I don't understand the question.

Given that table, you already have numbers for each Service and Location.  What else are you looking to compute?

{FIXED [Service],[Location] : sum([Number of Records])}/{FIXED [Service] : sum([Number of Records])}

Here it is:

Yes my Cal was wrong. I just did not read properly what you need, but looks you got what you wanted

Thanks

Deepak

Sure.

There are 2 parts to this calculation:

This formula returns the number of records for each combination of [Service] and [location]. I use this as the numerator to calculate a percentage.

{FIXED [Service],[Location] : sum([Number of Records])}

The rows it generates are:

Service 1, Location 1, 2 records

Service 1, Location 2, 1 record

Service 4, Location 3, 2 seconds

Service 4, Location 4, 1 record

and so on.

This next formula generates one record per [Service]. I use this as the denominator.

/{FIXED [Service] : sum([Number of Records])}

The rows it generates are:

Service 1, 3 records

Service 4, 4 records

Service 2, 2 records

service 3, 1 record

So for Service 1/Location 1, the numerator is 2, and for Service 1, the denominator is 3, returning 66%.

The way to understand this calculation - let's break it apart

{ represents a level of detail calculation

[Service] and [location] represent the level of detail (grain) of your result set. I'm asking for one record for each combination of [Service] and [Location]

: denotes the separation between the partitioning dimensions and the aggregate measure

SUM( define your aggregation when manipulating grain of data

[Number of Records] represent the measure I am aggregating

)} close the calculation

[Numerator]/[Denominator] returns a percent.

2/4 = .5

You can then format as percent in Tableau.

The level of detail calculations outlined above set the 'grain' for the numerator and the denominator.