COMMERCIAL IMPLICATIONS IN THE SALES FORECAST PROCESS: THE QUESTION GOAL VERSUS FORECAST AND THE IMPORTANCE OF SALES FORCE PARTICIPATION

Sales forecasting is gaining increasing importance in companies. In fact, this importance is disguised by complaints such as “one of our biggest problems is sales forecasting”. In fact, these statements occur not because everyone in the companies clearly visualizes the importance that the sales forecast has, but because the effects of a bad forecast can be felt throughout the entire organization.

Among the costs incurred with a low-quality sales forecast can be mentioned:

• In companies whose production schedule needs to be planned in advance, a forecast with low accuracy can result in planning that is very far from the actual need. This situation leads to two types of cost, which are not necessarily mutually exclusive: the cost of stocking a product manufactured in quantities greater than necessary and the cost of lost sales of products manufactured in quantities smaller than those actually demanded;
• Even in cases where the lack of forecast precision is detected in time to change the production schedule, this is done causing production costs higher than desired;
• This disturbance that affects production ends up also affecting the purchasing process. In times of globalization, many companies operate with foreign suppliers, whose supply lead-times exceed one month. In this way, the lack of precision in the sales forecast is reflected in larger safety stocks for imported products, unnecessary purchases or emergency purchases through the use of some premium transport. This extra supply cost may also apply to distribution and retail companies.
• Finally, for companies that have more than one stocking location – their own distribution network – poor forecasting quality can result in the misallocation of available products along the network, causing unnecessary logistical costs, such as transfer between facilities.

In view of all these disturbances, and associated costs, sophisticated forecasting packages were developed, capable of making statistical projections of demand, by item and by location. These packages use well-known quantitative techniques, and with the possibility of performing calculations quickly and simply, with little consumption of people's time during the entire forecasting process.

Despite appearing to be the complete solution to sales forecasting problems, statistical packages only make up the solution, which also involves restructuring the process as a whole, having as one of the most important points the active participation of the sales area.

The objective of this article is to present an approach for structuring a sales forecasting process that combines statistical forecasting methods with an active participation of the commercial area, with the objective of clearly demarcating the difference between sales forecasting and commercial goals, seeking at the same time make them compatible.
THE SALES FORECAST PROCESS

First, some relevant issues need to be discussed in terms of the sales forecasting process. Why not just do it based on quantitative methods present in the forecasting packages? What is the real need for the sales force to participate in this process? In this sense, it is important to emphasize that statistical methods are fundamental in the demand forecasting process. The purpose of this article is in no way to demean them, but to show that they should not be used alone, and their results should be criticized and validated by professionals and areas involved in the process.

With regard to quantitative methods, it is important to emphasize that they are based on the analysis of historical data and assume that the past behavior of demand tends to remain in the future, as long as the variables that influence demand do not change substantially. In this way, given a series of assumptions, the aim is to identify patterns in past sales and it is considered that these patterns will be repeated in the future, being then used in the sales projection. Thus, whenever changes occur in the environment in which sales take place, these will affect the accuracy of the forecast.

These changes, however, are often known in advance by the sales force or any other customer contact area. The contribution of the sales force is, therefore, to add to the “cold” forecast, calculated through historical series of demand, the market information available, and which may change the behavior of sales. This information may be related to the acquisition of products by a new customer, some promotion that will be carried out, stoppage for maintenance of equipment in production, among other factors.

The sales forecasting process can then be structured according to Figure 1. Initially, a forecast is calculated using statistical techniques, under the responsibility of a centralized team. This forecast must consider the historical behavior of demand and all the variables considered relevant in its behavior. Despite being obtained centrally, forecasts must be calculated with the necessary disaggregation levels, such as by customer, SKU, or sales region.

This “cold” number is then made available to the sales force, in a disaggregated form, so that it can be criticized and validated. This criticism is made based on the market's perception or on some concrete information that has not been used in the statistical calculation, and the professional is free to change the number as he sees fit. Regardless of whether or not any changes occur, the sales force becomes co-responsible for the forecast and, consequently, for its accuracy.

Once all forecasts are validated, they return to the place where they were initially generated to be aggregated again and arrive at the final consensus number for the company. This number is then sent to all areas that will use it, such as production and purchasing.

This final aggregation is due to the fact that it is not only necessary to know how much is expected to be sold of a given product at each point of sale, information necessary for distribution planning, but also the total quantity to be sold, information necessary for production or items reposition.

