By reading this book you will:
- See the common misunderstandings that undermine being able to combine these different forecast types.
- Learn how to effectively measure the accuracy of the various inputs to the forecast.
- Learn how the concept of Forecast Value Add plays into the method of combining the two forecast types.
- Learn how to effectively run competitions between the best-fit statistical forecast, homegrown statistical models, the sales forecast, the consensus forecast, and how to find the winning approach per forecasted item.
- Learn how CRM supports (or does not support) the sales forecasting process.
- Learn the importance of the quality of statistical forecast in improving the creation and use of the sales forecast.
- Gain an understanding of both the business and the software perspective on how to combine statistical and sales forecasting.
- Chapter 1: Introduction
- Chapter 2 Where Demand Planning Fits within the Supply Chain Planning Footprint
- Chapter 3: The Common Problems with Statistical Forecasting
- Chapter 4: Introduction to Best Fit Forecasting
- Chapter 5: Comparing Best Fit to Home Grown Statistical Forecasting Methods
- Chapter 6: Sales Forecasting
- Chapter 7: Sales Forecasting and CRM
- Chapter 8: Conclusion