How to Fix FORECAST.ETS Returning Zero Values in Excel
Question details
The user is attempting to forecast future data points but the FORECAST.ETS formula is returning zeroes instead of accurate predictive values.

- Product
- Microsoft Excel
- Device & OS
- not provided
- Scenario
- Using the FORECAST.ETS function to predict future trends based on historical time-series data.
- Observed behavior
- The formula yields zero or highly inaccurate results, typically caused by inconsistent timeline formats, incorrect seasonality settings, or referencing errors.
Before troubleshooting the formula arguments, ensure your historical data does not contain significant gaps or text errors, and verify that you have enough historical data points to establish a mathematical pattern.
Create and Use a Numerical Timeline
FORECAST.ETS requires a consistent, evenly spaced timeline to calculate correctly. Replacing text-based dates or inconsistent periods with a sequential numerical timeline often resolves zero-value errors.
When timelines consist of text formats (like 'March', 'April') or dates with irregular intervals, the ETS algorithm fails to calculate the data points accurately, frequently defaulting to zero.
Add a new column directly next to your existing text-based timeline or dates.
Enter sequential numbers starting from 1 (e.g., 1, 2, 3, 4) for each successive historical data period.
Modify your formula to use this new numerical column as the target date and timeline arguments.

Adjust the Seasonal Cycle Argument
An incorrect or absent seasonal cycle length can skew results or force a zero return. Adjusting the seasonality argument ensures the formula accurately detects your data's actual pattern.
Extend Formula References Properly for Future Periods
If only the first forecast period shows data and subsequent cells return zero, the historical data array may not be locked properly when copying the formula.
Forecast Data Accurately with WPS Spreadsheet
WPS Office offers a highly capable Spreadsheet application that fully supports advanced forecasting formulas, including FORECAST.ETS. If you need a reliable environment for data analysis, WPS Spreadsheet provides exactly what you need with full Microsoft Excel compatibility.
- 1. Open WPS Spreadsheet: Launch WPS Office and create a new spreadsheet or open your existing workbook containing the historical data.
- 2. Prepare your data: Ensure your timeline is arranged in sequential numerical order and historical values are formatted correctly.
- 3. Enter the formula: Select an empty cell and type =FORECAST.ETS() followed by your target date, historical values array, and timeline array.
- 4. Generate the forecast: Press Enter to calculate the forecast, then drag the fill handle to apply the predictive formula to subsequent periods.

Frequently Asked Questions
What is the purpose of the FORECAST.ETS function?
FORECAST.ETS predicts future values based on existing time-based data using the Exponential Smoothing (ETS) algorithm. It is especially useful for analyzing datasets that display seasonal patterns or steady trends over time.
Why do I get a #NUM! error when using FORECAST.ETS?
A #NUM! error usually occurs if the required arguments contain mismatched ranges. Verify that the size of your historical values array perfectly matches the size of your timeline array, and ensure the seasonality argument is set within acceptable parameters.
Can I use dates instead of numbers for the timeline?
Yes, you can use standard date formats, but the intervals between the dates must be absolutely consistent (e.g., exactly one month or one day apart). If intervals vary, the function struggles to identify the pattern and may return zero values, which is why substituting dates with a numerical timeline is a highly recommended fix.
How does automatic seasonality work in FORECAST.ETS?
By default, or when the seasonality parameter is set to 1, the function attempts to automatically detect the repeating seasonal pattern in your dataset. Keep in mind that automatic detection requires a sufficient number of historical data points to successfully recognize a cyclical pattern.