Although the figure only presents a degree of validation, this can occur at different levels of the organization, salespeople, regional managers, national sales manager. A regional sales manager can, for example, validate the total forecast for the area under his responsibility after the intervention of his salespeople. It should be borne in mind, however, that the greater the number of validation levels, the more costly and time-consuming the process.

CRITICAL FACTORS FOR THE SUCCESS OF THE SALES FORECAST PROCESS

In order for this process to achieve the expected results, two factors related to the company's commercial area must be considered, which will now be addressed: awareness on the part of the entire company with regard to the independence of the forecast in relation to the company's commercial goals (sales targets), and the participation of the sales force in the forecasting process.
THE FORECAST VERSUS GOAL ISSUE

Linking the sales forecast to the company's business goals is one of the most frequent problems encountered in the forecasting process. Despite having a very strong relationship with each other, there are significant differences that must be taken into account.

When a company's commercial objectives are not consistent with the behavior of its demand, there is a tendency to use these objectives as a forecast. This behavior is justified by the fact that companies need to pursue their growth, and one of the ways this can be achieved is through the establishment of aggressive commercial goals. By calculating a forecast based only on the past and current behavior of sales, there is a risk that they will stagnate at the level they are at the moment. The flaw in this reasoning is that the sales forecast is being related to the objectives without the intermediation of a structured sales plan.

Sales forecasting can be defined as a projection into the future of expected demand, given a set of environmental conditions. The sales plan, which expresses the commercial objectives, in turn can be seen as a set of managerial actions to be taken to achieve or exceed the sales forecast. In this way, while the sales forecast seeks the most likely scenario about future demand, the sales plan aims to reach, and in some cases exceed, the forecasted demand levels; that is, providing sales targets for marketing and the motivation for achieving and exceeding these targets.

Based on these two definitions, it is then possible to structure a process that expresses the existing relationship between the projection and the sales plan. Through given economic and environmental conditions, as well as behavior patterns, an initial projection of future demand is made. From this initial projection, the sales plan can be developed.

In case the initially structured sales plan does not reach the financial objectives of the company, a return to the sales forecast should be performed and examine what additional marketing and/or sales efforts can be taken in order to increase the projection to the levels necessary to achieve business objectives. This interactive process, figure 2, must be carried out until it is possible to combine the real behavior of the market with the company's objectives.

A point to highlight in this process is that the sales forecast can be raised to the level that reaches the previously set objectives, however this must be done through a plan that contains all the actions necessary for this. Among these actions, promotions, launching of new products, advertising and entry into new markets can be mentioned.

In order to ensure that the actions defined are actually able to generate the planned sales levels, it is necessary to know the magnitude of the impacts that each one can cause on the company's demand. This knowledge is obtained by storing information on actions taken in the past and the respective sales levels for the period. For example, sizing a promotion should be done by analyzing the impact of similar promotions already carried out.

However, it may happen that, even if a series of actions are planned, the objectives are not achieved. How then to ensure that in this case the forecast really expresses the market trend and not the company's desire?

This assurance must be provided by monitoring the accuracy of sales forecasts and their respective impacts throughout the company. Accuracy monitoring is very important, but by itself it is not capable of guaranteeing a commitment to high performance. Cause and effect relationships must be created that quantify the magnitude of the costs mentioned at the beginning of the article.

As already seen in the article published in this journal in February 2001 (see bibliographical references), it is possible to parameterize the stock policy, more particularly the safety stock, depending on the precision of the forecast, among other factors. With this parameterization, it is easy to quantify the cost of forecast accuracy in terms of inventory cost. As with the safety stock, the impacts on the supply activity should also be quantified - such as, for example, an increase in the acquisition cost due to the need for emergency orders - and on production - costs associated with production rescheduling and/or production batches not optimized due to urgent needs.

With these costs, it is easier for the entire company to charge for the quality of the forecast, regardless of who is responsible for its execution.
SALES FORCE PARTICIPATION

As seen previously, the participation of the sales force in the forecasting process is of fundamental importance, through the criticism of the statistical forecast based on the market information they have. In order to maximize the performance of this participation, four actions must be emphasized:

• Incorporate forecasting into the sales team's work;
• Minimize the influence of sales objectives;
• Keep the process simple;
• Ensure focus.

