Visuals different in Desktop and Online

[Gingerjeans88]

Hi all,

Can anyone help me understand why a visual is showing two completely different numbers in desktop and online? I published it direct, seconds later go to Service and the visual is wrong? 

Desktop:

Online

Additionally - my custom date dimension table seems to be counting more records than I want it to. Before I share the PBIX, are there any obvious reasons for this?

My report is designed to show volumes of applications each year - each application is linked to a Course, which has a School Year (just a text value in a dropdown list). This is how they used to count volumes, and in a way they still do - but more importantly they want to see how many were submitted per month (rolling total - hence the above visual). 

The numbers should technically always be the same, as they only open up course for applications once the custom school year (third monday in september) has started, so all applications tagged with the School Year should also only be after the beginning of the custom schol year. If I recreate a card visual in a new dataset without the custom date table and just use the 'date' field from the applications fact table, the number matches their CRM which is what the report is based on. 

In my main report, with the date dim table, the count shows about 90 more and I presume this is the culprit. The date dim is not trimmed.

Any ideas?

[Anonymous · Accepted solution]

Hi  @Gingerjeans88  ,

1. First, clear the Filter to ensure that the filters displayed on the data on Power BI Desktop and Power BI Services are the same.

2. In Power BI Services, UTC date time will change. If the dax function uses Today() or Now() these functions may encounter this problem, because Power BI Desktop uses our native language, and Power BI Services uses the language of the region where the account is located.

Best Regards,

Liu Yang

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

[Gingerjeans88]

PS. formula for my date dim.

im_Date =

VAR A =

ADDCOLUMNS (

CALENDAR ( DATE ( 2017, 1, 1 ), DATE ( 2025, 12, 31 ) ),

"AYEAR", YEAR ( [Date] ),

"Month", MONTH ( [Date] ),

"weekday", WEEKDAY ( [Date], 2 ),

"WEEKNUMOFMONTH",

WEEKNUM ( [Date] ) - WEEKNUM ( DATE ( YEAR ( [Date] ), MONTH ( [Date] ), 1 ) ) + 1

)

VAR B =

ADDCOLUMNS (

A,

"ISFirstday of a fiscal year",

IF ( [Month] = 9 && [weekday] = 1 && [weeknumofmonth] = 3, 1, 0 )

)

RETURN

ADDCOLUMNS (

B,

"FiscalYear",

SUMX (

FILTER ( B, [Date] <= EARLIER ( [Date] ) && [AYEAR] = EARLIER ( [AYEAR] ) ),

[ISFirstday of a fiscal year]

) + [AYEAR] - 1

)