Entrepreneurial Leadership and Management . . . and Other Stuff

RSS
Apr
05

Forecast Accuracy

Inevitably, after each fiscal quarter, a couple (or few, or many, or most) of the companies I work with inform me that their sales results for the quarter were below expectations. By itself, of course, this is usually not a catastrophe (although it sometimes can be) and leads to some reflection about what went wrong. What drives me completely nuts, though, is when numbers are missed after the company had been forecasting higher numbers during the quarter. I go total berserk when those higher numbers were still being forecast late in the quarter – sometimes until the last day. It’s not only when numbers miss on the low side that makes me upset. Sometimes, even when numbers miss on the high side I find cause for concern.

To me, forecasting accuracy is the single most important measure of the quality of a sales team. And, thus, is a reflection of the CEO’s ability to manage the sales process and sales team.

Being able to accurately predict bookings and, more importantly, collections in a small business is absolutely critical to maximizing a company’s opportunities and growth. The more accurate the prediction of income, the better spending can be aligned and the more efficiently the capital gained from the sale of equity or absorption of debt can be utilized. Why sell more of the company or take on more debt than is needed?

Aside from the financial issues, there are very few things that can hurt morale in a company more than missing a couple of quarters back-to-back. It’s quickly seen as a sign of weakness in the product, company or market and makes employees question what they are doing and how many hours of their lives they are pouring into the enterprise. Doing what you say you’re going to do, and sometimes doing better, makes everyone feel like they’re dedicating their efforts to a worthy cause.

Companies have sales plans, of course, which are generally set up a year ahead of time. A year is a very long time for a small company and accurately establishing sales predictions that far ahead of time is virtually impossible to do correctly. Forecasting, on the other hand, while often done for longer periods, generally is studied for the current or next quarter. This shorter term look at sales should be far more accurate than the longer term sales plan.

Forecasting for small companies without much of a pipeline is difficult. In this case, if a deal drops out of the forecast, it’s not likely to be replaced by another deal that comes in ahead of schedule. When companies are small, closing those all-important first key deals should be the main focus of the sales force. After those initial deals are closed, however, the focus should change to increasing the size of the pipeline in order to create a stable and predictable revenue/bookings/cash stream.

There is no excuse for a company with an established sales team (direct or indirect) to miss the numbers it predicted just a couple of months earlier by any significant amount. This means high or low. If the numbers are high (not including bluebird deals that were never in the pipeline), then spending wasn’t optimized and growth through increased spending could happen sooner. If the miss is low, then the company potentially spent more than it should have and may run into cash problems earlier than planned.

There should be enough visibility into the deals in the pipeline and enough of an understanding of the sales cycle to be able to determine which deals are going to come in during a quarter and which deals aren’t. Certainly, as the company gets further into its quarter, the more accurate the forecast for that quarter should become. With this logic, one should assume that the forecast one week before the end of the quarter should be basically right on. It’s shocking to me how often that it’s not.

The CEO and VP Sales need to be held responsible for the accuracy of the forecast. In my experience, they are often either judged on their sales achievement with respect to their plan only or, even more often, with respect to some abstract level of sales that seems right for the maturity of the company in its particular market. This is not sufficient. Accuracy of forecast is an indication of how close the CEO and Sales VP monitor the pulse of the business. It demonstrates how well the two understand the customer, the sales cycle and the market. Without a detailed understanding of these, there is little chance that the company will be successful. Therefore, it needs to be one of the key and, at certain stages of the company, the primary key factor in judging the success of company management.

The way to make this work, of course, is through compensation plans, making sure the company is tracking the right metrics and always asking good questions. I’ll talk about some of this stuff in the future in a post about compensation plans and unintended consequences.

Can anyone else you know use this information? Please share . . .
Share on Facebook
Facebook
Share on Reddit
Reddit
Tweet about this on Twitter
Twitter
Pin on Pinterest
Pinterest
Share on LinkedIn
Linkedin
Buffer this page
Buffer
Email this to someone
email
 April 5th, 2006  
 Will  
 Management, Selling  
   
 11 Comments

11 Responses to Forecast Accuracy

  1. The case where this is particularly hard is in an early stage company that does large enterprise sales.  In that case there’s no granularity and the actual number depends a lot on a couple of key deals, and there’s always uncertainty with those. Your comments suggest a terrific sales comp plan, though – have a significant component that is based on the inverse square of the difference between actual and forecasted sales – possibly based on forecasts at beginning of quarter, one month in, and two months in.

