What are the key performance indicators (KPIs) that you use to measure your revenue forecasting accuracy?

I disagree with removing data points. Changing the data doesn't help you improve, which should be the goal, not making your forecasts appear more accurate than they actually were. Instead of removing outliers, use WMAPE. WMAPE = Weighted Mean Absolute Percentage Error Pros: allows you to consider significance, robust to outliers or extreme values (disabling any data points from disproportionately impacting your results), highlights areas to prioritize accuracy improvement, customized approach. MAPE = Mean Absolute Percentage Error Pros: Measures your average forecast accuracy, straightforward comparison between forecasting methods and model, can be used with a wide variety of aggregate data points.

As CFO and VP of Operations for a furniture production and distribution company, I used KPIs like sales, gross profit margin, operating expenses, and ROI to measure the company's success. As president of nonprofit making garments for the US military, I use KPIs like the number of garments produced, the percentage of garments that meet quality standards, and the percentage of garments delivered on time. By tracking these KPIs, I can measure both organizations' success and make better decisions about allocating resources and improving performance.

I recommend removing any outliers caused by unforeseeable events when calculating forecast variance. For example, let’s say your company’s website was down for a day. That would negatively impact your forecast variance, but that’s not something you could have accounted for. So I’d remove that data point. If, however, the website being down becomes a recurring theme, make sure you account for that in your model moving forward.

The most essential KPI for revenue forecasting accuracy is the comparison between your forecasted and actual revenue. Essentially, this KPI calculates the difference between your predicted and real earnings and illustrates how accurate your predictions were. It's calculated by dividing the difference between your forecast and actual revenue by your forecasted revenue and then multiplying by 100 to get a percentage. Ideally, the percentage should be as low as possible.

Forecast variance measures the difference between what was predicted and what actually occurred. It's a valuable KPI in various fields, highlighting the accuracy of predictions. A lower forecast variance signifies a closer alignment between forecasts and actual outcomes, which is often a key goal in many professions, including accounting.

Forecasting Accuracy KPI's even though critical to measure how we are forecasting against actual revenue for a given period, lot of thought & analysis is needed to identify the forecast lag to use to compare it against actual revenue incurred. As we move closer to the month the sales are actualized the forecast becomes better but most of the buying decisions are made long time before that due to long supplier lead times. So, forecast bias provides a good trend and measure our forecasting if we are consistently over or under forecasting. If we are minimizing the gap between our projected forecast and actuals every month we are moving in the right direction or else, we need to come up with better ways to incorporate market intel.

Forecast bias measures the tendency of forecasts to consistently overestimate or underestimate actual revenues. It quantifies the systematic errors or tendencies in the forecasting process and provides insights into the direction and magnitude of forecasting errors. Positive bias indicates that forecasts tend to be too high, while negative bias suggests that forecasts tend to be too low. Monitoring forecast bias helps identify potential sources of error, such as data inaccuracies, model assumptions, or behavioral biases, and enables organizations to adjust their forecasting methods accordingly.

MAPE is an extremely useful summary measure when managing forecasting projects with many points to compare. I used this heavily while evaluating store level accuracy for products being distributed through major retails such as #walmart and #costco. It became the primary target for my team to shoot for improving.

Measuring revenue forecasting accuracy is crucial for evaluating financial projections. Key performance indicators (KPIs) are essential for this assessment. One fundamental KPI is the Mean Absolute Percentage Error (MAPE), quantifying the average deviation between forecasted and actual revenue as a percentage of actual revenue. Tracking the percentage variance between forecasted and actual revenue provides a straightforward measure of accuracy. A lower MAPE signifies higher forecasting accuracy. Regularly tracking MAPE allows companies to pinpoint inaccuracies, refine forecasting techniques, and enhance overall financial planning and decision-making processes. By focusing on MAPE, businesses can ensure their revenue forecasts are reliable.

