MSPowerBI

Information

• extension for local saves is *.pbix

Concepts

• https://docs.microsoft.com/de-de/power-bi/guidance/star-schema
• There are three different options on how your data should be treated: Import, DirectQuery, or Live Connection. This section will focus specifically on the Import option.

Power Query

Eine Power Query benutzerdefinierte Funktion ist eine Zuordnung aus einem Satz von Eingabewerten zu einem einzelnen Ausgabewert und wird aus nativen M-Funktionen und -Operatoren erstellt.

M Language

M ist eine funktionale Sprache:

• Der gesamte Code ist in Schritte unterteilt, die alle innerhalb eines Let-in-Statements definiert sind
• Jeder Schritt hat einen Namen, z. B. Quelle, oder auch #"Geänderter Typ"
• Schritte bauen zumeist (aber nicht zwangsläufig) aufeinander auf. Daher sieht man es häufig, dass der Name eines Schrittes im darauffolgenden Schritt innerhalb der Funktionen benutzt wird
• Derjenige Schritt, der nach dem in benannt ist (letzte Zeile im Skript), stellt das Ergebnis der gesamten Abfrage da. Das Ergebnis dieses Schrittes wird dem Nutzer – zumeist, aber nicht notwendigerweise – als Tabelle zurückgegeben.

Relationsships

Import aspects are:

• Auto-detected relationships
• There may be only one active relationship between two tables
• There may be an unlimited number of in-active relationships between two tables
• Relationships may only be built on a single column, not multiple columns
• Relationships automatically filter from the one side of the relationship to the many side Relationships cannot be built directly between tables that have a many-to-many relationship
• Cross-Filtering means a filter from a 1-to-many and many-to-1 relationship like filter on one dimension table to filter another dimension table via a fact table

The arrows defines the direction of filtering:

Hierachies

• https://radacad.com/what-a-power-bi-hierarchy-is-and-how-to-use-it

Modelling Organizational Hierachies

• https://ssbi-blog.de/blog/business-topics/wie-du-unregelmaessige-hierarchien-in-power-bi-und-power-pivot-nutzen-solltest/ fixed number of levels
• https://www.daxpatterns.com/parent-child-hierarchies/
• https://radacad.com/parsing-organizational-hierarchy-or-chart-of-accounts-in-power-bi-with-parent-child-functions-in-dax with two columns representing a parent relationship
• https://radacad.com/removing-blanks-from-organizational-ragged-hierarchy-in-power-bi-matrix-visual
• https://radacad.com/dynamic-row-level-security-with-organizational-hierarchy-power-bi

Measures

Calculated measures are very different than calculated columns. Calculated measures are not static, and operate within the current filter context of a report; therefore, calculated measures are dynamic and ever-changing as the filter context changes. Calculated measures can do the following:

• They can be assigned to any table
• They interact with all the relationships in the data model automatically, unlike calculated columns.
• They are not materialized in a column, and therefore cannot be validated in the Data View.

• https://www.enjoysharepoint.com/power-bi-measure-examples/ examples

Filter Context

Der Filterkontext beschreibt die Filter, die während der Auswertung eines Measures oder eines Measureausdrucks angewandt werden. Filter können direkt auf Spalten angewandt werden. Darüber hinaus können Filter auch indirekt angewandt werden. Dies passiert immer dann, wenn Filter durch Modellbeziehungen an andere Tabellen weitergegeben werden.

Change of the Filter Context

Es gibt zwei mögliche Standardergebnisse, wenn Sie der CALCULATE-Funktion Filterausdrücke hinzufügen:

• Wenn sich die Spalten (oder Tabellen) nicht im Filterkontext befinden, werden dem Filterkontext neue Filter hinzugefügt, um den CALCULATE-Ausdruck auszuwerten.
• Wenn sich die Spalten (oder Tabellen) bereits im Filterkontext befinden, werden die vorhandenen Filter durch neue Filter überschrieben, um den CALCULATE-Ausdruck auszuwerten.

CALCULATE(<EXPRESSION>, [<FILTER>*])

FILTER

The following definitions are equivalent

Revenue Red = CALCULATE([Revenue], 'Product'[Color] = "Red")
Revenue Red = CALCULATE([Revenue],
   FILTER(
       'Product',
       'Product'[Color] = "Red"
   )
)

ALL

• https://radacad.com/how-to-use-all-in-a-dax-expression-in-power-bi Many examples to ignore filter by specific slicers or other visuals or in total

DAX

• https://www.powerbi-pro.com/calculate-das-schweizer-armeemesser-in-dax/

Views

• 
  • Report View (Berichtsansicht)
  • Data View (Datenansicht)
  • Relationship View (Beziehungsansicht)

Ressources

• Microsoft Documentation
• Beginners Tutorial
• examples for data manipulation at import
• irregular hierachies e.g. for org model

Operation

Import Data from Excel

General Information

Power Query Editor always works with a preview of the data, to make the development process fast. When you load the data in Power BI, transformations will be applied on the entire dataset. Hence errors might have not been identified before the full import of data.

Steps

Define Import

• select import and file
• select transform (not load)

Delete Rows e.g. Header

• delete rows above headers (e.g. filter sums) by
• set first row as header line by

Delete Columns

• delete columns not needed by right click -> delete

Add Columns

• see https://docs.microsoft.com/de-de/power-bi/transform-model/desktop-tutorial-create-calculated-columns
• 
• example

= Table.AddColumn(#"Geänderter Typ", "Charge Fact", each if Value.Is([#"Chargeability (FTE)#(lf)"], type number) then [#"Chargeability (FTE)#(lf)"] else 1)

= Table.AddColumn(#"Geänderter Typ2", "Activity Type", each if [#"Partner-PSP-Element"] = "9914.P10048.004" then "ES Admin" else if [#"Partner-PSP-Element"] = "9914.P10048.005" then "ES Training" else if [#"Partner-PSP-Element"] = "9914.P10048.006" then "ES Relationship Management" else "tbd")

Delete Rows

• filter specific value (click to right down arrow in the column header and define the filter)

N charactors from left

• 
• next step delete the rows to the right
• rename column to the original name (w/o .1 suffix)

Adjust Query

• 

Not summing up a column like personnel number

• 

Error Handling

• example see here

Export Data to Excel

• move or right click in header
• select dots
• select export data

Relationsships

Change Default Summarization

• 

Visuals

Sorting according to Values

• Enter Data and give the table a name
  • if needed edit the data by which opens Power Query
  • select
• Create Relationship
• Create new Sort Column with e.g.

ES Activity Order = RELATED(ES_Activity_Order[ES Activity Order])

• set this column as sort column by

Slicer

• 
• Selection Behavior

Matrix

• is neary the same as pivot
• https://www.goodly.co.in/create-pivot-table-in-power-bi/
• grand total and subtotals

Mulit-Row Card

• Title On/Off: caption of multi-row card on top
• Visual Data -> Fields -> Right Click -> Rename for this visual: text below number
• Visual Data -> Category Label On/Off: show text below number
• Visual Data -> Category Label: format of text below number

Pie Chart

• sequence of fields defines the position, starts at 00:00
• Format -> Visual -> Legend -> Position
• Format -> Visual -> Slices -> Colors

Single Card

• 'Collout value' -> Font: size of number
• 'Category label' on/off: text below number
• 'Title': text above number

Stacked Bar Chart

• figures in bar by switching on 'data label'

Projects

ES KPI Reporting

• Table for Activity Sorting: Report View -> Enter Data -> ... and edit it via these steps

English/German

Slicer	Datenschnitt
E1	G1
E2	G2