Discretisation

Hi! We recently shipped the ability to 'bin' numeric fields, right click the field and choose 'Group'. We might continue this work to support measures if there is more support for it.

Hi Will, Thanks for the review. There are many articles on Discretisation written by prominent Tabular Gurus. Being able to dynamically bucket aggregate data is of vital importance. Even in SSAS - Multidimensional this was available, but the buckets could not be controlled by a user interface. For example if I want to know all my customers who purchase in the range 0-1000, 1000-2000, etc.. it is not possible to do this without a lot of acrobatics. However bucketing leads to 'exception' reporting which is the most important way in which exceptions can be highlighted. We did some work on this, for example if a customer has purchased less than previous year, we tagged as 'Reduced', if he has purchased more, we tagged as 'Increase' etc.. But these tags come in the data area of the pivot. What I need is a mechanism by which such tags can be available in the dimension area. The discretization should not be linked to any specific dimension, but it should work across any selected dimension. let me know if you need more details on this.

I'd really like to see this!

I have an idea related to this. I my opinion this and many other requirements could be solved by having calculated/dynamic tables. Those tables could have columns calculated at report time and affected by filters and measures calculated in this new deeper context. Coming from Tableau the measure/column distinction in Power BI is a bit odd to me.