MAPE is a widely used metric for quantifying the accuracy of revenue forecasts relative to actual outcomes. It measures the average percentage difference between forecasted and actual revenues, providing a standardized measure of forecasting error. MAPE is calculated by taking the absolute percentage difference between forecasted and actual values for each observation, averaging these differences across all observations, and expressing the result as a percentage. MAPE allows organizations to assess the overall magnitude of forecasting errors and compare the relative performance of different forecasting methods or models.

Instead of using MAPE, always use sMAPE (symmetric-MAPE) since it takes care of the DIV 0 errors in MAPE calculations and makes sure the correct percentage error is reported

MASE is another performance metric commonly used to evaluate the accuracy of revenue forecasts, particularly in the context of comparing forecasts across different time series or forecasting horizons. Unlike MAPE, which measures forecasting error relative to the magnitude of actual values, MASE accounts for the variability and scale of the time series itself. By comparing the mean absolute error of a forecast to the mean absolute error of a naive or benchmark forecast, MASE provides a standardized measure of forecasting accuracy that is robust to changes in data scale and variability.

MASE assesses forecast accuracy by comparing it to a simple benchmark. A value less than 1 indicates better-than-benchmark performance, making it a valuable tool in various fields for refining forecasting methods.

FVA is a metric that quantifies the value added by our forecasting process. It helps identify areas where our forecasts significantly contribute to decision-making and where improvements may be needed. By focusing on FVA, we can optimize forecasting efforts for greater impact in our profession.

Forecast Value Added assesses the value contributed by the forecasting process relative to a naive or benchmark forecast. It measures the extent to which the forecast improves upon a simple, baseline forecast, such as a moving average or historical average. By comparing the accuracy of the forecasted values to the benchmark forecast, FVA helps organizations evaluate the effectiveness of their forecasting methods and identify areas where additional improvements can be made. Positive FVA indicates that the forecast adds value beyond what could be achieved with a simple baseline forecast, while negative FVA suggests that the forecast performs worse than the benchmark.

From my perspective, the reason to use revenue forecast during month end closing, just because some revenue is unknown at the time of month end closing, due to some business nature. But sooner or later, actual revenue will be available. Therefore, afterwards revenue true-up is necessary to test your forecast accuracy. I would like to use true-up variance as the most important accuracy indicator. Generally speaking, if true-up variance is 15% or less, I would consider the forecast acceptable, if true-up variance is bigger than 15%, I’d say you need to improve your forecasting accuracy. Otherwise, external auditors may question your forecasted revenue as well.

It's important to consider how the forecast was used every month. What business decisions did it drive? What's the expected profit or revenue increase that the business will realize because of the forecast?

Consider Cycle on Cycle Stability and Forecast Reasonability along with Batch Stability in addition to Forecast Accuracy and Bias

En mi experiencia los KPI para medir la precisión de la previsión de ingresos incluyen: 1. Tasa de crecimiento de ingresos: Mide el aumento porcentual de los ingresos en un período específico, lo que permite evaluar el rendimiento financiero de la empresa. 2. Margen de crecimiento de beneficios: Proporciona una instantánea rápida de la viabilidad económica de la empresa al calcular los ingresos totales y restar el costo de la mercancía vendida. 3. Generación de leads: Mide la cantidad de clientes potenciales que muestran interés en los productos o servicios de la empresa. 4. Participación en el contenido de marca: Evalúa el compromiso de los clientes con el contenido de la marca.

Organizations may also consider other factors when evaluating revenue forecasting accuracy. These may include the timeliness of forecasts, the granularity of forecast data, the stability of forecasting models over time, and the ability to capture and adapt to changing market conditions or business dynamics. Moreover, it's essential to consider the broader impact of forecasting accuracy on business outcomes, such as inventory management, resource allocation, and strategic decision-making. By adopting a holistic approach to measuring and improving revenue forecasting accuracy, organizations can enhance their ability to anticipate and respond to market fluctuations and drive sustainable growth.