The problem with probability

Probability is the measure of the likeliness that an event will occur. It’s very common for sales leaders to use probability to measure the likeliness that deals will close and that revenue will occur. It’s an easy, common, but highly inaccurate method of reporting a sales forecast.

Here’s why using probability is not the best method for sales forecasting:

  • Selling is binary. It’s a yes or a no. You either win the deal or lose the deal. You don’t win a percentage of a deal.

  • Probability weighting assumes all opportunities have the same level of risk in each sales stage of the sales process which is never true.

  • We all know that as an opportunity progresses through the buyer’s journey, that the likelihood the opportunity will be won does increase as the opportunity moves through the process. Also, the inherent risk of an opportunity should be decreasing as it progresses. If every sales person and deal was exactly the same, these would be correct statements. But as we all know, they are not.

It is possible to look at your historical performance and use conversion rates by stage to weight a pipeline or a forecast. This works for the CXO, where they just need a high level number and are not concerned about what makes up that number. At the frontline, weighting does not work. Managers and reps need to speak in terms of real deals and the risks inherent to those deals. Explaining what makes up 60% of your proposal/quote stage or 80% of your negotiation stage does little to help sales managers and reps focus on the right deals and forecast accurately.

So what’s better than weighting?

Use a forecast methodology based on risk scenarios. Here’s how it works:

  1. Once an opportunity moves into a stage in your sales process whereby it should be “forecastable” and has a close date of the particular forecast period, that opportunity goes on a potential forecast list.

  2. The sales rep then assigns a risk scenario (for example Commit, Likely, Best Case) to the opportunity. If there is no risk scenario assigned, then the opportunity most likely is not closeable in that forecast period and should be assigned a more accurate close date where it is forecastable. Best practice risk scenarios include:

    Commit – Which includes revenue already won/booked plus deals where the sales rep has secured a verbal.

    Likely – Commit scenario total plus opportunities that are closeable in the forecast period, but still require some work and carry an acceptable level of risk. An example would be an opportunity where your solution is in the finals and the decision is due in the forecast period.

    Best Case – Likely scenario total plus opportunities that will be a stretch to close in the forecast period. A best case opportunity example would be an opportunity that you can pull into the quarter if needed, but it would take some work to get done. Maybe discounting, perhaps different legal terms, etc.

  3. All 3 scenarios are totaled at the rep, manager and CSO levels. The dollars are backed by fact – the actual deals and values. When managers meet with reps to discuss the forecast, they talk about actual forecast opportunities and the risks specific to each opportunity.

  4. If a forecast over-ride is required by a manager, that manager can add or take opportunities out of the forecast. If a manual forecast (judgment) is required (where a sales manager can add or take away dollars from the forecast that are not opportunity-backed), then the forecast should include a gap (lower than quota) or upside (higher than quota) calculation when compared to actual opportunities in each risk scenario that the manager has to explain. Manual forecasts are common for many businesses that have:

    A. Run-rate (renewal) revenue where management wants to account for those dollars.

    B. Business that closed faster than the reps can get it entered/updated in CRM.

    C. And Large businesses with varying behavior across the senior sale manager level.

There’s a lot more than just picking scenarios to ensure forecast accuracy. You also have to make sure you that your opportunities are in the right sales stages, have the right value and an accurate close date selected. This is directly related to how your sales team adheres to your sales process and the data entry requirements for your CRM system.

Don’t weigh you sales team down with probabilities and discounted forecast weights that do not add to their ability to improve forecast accuracy. If you know the numbers are inaccurate and based on a flawed methodology, then so do the reps. Give them a process and a model that really helps them to give you the accurate insights into your forecast, in the most simple and intuitive manner possible.