B2B sales forecasting is challenging because most B2B companies lack the large sales volumes required for statistical forecasting techniques.
As a consequence, those companies that care about their forecasts usually choose one of two methods, the weighted pipeline and forecast categories. Both are easy to implement but have notable drawbacks.
The weighted pipeline approach calls for applying a closing probability to your live opportunities (usually related to their pipeline stage). The sales forecast for a given quarter equals the amount of the opportunities closing in that quarter multiplied by their closing probability.
The advantage of this method is its simplicity. The drawbacks are the judgmental amounts, closing dates, and closing probabilities. It also ignores the all-or-nothing nature of B2B opportunities.
The forecast categories approach involves allocating your live opportunities to categories that represent various degrees of forecast certainty, depending on the opportunities’ pipeline stage.
This method is also simple, and it acknowledges the all-or-nothing nature of B2B opportunities. The drawbacks are the judgmental amounts and closing dates.
Several elements can help companies mitigate the drawbacks of these forecasting techniques:
1. A documented sales process
Both of the above methods require that your opportunities be in the "right" pipeline stage, i.e., the stage that accurately reflects how advanced they are in your sales process.
For that, a well-documented sales process is key. Sales reps must understand where they are and what they should do based on their prospects' buying processes. A short list of requirements by stage is indispensable, but don't overdo it: Your sales pipeline must remain an operational tool, not an administrative ledger.
Some CRM applications let you create validation rules for advancing opportunities from one pipeline stage to the next. While such validation rules are stricter than simple requirements, they can be quite cumbersome for sales reps handling dozens of opportunities. You don't want them to degrade data quality, with sales reps not updating their pipeline because of "fastidious" clicks. So we recommend that validation rules be used sparingly, for really important requirements, like contractual documentation or compliance procedures.
2. Reality checks for closing dates
Both techniques are also exposed to bad closing dates. This is a serious issue. Respondents to CSO Insights' 2011 Sales Performance Optimization Survey typically indicated that fewer than half of their sales opportunities closed in the time frame predicted. We all know what "slippage" does to sales forecasts.
I would like to suggest three reality checks for closing dates, which should improve your sales forecasts significantly.
Hunt down outdated closing dates. SalesClic research shows that at any given time, 30 percent of the opportunities in B2B sales pipelines have outdated closing dates. This is clearly unacceptable. Standard CRM software reports and alerts will help you eradicate those pests.
Identify stagnating opportunities quickly. Both SalesClic research and Sales Benchmark Index research show that, on average, it takes twice as long to lose an opportunity as it does to win an opportunity. The reason is simple: The opportunities that are eventually lost tend to stagnate for a long time before they are determined to be lost, which is why you must spot them as early as possible in your pipeline. Doing this intelligently (i.e., based on the analysis of your historical data) is quite complex, since it implies measuring and interpreting your sales velocity in real time. But selected sales analytics applications can help you.
Backpedal from the closing dates. You know your sales process: the requirements for each pipeline stage and the time they usually take. For a given opportunity, start from your closing date and backpedal through your sales process to the current stage of that opportunity. Do you arrive at today's date? That is usually an effective check.
3. Closing probabilities: Beware decomposition
Forecast categories are immune to bad closing probabilities, but the weighted pipeline is very exposed. Copious research in behavioral economics has established that humans are bad at assessing probabilities, which, regardless of the politics involved, explains why weighted sales pipelines are frequently biased.
As a correction, some sales consultants recommend that you decompose the closing probability into (a) the probability that your prospect will complete the corresponding project, and (b) the (conditional) probability that you will be the preferred supplier for this project:
P(closing) = P(project) x P(supplier)
I am not convinced, for reasons that are detailed in this article. I don't necessarily recommend that you drop decomposed closing probabilities altogether if you are using them (as a "forecast accounting" tool they can be interesting), but at a minimum suggest that you combine them with alternative, ideally data-based methods.
Thomas Oriol is the director and CEO at Nimble Apps Limited, the Dublin-based publisher of SalesClic, a sales pipeline visualization, analysis, and forecasting solution for B2B companies. Prior to founding Nimble Apps, he spent spent 10 years advising European technology companies on mergers and acquisitions.