Power BI Time Intelligence for the Indian Financial Year (April–March)

⏱ 2 min readUpdated 28 September 2026

Power BI’s time-intelligence functions assume a January–December year unless you tell them otherwise. Indian businesses run April–March, so default YTD numbers reset in the wrong month. Here is the setup that gets it right.

In this article
  1. 1. A calendar table with fiscal columns
  2. 2. Measures with the fiscal year end
  3. 3. Visuals

1. A calendar table with fiscal columns

Calendar =
ADDCOLUMNS (
    CALENDAR ( DATE ( 2019, 4, 1 ), DATE ( 2027, 3, 31 ) ),
    "FY",       "FY " & FORMAT ( IF ( MONTH ( [Date] ) >= 4, YEAR ( [Date] ) + 1, YEAR ( [Date] ) ), "0" ),
    "FY Month", MOD ( MONTH ( [Date] ) - 4, 12 ) + 1,
    "FY Qtr",   "Q" & ( INT ( ( MOD ( MONTH ( [Date] ) - 4, 12 ) ) / 3 ) + 1 ),
    "Month",    FORMAT ( [Date], "MMM" )
)

FY Month runs 1 (April) to 12 (March). Sort the Month column by FY Month, and Mark as date table.

2. Measures with the fiscal year end

Sales FYTD   = TOTALYTD ( [Total Sales], 'Calendar'[Date], "3/31" )
Sales LY     = CALCULATE ( [Total Sales], SAMEPERIODLASTYEAR ( 'Calendar'[Date] ) )
Sales FYTD LY= CALCULATE ( [Sales FYTD], SAMEPERIODLASTYEAR ( 'Calendar'[Date] ) )
Sales QTD    = TOTALQTD ( [Total Sales], 'Calendar'[Date] )
Growth %     = DIVIDE ( [Sales FYTD] - [Sales FYTD LY], [Sales FYTD LY] )

The "3/31" argument ends the year on 31 March. Quarters already align (Apr–Jun = calendar Q2 = fiscal Q1) because the fiscal year starts on a quarter boundary — only the labels change, which the FY Qtr column handles.

⚠️ Time-intelligence functions need a continuous date table with every day. Gaps or using the fact table’s date column directly cause blanks or wrong totals.

3. Visuals

  • Matrix: FY and FY Qtr in rows, Sales FYTD and Growth %.
  • Line chart: Month (sorted by FY Month) on the axis, current vs last year.
  • Slicer on FY.

Background: CALCULATE and filter context and star-schema modelling.

✨ Ask AI about this article

Stuck on a step? Ask a question and the AI answers using this article.

Free · AI can be wrong