I've been struggiling with this issue for some time now and not sure what the issue is.
I have got monthly data extracts, which provide submission data. One column provides 'start of the week' and I use this for the timeline and then a Count of submissions to create a basic line chart. My aim was to add the Forecasting feature within Power BI to Forecast 6 months.
My issue is, the Forecast line will only appear if I ignore the last months data. If it is set to 0 (which I want, as I want it to create a forecast taking into account every month) it will error and say there is a date issue. If I ignore the last month, it works..
I have got 5 months worth of data, and everytime I update it, it works, but only if I ignore the latest month. Any ideas?
Hi @Kai_PBI ,
I agree with SamWiseOwl , and here's what I'd like to add.
The first thing that needs to be applied to the x-axis is the date hierarchy
Then on top of that you need to drill down to the month level
Then turn on the predict button to make it happen
As you reported, it is not possible to make predictions without regular intervals, just as if you drill down to the day level again, it will suggest that it is not possible to make predictions because the date of the previous data point, is not the same as the date interval of the next date point. The only way to do normal separation is by month, year or quarter.
Best regards,
Albert He
If this post helps, then please consider Accept it as the solution to help the other members find it more quickly
Hi @Kai_PBI ,
I agree with SamWiseOwl , and here's what I'd like to add.
The first thing that needs to be applied to the x-axis is the date hierarchy
Then on top of that you need to drill down to the month level
Then turn on the predict button to make it happen
As you reported, it is not possible to make predictions without regular intervals, just as if you drill down to the day level again, it will suggest that it is not possible to make predictions because the date of the previous data point, is not the same as the date interval of the next date point. The only way to do normal separation is by month, year or quarter.
Best regards,
Albert He
If this post helps, then please consider Accept it as the solution to help the other members find it more quickly
The forecast line dissapears entirely. I then get a error staying the data is too irregular.
Helpful resources
Fabric Community Update - July 2025
Find out what's new and trending in the Fabric community.
Power BI Monthly Update - July 2025
Check out the July 2025 Power BI update to learn about new features.