Incorporate forecasting into your sales team's work

It is common for the sales force to perceive that forecasting is not part of their job. The time spent on this activity is seen as time taken away from its “real function”, that of effectively selling products or services.

It is essential that sales management communicate to its people that sales forecasting is also part of its activities. A salesperson's role can be seen as containing three axes of activities. The first axis is effective sales, that is, generating revenue and profit for the company. The second is building and maintaining customer relationships – an investment activity that will ultimately result in increased customer purchases. The third axis is precisely to provide information to the company about customers, competitors and the market in general, in which the participation of the sales force in the forecasting process is inserted.

There are a few ways in which you can convince the sales force that the forecasting process is part of their job. Initially, you must ensure that this type of activity is contained in your job description. However, this formalization is of no use if there is no direct charge for this activity. In this way, just as it is necessary to monitor the accuracy of the forecast by product, as has already been seen, there must also be the monitoring of the accuracy of the forecast for each seller. In the event that salespeople have variable compensation, the sales forecast accuracy indicator must be included in the compensation calculation. In parallel, feedback should be provided regarding the performance that the salesperson has been achieving in this regard, as a way to encourage improvement in terms of forecast accuracy.

Finally, communication must also be done through training. This training can have two aspects. The first, and which should not be overlooked, concerns raising awareness about the importance of sales forecasting, highlighting its impacts on inventory management, production scheduling, supplies and capacity planning. The costs resulting from the lack of quality of the forecast must be explained. The second aspect has to do with training salespeople on techniques through which salespeople can improve their forecasts. This second aspect should only be worked on when the sales force is expected to do more than just criticize an already calculated statistical forecast.
Minimize influence of sales objectives

This aspect is nothing more than the unfolding of the confusion between the sales target and the forecast discussed earlier, with regard to the sales force. In this type of environment, two types of situations can occur: underestimation of the forecast so that they can be easily overcome, giving the impression of a great sales performance, or else an elevation of the forecast towards the target, giving the impression that this will be achieved, even if this is not realistic.

It is worth mentioning that both situations are harmful. The first will likely create conditions for product shortages, since material planning was carried out based on a lower number than sales, while the second will incur unnecessary inventory costs.

Once again, the answer to this question lies in monitoring and rewarding the sales force based on the accuracy of its forecasts. This monitoring must be dissociated from the achievement of sales targets, in this way, the salesperson can be rewarded even if the target is not achieved, provided that he has already indicated that this quantity would not be reached.
Keep the process simple

This issue is related to occupying the least amount of time possible for the salesperson in this activity, because despite being part of their duties, it is not a priority. For that, it is enough that the process is structured as shown above, with the participation of the sales force restricted to the validation of a forecast that has already been made.
ensure focus

This aspect is also related to the workload passed on to the sales force. Only the participation in the forecast of the most significant customers/products of the business should be charged. If the Pareto rule (80/20) is applied, the participation of salespeople must be restricted to the 20% of customers/products that correspond to 80% of sales volume, that is, customers/products for which small differences in accuracy in forecasting can generate large differences between the total volume forecast and the volume actually sold. Forecasting for the remaining 80% of customers/products can only be done by statistical methods as the impacts of your forecasts are not as significant on the total volume. In addition, a smaller number of forecasts to be validated becomes less costly for the company and less resistance on the part of the sales force.
CONCLUSION

This article sought to demonstrate that the use of statistical packages alone is not enough to obtain accurate forecasts, constituting only the easiest step in the search for quality forecasts.

As important as the use of statistical techniques is the structuring of a process that deals with existing organizational issues, such as awareness of the importance of sales forecasting, its independence in relation to the company's goals and the need for the sales force to participate in the process. process.

In order to guarantee its effectiveness, the forecasting process must be monitored not only with regard to its performance as a whole, but also in relation to the participation of the professionals involved in it.

If these points are not considered, there is a risk that the calculated statistical forecasts will not be effectively used, causing not only that all the effort to arrive at them was in vain, but also that the problems generated by the bad forecast continue to exist and entail unnecessary costs.
BIBLIOGRAPHY

• MENTZER, John T., BIENSTOCK, Carol C.; Sales Forecasting Management; SAGE Publicatins, 1998
• SAGGIORO, Eduardo, LACERDA, Leonardo, AROZO, Rodrigo; “Managing Uncertainties in Logistics Planning: The Role of Safety Stock”; Tecnologística Magazine, February 2001