How To Forecast Sales

The conventional wisdom - "Sales managers can't forecast accurately because there are too many uncertainties involved."

The reality - "Sales forecasting can indeed be turned into an accurate, reliable process."

Forecasting is becoming all the more difficult because customer loyalty is on the wane. Global competition has increased such that companies are less sure of where future sales will come from. Moreover, distribution channels have become more complex and the lifespan of products has decreased, all resulting in greater uncertainty.

Indeed, research by Sales Benchmark Index has found that roughly two-thirds of all sales forecasts have a margin of error that exceeds 25 percent. Amazingly, more than 10 percent of forecasts have a margin of error of greater than 75 percent!

Many companies continually suffer from sales forecasts that are inaccurate and unreliable. When projected numbers are unrealistically optimistic, manufacturing divisions ramp up operations for products that end up sitting in warehouses collecting dust. Or, conversely, the demand for a hot item shoots through the roof but the company is caught off-guard, thus missing a crucial window in the market.

And it's not just big mistakes that hurt the bottom line. Sometimes even a small increase in the accuracy of your forecasts can lead to substantial savings, because your distribution chain will be returning fewer products, thus decreasing your shipping, handling and storage fees.

Let me put it this way: I have never heard a CEO or senior manager complain that the forecasts from their sales group were too accurate, but I have heard countless executives grouse that they simply couldn't rely on their company's sales projections. An inability to forecast sales usually means you've lost touch with your customers - a deficiency that can lead to disaster when the market makes a turn in one direction and your firm is still headed down a different path.

The result: You end up developing and marketing products nobody wants.

It's a cop-out for sales managers to claim that sales forecasting is inherently impossible. The simple truth is that companies can indeed reliably forecast their sales, and all the leading organisations do it because they absolutely need that crucial information. Otherwise, a business can't be run efficiently. How, for example, can the manufacturing department plan its resource allocation without knowing the volume of shipping orders for the upcoming quarters?

The trick to accurate sales forecasting is having the right system in place.

Understanding the sales funnel

Each location of the funnel (bottom, middle or above) has a quantitative metric for the likelihood of the deal closing in a given amount of time. That period can be based on a typical sales cycle. For example, a typical sales cycle for your product might be eight months.

• Bottom - At the bottom are deals that you've almost closed. They might generally have a 70 percent probability of closing, within half the sales cycle.
  
  
• Middle - In the middle of the funnel are other prospects that are in the works. They might have a 40 percent chance of closing within the sales cycle.
  
  
• Above - Above the funnel are numerous leads that need further investigation. Above the funnel prospects might have just a 10 percent chance of becoming a finalised deal within the typical sales cycle.
  
  

To obtain an accurate projection of your sales:

1. Categorise every deal into the right location (bottom, middle or above)
   
   
2. Estimate of the size of each potential order
   
   
3. Add-up the value of the opportunities for each location of the funnel 
   
   
4. Apply the appropriate probability and time period
   
   
5. The total sum of those numbers is your sales forecast
   
   

Some may be skeptical about how such a simple concept could be effective in practice, but I have seen numerous companies dramatically improve the accuracy of their sales forecasts by implementing it.

Managing the sales funnel

You must first classify your customer opportunities accurately - easier said than done. Many sales professionals fool themselves into thinking a deal is closer to being closed than it really is. You need to get them to be realistic by requiring good metrics, both qualitative and quantitative.

Everyone has to agree to the criteria, and each person has to abide by them. You should consider having a standard form for salespeople to fill out for each solid lead, especially for potentially large deals. That form would include questions that help determine the location of that opportunity with respect to the sales funnel.

Consider an acronym like "DUNCE"

• D - The customer has dollars allocated to pay for the project.
  
  
• U - There's urgency on the customer's part.
  
  
• N - You understand the customer's true need.
  
  
• C - You have coaching in place.
  
  
• E - The identity of the economic buyer or final decision-maker is known.
  
  

The criteria should help managers refrain from second-guessing the sales estimates from their staff. "Because Roger always estimates low and sandbags his numbers, I'll adjust his forecasts upward by 20 percent. And because Marcia is always wildly optimistic, I'll cut her projected sales by half."

The funnel criteria, when selected properly, will help prevent adjusted numbers.

But you also need someone in charge of the funnel system to ensure that all the salespeople are using it and that everyone is abiding by the same criteria, someone who can hold people's feet to the fire. The important thing here is that the person has to have enough clout to hold people accountable for their individual funnels.

The funnel "meister" should be empowered to hold people accountable. If, for example, customer prospects must have a 50 percent probability of closing before they can be placed in the middle of the funnel, then a salesperson who is closing just 25 percent of those deals needs to be taken to task. Could that salesperson, for instance, be prematurely placing those leads into the middle of the funnel before they've been properly qualified? That's why the criteria needs to be specific enough to make such assessments, and you need at least three or four for each funnel location.

Interestingly, experienced sales managers usually know what those criteria should be, because they have a sense for the typical obstacles that tend to hold up deals for their company's products.

Large businesses should consider having a different funnel for each major product line, especially as some products have different selling cycles. Additionally, they might need to have a separate funnel for each major geographic region. The criteria needs to reflect changing markets. In normal times, for example, your sales cycle might be six months. But in a recession, that time period could easily balloon to over a year. That's why in volatile markets, you're better off using a sales cycle that is adjusted, using a moving average over several cycles or, better yet, a weighted moving average that places more emphasis on recent periods of time.

In addition, remember that your sales cycle is just an average. In general, larger deals will tend to take longer to close because they will usually involve more people in the approval process. Also, sales to new customers will typically take much longer than sales to existing accounts, especially when those deals involve products that are new to the market.

When implementing a funnel system, companies should consider having at least weekly meetings for people to give updates of their funnel activity. Peer influence helps keep people honest to their commitments. The real power of the funnel comes over time when people can track the progress of various potential deals, and identify what's stalled so that a plan of action for moving prospects along can be developed.

World-class sales organisations place a value on process, and accurate forecasting should always be part of that process. To improve your forecasting, the use of the sales funnel concept can help tremendously, but implementing it takes concerted effort and an awareness of potential pitfalls. In particular, companies should be on the lookout for the following common mistakes: 

1. Allowing prospects to "whirlpool"
   
   Every sales organisation has customer prospects in the middle of the funnel that go round and round but make little progress towards moving to the bottom and getting closed. You might have an uncovered base, such as a key executive at the customer company who is unconvinced of the need for change. 
   
   
2. Confusing selling with buying
   
   Both the selling and buying process has seven basic steps, but they are markedly different:
   
   

   	Selling	Buying
   1.	Target prospects	Monitor status quo
   2.	Qualify leads	Recognise the need to change
   3.	Cover the bases	Define problem
   4.	Make proposal	Evaluate options
   5.	Close deal	Select best solution
   6.	Fulfill order	Implement solution
   7.	Up-sell and cross-sell	Assess value of solution

   
   
   The problem is that there's often a mismatch in where the seller is, versus where the buyer is. The classic mistake occurs when the seller is on step 5 (ready to close the deal) while the buyer isn't even on step 2 (that is, the customer isn't fully convinced that he has a problem). 
   
   
3. Treating all products the same
   
   Companies should consider ranking their product lines (i.e. order of potential revenues or profits) so they can spend more time tracking and forecasting those products that will have a bigger impact on the bottom line. 
   
   
4. Not making special allowances
   
   Sales forecasts based on the funnel concept will be accurate if you have numerous prospects that are all about the same order size. In such cases, the law of averages will prevail: some prospects will drop out while others will reach fruition so things eventually even out. When some of your sales opportunities are much larger than others, separate them from your forecast, perhaps by placing an asterisk (*) next to your quarterly projection.
   
   Deals that represent substantially more in revenue gain often require their own individual funnels so they can be tracked separately.
   
   
5. Failing to properly prioritise activities
   
   In general, salespeople tend to work the funnel from the bottom up, concentrating on the surest opportunities first and leaving the less certain stuff for last. That approach might seem to make sense, but the truth is that it leads to unnecessary volatility.
   
   Sales professionals soon realise their funnel is drying up. Panic then ensues, as everyone scrambles to find new business. Finding new leads could take months, and working them down the funnel requires more time - a lag that could result in a sales shortfall and missed forecast.
   
   To prevent that, always prioritise the three areas of the funnel in the following order:
   
   
   1. Bottom
      
      
   2. Top
      
      
   3. Middle
      
      
   The only way to ensure prospecting gets done is to prioritise it ahead of the work that needs to be done in the middle of the funnel. 
   
   
6. Not having a funnel "meister"
   
   You need someone in charge of the funnel system with a lot of clout and who knows the process and customer accounts. This person needs to keep people honest by asking questions like, "How can this customer be in the middle of the funnel when you don't really know the final decision-maker?". In other words, the funnel meister can't just blindly accept information from salespeople. They need to constantly question funnel information, because salespeople are absolutely notorious for being overly optimistic about their prospects. 
   
   
7. Making the funnel process burdensome
   
   On the other hand, you don't want to make the funnel process an onerous or thankless chore. The funnel has got to be a tool that enables the sales force to work more efficiently and effectively. Otherwise, people will do everything they can to avoid using it.
   
   The trick is to integrate the funnel process with what the sales force already does. Many companies, for instance, use an internal process to track the status of customer prospects. Much of the information required by a funnel review can be obtained from such a system. Important data can be transferred directly to a customer relationship management (CRM) system that many companies already use. CRM applications typically have the capability to perform sales forecasting, and that information can be used in your funnel reviews. 
   
   
8. Over-relying on CRM
   
   That said, sales organisations must be careful about relying too much on data from a CRM system. As with any type of application, the software is only as good as data input. When a CRM application is too cumbersome or difficult to use, salespeople will often input information without really thinking about what they're doing, because they just want to get the task over with. Inaccurate information can lead to inaccurate forecasts.
   
   To prevent that, sales managers have to be involved in the design of the CRM tool and should never completely relinquish that job to the IT department. If they do, they risk ending up with a system that salespeople don't really take seriously because the perceived benefits do not outweigh the effort required to use it.
   
   A good rule of thumb is that, if you can't teach salespeople the basics of how to use a CRM system within five or ten minutes so that they can hit the ground running (later, they can learn additional functionality), then the system is probably too complex. Any sales application needs to be intuitive enough so that people can essentially learn how to use it simply by using it. And ideally the software should interface with other tools that your salespeople regularly use, such as their mailing system, as well as portable devices. In short, both usability and accessibility are crucial.
   
   

Sales forecasts will always have some degree of uncertainty. After all, predicting the future is, at best, an inexact science. Indeed, you can't foresee and prepare for every possible contingency - nobody's crystal ball is ever that clear. But the sales function is a definable, repeatable process that can be tracked and managed, using a simple concept called the sales funnel. And if a process can be tracked and managed, it can be monitored to extrapolate the future from the present.

Author Credits