New DAX Function to Compare Partial Month-to-Date with the Same Partial Period Last Year

[MauriSchulz]

Power BI currently does not provide a native DAX function that allows comparing month‑to‑date values with the same proportional period of the previous year. Existing time‑intelligence functions such as SAMEPERIODLASTYEAR, DATEADD, and PARALLELPERIOD always return the full month from the previous year, which creates inaccurate comparisons when the current month is still in progress.

This limitation directly affects one of the most important analytical needs in business: measuring operational and commercial performance by comparing the same period across different years.

For example, if today (or the last sales date) is May 13th, 2026, I need to compare:

• May 1st–13th, 2026 with

• May 1st–13th, 2025

  However, SAMEPERIODLASTYEAR returns May 1st–31st, 2025, which makes the comparison invalid because the periods are not aligned. This prevents accurate performance evaluation, especially in sales, retail, finance, and operational monitoring.

  Additionally, when navigating future months (e.g., June 2026), Power BI still returns values after May 13th, 2025, which should not be included because the current year has no data beyond May 13th.

  
  Example of the measure I currently need to write to solve this manually:
  

M_Margem_Vs_LY_partial =
VAR MaxDataGlobal =
    CALCULATE (
        MAX ( fSales[Data] ),
        REMOVEFILTERS ( dProdutos ),
        REMOVEFILTERS ( dSubCategoriaProdutos ),
        REMOVEFILTERS ( dCategoriaProdutos ),
        REMOVEFILTERS ( dLojas ),
        REMOVEFILTERS ( dLocalidades ),
        REMOVEFILTERS ( dPromocoes )
    )
VAR MaxDia =
    DAY ( MaxDataGlobal )
VAR MaxMes =
    MONTH ( MaxDataGlobal )
VAR MaxAnoLY =
    YEAR ( MaxDataGlobal ) - 1
RETURN
    CALCULATE (
        [sales],
        SAMEPERIODLASTYEAR ( dCalendario[Data] ),
        FILTER (
            ALL ( dCalendario ),
            YEAR ( dCalendario[Data] ) = MaxAnoLY && MONTH ( dCalendario[Data] ) <= MaxMes
                && NOT (
                    MONTH ( dCalendario[Data] ) = MaxMes && DAY ( dCalendario[Data] ) > MaxDia
                )
        )
    )

This measure works, but it is long, complex, hard to maintain, and not intuitive for most users. A native DAX function would solve this elegantly.

Proposed Solution:

Introduce a new DAX function such as:

PARALLELPERIOD_MATCHDAY( <measure>, <reference_date>, [<offset_years>] )

This function should:

• Automatically align the previous year’s period to the same day of the current month

• Prevent returning dates after the last available date of the current year

• Support month, MTD, and YTD contexts

• Enable accurate performance measurement for the same period across years

• Eliminate the need for complex manual filtering

• Improve accuracy and usability for real‑world business scenarios

Benefits:

• Accurate year‑over‑year performance evaluation

• Cleaner and simpler DAX

• Better performance

• Easier adoption by analysts and business users

• Reduced risk of incorrect calculations in executive dashboards
  
  Example: