More articles on Forecast Accuracy:
Forecast Accuracy in the InsightSquared Platform
Forecast Percent Error
Our Forecast Percent Error calculates the percentage difference that conveys how close your first OR last forecast was to your final bookings. For IS2 to perform this calculation, you must have data entered in Forecast Submission.
First/Last Forecast Booked
Our First/Last Forecast to Booked calculations show how much you booked of either the first forecast submitted in the period or the last forecast submitted in the period. If you booked more than was forecast, this will be >100%, if you booked less than was forecast, this will be <100%.
Where can I see Forecast Percent Error & First/Last Forecast Booked?
You can add either calculation to any Forecast Submission report as a filter, variable, or graph/table metric.
How do I interpret Forecast Accuracy?
Long story short:
0% = perfect
>0% = the farther away you are from zero, the less accurate your forecast was
Forecast Percent Errors convey how close your first or last forecast was to your final bookings, answering how right (or wrong) were you?
Consider the forecast as the zero-state for accuracy; eg. if the bookings exactly match the forecast, there is a ZERO percent error between them. So in this case, 0% = perfect. Obviously, forecasts are rarely perfect, but the goal is to keep your error percentage low, meaning your forecast was very close to what you booked.
Once you've determined the accuracy of your forecast, you can look at First/Last Forecast Booked to determine whether the forecasts were more or less than final bookings and continue inspecting forecast data to improve accuracy. Keep reading for some examples of how to use these metrics to improve your forecast.
Examples
1. The Formula at Work
If Sarah initially forecasted $10,000 (first forecast) then booked $14,000 (bookings), her Forecast Percent Error is 33%, and her Percent of Forecast Booked is 140%.
Forecast Percent Error: This calculation answers how accurately Sarah forecasted for the period. Using the percentage difference formula, this example's calculation would look like this:
Percent of Forecast Booked: This calculation answers how much of the forecast was booked, so we calculate what percent of the initial forecast ($10,000) was booked ($14,000).
Sarah booked 100% + an additional 40% of her initial forecast = 140%
2. How a manager can use Percent Error to improve a rep's Forecast Accuracy
Here's an example of how we can use the Forecast Accuracy dashboard & reports to improve a rep's forecast accuracy. So first, we need to know who needs coaching.
Who on my team has a high average forecast error percent? This establishes which reps have higher error percentages so we know who needs further coaching. Looks like Paul has a high percentage, we need to know if that's a trend.
Was the rep's error percentage consistently high? Or was there a month this year that had an unusual outcome that offset their average? (Sometimes there are surprises that can throw off a forecast at no fault to the rep, like a company unexpectedly losing a key exec and a big deal disappearing.) So we open the Forecasting Percent Error Over Time by Person report, click on Paul's name to focus on his data only, and discover that his error percent is consistently high so this average is an accurate representation of consistently inaccurate forecasts. We need to know where they're going wrong.
Is their forecast consistently high or low? While percent error shows how far off a forecast was, Percent of First Forecast Booked shows whether a rep booked over or under their forecast. (If a rep books more than they forecasted, this number is >100%, if they booked less than they forecasted, it is <100%.)
Paul's Percent of First Forecast Booked is 122.7%, so it looks like his forecasts have been too low. We need to know why.
What information is the rep using to determine their forecast? Now we have a chance to communicate with Paul, armed with information, and coach him on what he should be using to determine his forecast and how he can improve his process. We need to track for improvement.
Do we see an improved trend? As Paul works to improve his process, we should see his percent error trend down as his forecasts become more accurate.
More articles on Forecast Accuracy:
If you run into any issues, please contact support.
Comments
0 comments
Please sign in to leave a comment.