As mentioned in my last post, Peter Michie of PERFORMAX will be discussing forecasting.  Peter has many years experience in sales and sales management and utilized that experience to build a successful consulting company named PERFORMAX. 

Background

In my experience over years of working for sales organization either as an employee or as an outside consultant I’ve found that one of the most potentially valuable company processes, Sales Forecasting, is often

-        Used as a “whipping post” for sales people by their managers

-        Totally inaccurate in the resulting forecasts.

Although large amounts of effort & time are spend on collecting, reviewing and aggregating Sales Forecast, often management either ignores the inputs from Sales, or modifies them based some secret algorithms that he or she have developed.

So the purpose of this paper is to review why this is the case and an outline of a better Sales Forecasting Process.  Part I of this paper will look at forecasting today and Part II I will introduce a better forecasting process.

Sales Forecasting Overview  – Today

Just to be clear, I am talking about Sales Forecasting based on inputs from individual sales reps concerning the set of sales opportunities with which they are engaged; not Market Forecasts based on estimates of possible market share & penetration etc.

Most the sales force uses some kind of forecasting system—either a homegrown system or a Sales Forecasting System or a Customer Relationship Management System (CRM).  I will refer to all these collectively as CRMs.

Often the CRM is a large $$$ ticket item purchased by senior management in an attempt to manage sales from the executive suite. The sales people are expected to work with the system, to populate the system with all of the customers & prospects that they are working on (their pipeline or funnel), to submit an estimate of when the sale will close (their sales forecast) on some regular basis for upcoming 30, 60 and 90 day periods (or beyond); and to record all of their activities (by sales step) into the system.

Table1 below shows a simple sales forecast for one sales person.

Table 1: Simple Sales Forecast:

Most CRMs provide much more info than the above, but even this simple format allows the sales person and their sales manager to discuss specific opportunities, develop a strategy on how to win the accounts, and provides some informational outlook of future $ revenue expected from each sales opportunity.

Of course there need to be different kinds of layouts to provide similar information for recurring revenue stream companies, and for companies whose sales forces sell many more transactions than the few shown here.

The individual sales forecasts are then aggregated up to the sales manager level, regional director level, and so on until up to the Corporate Sales Forecast for use by senior management.

Reasons for a Sales Forecasting Process – Today
The Sales Forecast is supposed to predict the revenue & product sales that will be coming in during a specific period (typically 30, 60, 90 days into the future), so that senior management can then feel confident that they are “on top of the business” and that they can move forward with plans for hiring, expansion, training, acquisition etc. based on this information. 

In manufacturing companies, the other critical dimension besides the $ Forecast is the Product Forecast as most products have manufacturing lead times, so a future sales forecast is critical in terms of making sure the right products are available at the right time.

Obviously, in all cases this data is vital to the successful management of the business. 

Unfortunately the information is often not used: in over 65% of our client companies, the Sales Forecast, as rolled up to Sales Management, was mostly ignored at the executive level for several reasons:

1. It was very inaccurate or unreliable.
2. It just simply was not what they wanted to see, as in order for public companies to hold their share value, they need to show that their forecast is either close to, or exceeds their business plan. 

As per the Introduction, instead of being a real management tool with consequences and accountability, the sales forecast from the Sales Department often becomes just another exercise in futility.

Inevitably someone in finance or product management, or in the senior executive’s office, or an analyst with a spreadsheet produces  “The Forecast”, based on historical trending or imagination or some other unreliable source, and that is used by the Executive Team as the basis for their planning.

Why are Sales Forecasts inaccurate?

There are 2 major reasons that inhibit forecast accuracy:

1. Uncontrollable, External reasons: Forecasting tries to predict the future – and most of us are not very good at that because of uncertain economic times, customers changing their minds, of being taken over, competing priorities, typical sales obstacles, etc. etc. 

1. Internal reasons:  Corporate culture, cumbersome systems or ones with flawed logic, confusion about what to forecast, fear about not forecasting enough by poorly trained sales people & sales managers, shifting emphasis, confusion and unclear goals and direction, and resistance because no one has addressed WIIFM—What’s In It For Me?

In many cases the tool has become proof that something has been done to address the sales issue.  “We spent a lot of money on a system, so we must have solved the problem!”  But what value is actually being generated by a tool with such questionable results?

Within most corporate sales departments, many sales managers create their own personal spreadsheets, often with different assumptions, to come up with a forecast that they are comfortable sending up the chain. Even though this process has been integrated into some of the CRMs, most Sales VPs will not place significant bets on the accuracy of the automated forecast. In at least one company, the promotion to sales manager is referred to as “death by spreadsheet”. 

Other Sales Forecasting Issues

If some of the above problems were not enough, within several CRM Forecasting modules there are at least 2 inherent problems that add to the inaccuracy of the overall Forecast:

1. Using Weighted Forecast $ where each sales opportunity’s Forecasted $ Amount is multiplied by the percentage probability of the sale closing
2. Using fixed probability percentages attached to different stages of a sales process.

1. Why (Forecasted Amounts) X (% close probability) does not work!

As you will see, the next step of forecasting process (with a Forecast like the one above) causes the most common contributing problem for the accuracy of a sales forecast, as it uses totally false logic….

In order to assist Sales Managers, and ultimately Senior Management, in the creation of a viable company forecast, all of these items must be translated into a realistic $ forecast.  A common technique is to simply multiply the individual opportunity $ by the % probability shown in the table to come up with a weighted forecast $. 

Using this technique for the data shown in Table 2 (below), a weighted total line would be:

Table 2: Weighted Forecast

This one line result is then consolidated at the Sales Manager levels; then up the management chain until at the executive level the 30/60/90 day projections are available for supporting the company objectives.

So what’s wrong with this???                                                                …EVERYTHING !

First, even my bad degree in mathematics taught me that the result from the multiplication of a $ dollar amount and a % probability yields a totally meaningless result when it is applied to a sales forecast! The results on a line by line basis are actually as meaningful as multiplying the average temperature of the Atlantic Ocean by your age; or your car speed by the width of the road you drive on.

In the sales context shown in Table 1, probability refers to the likelihood of a sale taking place; not the percentage of the revenue if the sale happened. So if you have 3 pending sales each forecast at a 33% probability, and you really think only one sale will occur, then how much are you actually forecasting?  Of course the difficult thing is to know which ones will close!

To illustrate how misleading this technique can be, the example above nearly saw me fired as a Sales Manager, and which became the primary driving force behind my quest to seek a better way to forecast consistently with greater accuracy.   

So let’s look closely at the 30 day time frame, assuming the company uses this data to forecast $1,617,000 of business closing from this one sales person. If you look at the possible scenarios depending on which of these 3 deals actually close, and since none of the opportunities are forecast at 100%, only if the 3^rd, large transaction closes, does the forecast come anywhere close to the Weighted Probability Forecast. 

Also, the consequences of a shortfall in forecast of this order would probably mean drastic measures for the company: job cuts, bonus cuts, delayed purchases, delayed investment in R&D etc.

 This sample is built around one salesperson’s forecast.  Imagine if you have a large sales force with similar shortfalls… and many of you do!  How does a shortfall of $1,500,000 times 10 or even 100 salespeople look?

 Perhaps this gives a very good reason why Sales Forecasts are so often ignored at the company level by the CFO who creates a Finance Forecast with no input from the Sales Department.

 In Part II of this two part article I will introduce a better forecasting methodology.

| | | | | |

No related posts.

Related posts brought to you by Yet Another Related Posts Plugin.

Tags: Degree of Qualification, Forecast, forecasting, qualified opportunity, qualified prospect, sales forecast, sales forecasting

Categories: Uncategorized