  2. The case where this is particularly hard is in an early stage company that does large enterprise sales.  In that case there’s no granularity and the actual number depends a lot on a couple of key deals, and there’s always uncertainty with those.

    Your comments suggest a terrific sales comp plan, though – have a significant component that is based on the inverse square of the difference between actual and forecasted sales – possibly based on forecasts at beginning of quarter, one month in, and two months in.

  3. You’re, of course, right about the granularity of early stage enterprise deals.  You can’t value the forecast the same way at an early stage with that type of sales model.  Perhaps this says more about how a company enters such a market than how it forecasts sales in it. Your math for the comp plan makes a lot of sense.  Perhaps avoiding the term “inverse square of the difference” might make it sell-able to a sales team  😉

  4. You’re, of course, right about the granularity of early stage enterprise deals.  You can’t value the forecast the same way at an early stage with that type of sales model.  Perhaps this says more about how a company enters such a market than how it forecasts sales in it.

    Your math for the comp plan makes a lot of sense.  Perhaps avoiding the term “inverse square of the difference” might make it sell-able to a sales team  😉

  5. In thinking more about this, you could just take whatever decent comp plan you already have, and divide that by the square of the difference.  This way there’s no incentive to make “0” the target. Just tell them it’s like gravity…

  6. In thinking more about this, you could just take whatever decent comp plan you already have, and divide that by the square of the difference.  This way there’s no incentive to make “0” the target.

    Just tell them it’s like gravity…

  7. Sorry to crash your party with an opposing point of view, but this discussion rings of a typical theoretical response from someone who has never been in sales and had try to accurately forecast down through a couple layers of sales directors, managers and salespeople. Forecast accuracy is a real problem.  If it was an easy thing to solve, it would’ve been solved by now. Trying to solve it with a comp plan won’t work.  No matter what comp plan you put in place, sales people will find a way to use it to their advantage.  If you try to punish them for sandbagging, then they’ll just hit their forecasted number and withhold the rest of the business until the next quarter.

  8. Sorry to crash your party with an opposing point of view, but this discussion rings of a typical theoretical response from someone who has never been in sales and had try to accurately forecast down through a couple layers of sales directors, managers and salespeople.

    Forecast accuracy is a real problem.  If it was an easy thing to solve, it would’ve been solved by now.

    Trying to solve it with a comp plan won’t work.  No matter what comp plan you put in place, sales people will find a way to use it to their advantage.  If you try to punish them for sandbagging, then they’ll just hit their forecasted number and withhold the rest of the business until the next quarter.

  9. Yup, agreed.  There’s no particular solution and it’s a moving target.  Dealing with the unintended consequences of sales plans take a supercomputer to consider all permutations.  Have built many of these plans in the past, I have sorta figured out when I hit diminishing returns and then put the plan into action.  I don’t think I’ve ever implemented a sales plan, though, where I didn’t yell “doh!” at some point in the year when behavior I didn’t plan for was inadvertently rewarded by the plan. In my experience, implicit punishments in plans never work.  Rewards for exhibiting the desired bahavior generally do.

  10. Yup, agreed.  There’s no particular solution and it’s a moving target.  Dealing with the unintended consequences of sales plans take a supercomputer to consider all permutations.  Have built many of these plans in the past, I have sorta figured out when I hit diminishing returns and then put the plan into action.  I don’t think I’ve ever implemented a sales plan, though, where I didn’t yell “doh!” at some point in the year when behavior I didn’t plan for was inadvertently rewarded by the plan.

    In my experience, implicit punishments in plans never work.  Rewards for exhibiting the desired bahavior generally do.

  11. Pingback: Forecasting Sales in 2009 « 2-Speed

%d bloggers like this: