Three Forecasting Models (and the Forecasting Accuracy Myth)

3 min read

Kluster‘s Co-Founder and CCO, Rory Brown dispels the forecasting accuracy myth and presents three forecasting models to help you make revenue-defining decisions. Watch the video or read the transcript below.

Today on Klog, we’re going to talk about the forecasting accuracy myth. Time, as we know, is constant. But forecasting accuracy can only happen in a particular moment in time. That point of accuracy will be different for every company. But if your company has a wavering accuracy curve, the quality and integrity of the decisions you make will be impacted.

Here we present three forecasting models you can use to make sure that you don’t end up in this chasm 👇

Forecasting method 1: Category Forecasting

Many of you will know this as using best and commit in your CRM.

Here is the fundamental flaw of category forecasting. As illustrated by the red dotted line above, the first problem is that we need total pipeline available to be visible to us in order to make judgments on that pipeline.

The second problem is you’ve then got to wait until all the salespeople have made their judgments on that pipeline.

So there’s a big gap in terms of visibility in this area 👇

Forecasting method 2: Weighted Forecasting

Very simply, weighted forecasting is applying historic stage conversion rates as a weighting to your live sales pipeline.

The cons of a weighted forecast are we still have the wait for all the pipeline to be added before the weighting is accurate.

However, the visibility gap is smaller than the category forecast, so we do have a little bit more visibility.

Pros are that the weighted forecast is statistically sound, which means your arguments hold up in court.

The second pro is we don’t actually have to wait for the judgments of the sales team to get an accurate outlook.

Forecasting method 3: Machine Learning Forecasting

Essentially, this method takes lots of data points and historic trends to then make a prediction of where your revenue is likely to end up.

What are the pros of a machine learning-based forecast?

Firstly, we remove the visibility gap that we saw in the category and weighted forecasts. And that’s for one simple reason. Before the time period begins, machine learning uses data and activity to make a prediction of how much pipeline you’re going to end up with in the funnel and then uses that to make a prediction.

So what we end up with in the machine learning forecast is a slightly wavering line in terms of accuracy. But with much greater confidence in the early stages of the time period than we were getting with the category and weighted forecasting models.

Find the right sales forecasting software for your business

Forecasting software is no longer a nice to have. It has become an essential component of high-growth businesses.

But finding the right provider for your business and sales organisation can be a daunting task.

Download our Buyer’s Guide to Choosing the Right Sales Forecasting Software for 9 considerations when evaluating and selecting the right forecasting software provider.

Share your thoughts

Up ↑

%d bloggers like this: