September 7, 2022

A

Gustavo Melendez

TAGS

No items found.

Sales forecasting is an essential part of any business, yet it can be challenging to create meaningful forecasts. In this blog post, we will discuss a simple method for creating sales forecasts using Monte Carlo simulations. This approach relies on limited data and produces statistically meaningful forecasts with just a couple of clicks. With Monte Carlo simulations, you can model complex scenarios and get a better understanding of the range of possible outcomes and their probabilities.

A sales rep's pipeline includes every open opportunity that they are working on. In most CRM's, like Salesforce, an opportunity has 3 main data points that are used for sales forecasting: **amount**, **close date**, and **stage**. The opportunity stage is often associated with a win probability such that deals that are early in the sales cycle have a lower probability of closing vs those towards the end, which have a higher probability of being won. In other words, opportunity stage forecasting assumes that deals have a different likelihood of success at different sales cycle stages.

For example, if you know that you typically win about three-quarters of deals that enter the 'Negotiation' stage, you can predict that you have a 75% shot for all the deals in that stage. An expected revenue forecast will use this probability to determine an aggregate forecast amount across your pipeline.

Calculating an expected revenue forecast requires very little math:

- For each open opportunity in your quarter, multiply the amount by the stage probability.
- Sum each of the amount/probability products for each opportunity
- Add in the amount of any closed won opportunities

This method is straightforward but creates a sales forecast that is a single-point estimate. In other words, it only gives you a single number and does not take into account the range of possible outcomes. They also often produce forecasts that would not be possible. A deal can either be won or lost; there is never an outcome where your deal can close at 75% of the opportunity amount.

One common scenario that distorts expected revenue forecasts is when your pipeline has opportunities with drastically different amounts. So-called whales, or very large opportunities compared to the rest of your pipeline, can often drive an expected forecast number that would not be possible if those opportunities did not close on time. If 20 of your deals in the pipeline are around $10K each and 6 opportunities are $50K, just 1 or 2 of the bigger deals can easily sway the revenue for the quarter if they don't close on time. Visualizing the impact that each opportunity can have on the pipeline can be a time-consuming effort if done manually.

This is where Monte Carlo simulations come in. Monte Carlo simulations can be used to generate a distribution of possible revenue outcomes and their probabilities instead of a point estimate. It helps us visualize the possibilities across the pipeline and make more informed decisions.

Sales forecasting with Monte Carlo simulations is straightforward. A good monte Carlo simulation will run thousands of trials and randomly adjust variables in a model to determine the probabilities of each outcome. The outcome of each simulation is placed on a histogram, which allows us to visualize the most probable outcomes.

For example, in a pipeline with 20 opportunities slated to close this quarter, we might run 20,000 simulations where in each trial we randomly adjust 15% of our pipeline to close up to 30 days later than we currently expect. Each trial produces a new revenue forecast which is placed on the histogram. The more simulations that produce a similar outcome (forecast range), the more probable it is to happen. At the end of the simulation, we can visually see how probable each forecast is for our given scenario.

Of course, adjusting the simulation settings will produce different results, but in all cases, the insights produced by a Monte Carlo simulation will far outweigh the point estimates produced by an expected revenue forecast. You could choose to run a simulation that only adjusts a small portion of the opportunities in your pipeline, or one that adjusts all open opportunities. Simulations could be even more nuanced by focusing on opportunities with close dates near the end of the month, which are often the most likely to carry over into the next quarter or reporting period.

If you're already using Salesforce and are interested in trying out this approach to generate your sales forecasts, we've developed a simple app called Confidence, which can help you run thousands of simulations on your sales pipeline in just a couple of clicks. You could be generating sales forecast simulations in just a matter of minutes. Take a look at our walkthrough of Confidence below and sign up for free here.

Next:

LaborFirst