Calculated column in transform data is increasing the data refresh by 4 hours

Hi marktait ,

What lbendlin  said does work, you can try to optimize performance by modifying your M function.
You can also try to try to optimize both the data source and the refresh.
1. Add the four columns you need ahead of time in the data source, as the data source is definitely running under a better load than the performance in Power BI.
2. Based on your description, the database you are using seems to be SQL Server. then you can try to use incremental refresh when doing the refresh. If you're using Power BI Pro or Premium, you might benefit from setting up an incremental refresh policy for your dataset. This way, only data that has changed will be refreshed, which could significantly reduce the overall refresh time. However, this approach requires careful planning around how your data is partitioned and updated. 
Incremental refresh for semantic models and real-time data in Power BI - Power BI | Microsoft Learn

Best Regards,
Dino Tao
If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.