The article in this collection, “All Probabilities are the Same,” stated that all sales probabilities are 50% - either a sale will occur or not. Assuming that this statement is true, or alternatively, that calculated or feeling-based probabilities are not accurate, how then, can sales forecasts be made? Clearly, sales forecasts are critical to running any business to help determine material requirements, production capacity needs, customer support levels, and, of course, cash to fund all operations. Large variances between forecasts and actual results can quickly drive a company out of business.
A far more accurate forecasting method relies on determining the actual progress that an opportunity has made through the entire sales process. As the title of this article states, progress is the predictor. This concept is based on one major assumption: all sales move through a number of discrete steps of a known sales process. With some thought and experience, those steps can be easily identified and tracked. From past experience with a number of customers serving widely varying market segments and customer types, it has been found that about 35 discrete steps or activities take place in a typical sales process. Obviously, the complexity of those steps and time it takes to resolve each one can vary considerably. However, in virtually every case, each step had to be successfully passed on the way to success.
This concept is nothing new. Sales Funnels or Sales Pipelines or even Sales Probability assignments are often used to implement this concept. The difference described in this article is the removal of the time element that is customarily assumed in the common funnel and pipeline approaches. With these methodologies, the sales process is thought of as a number of sequential steps that are followed in a specific order. In the Progress as a Predictor model, the steps can be completed in any order with some of them actually pursued out of order to significantly increase the confidence of the forecast.
A pipeline analogy illustrates this point. Think of a pipeline with each step (perhaps 35, as mentioned above) appearing as a valve that can be fully opened or fully closed to impede the flow of the pipeline. In order for the liquid to flow from one end to the other (an order to take place), it doesn’t matter if any number of valves are closed nor does it matter which valves were opened when. All that matters is that all valves are opened at the same time. So, any one valve being closed can permanently stop the flow.
By extension, this model shows the fallacy of taking into account the historical average sales cycle when considering the likelihood of receiving any specific order. There is simply too much time variance associated with each step that can occur and, with some planning, some of those steps can be pursued in parallel. A real world example illustrates this issue. In this situation, the sales rep had won the approval of all financial, operational, and senior executives and was confident that we would receive the order in the next few days. I asked one question: “Has the prospect’s legal department reviewed our standard contract and found it acceptable?” You know the answer I received; we never were able to come to acceptable terms on this “done deal”. In hindsight, we could have begun discussions with the company’s attorneys on the legal aspects of the contract weeks before and discovered the showstopper issue (their required ownership of our intellectual property). Had we known that issue, perhaps we could have suggested a different relationship or walked away earlier knowing that “that valve” would never be opened.
By examining all valves (steps) simultaneously instead of in their “logical” order, parallel activities or potential showstopper issues can be addressed in an out-of-sequence order to remove as many variables as possible.