Have you ever received the error in Tableau? This can be kind of frustrating if you need to normalize any of your values for a dashboard.

What exactly is this error? It occurs when you try to mix an aggregated argument such as SUM([Sales]) with a non-aggregated function such as Discount or Category.

You have to make them **all aggregates** or make them **non-aggregates / row level variables**.

So you can't have **[Profit] / SUM ([Sales]),**

You will need to use **SUM ([Profit]) / SUM ([Sales]) **instead.

Or

You can't have **[Profit] / SUM ([Sales])**

But you can have **[Profit] / [Sales].**

Or

You can't use **[Profit] / SUM ([Sales])**

You have to use **[Profit] / { fixed: SUM ([Sales]) } **

The examples above were taken from this Tableau article __here__. If you prefer a YouTube video, Zen Master Chris Love has a video __here__. They both review the Superstore Data set and it is very easy to follow along.

I wanted to use my own use case in this blog article.

In the example below I used __this data set__ from __Data World __capturing the number of **Deaths** in **Game of Thrones**. I added a Rowcount of 1 for every row and I added the following calculation to my Excel sheet with Alteryx to condense the number of Houses.

**Revised House Name**

If Contains([Allegiance],"Free Folk") then "Free Folk"

ELSEIF Contains([Allegiance],"House Baratheon") then "House Baratheon"

ELSEIF Contains([Allegiance],"House Bolton") then "House Bolton"

ELSEIF Contains([Allegiance],"House Frey") then "House Frey"

ELSEIF Contains([Allegiance],"House Greyjoy") then "House Greyjoy"

ELSEIF Contains([Allegiance],"House Lannister") then "House Lannister"

ELSEIF Contains([Allegiance],"House Stark") then "House Stark"

ELSEIF Contains([Allegiance],"House Targaryen") then "House Targaryen"

ELSEIF Contains([Allegiance],"Night's Watch") then "Night's Watch"

ELSEIF Contains([Allegiance],"None") then "None"

ELSEIF Contains([Allegiance],"House Arryn") then "Other Houses"

ELSEIF Contains([Allegiance],"House Baelish") then "Other Houses"

ELSEIF Contains([Allegiance],"House Clegane") then "Other Houses"

ELSEIF Contains([Allegiance],"House Hollard") then "Other Houses"

ELSEIF Contains([Allegiance],"House Martell") then "Other Houses"

ELSEIF Contains([Allegiance],"House Moore") then "Other Houses"

ELSEIF Contains([Allegiance],"House Mormont") then "Other Houses"

ELSEIF Contains([Allegiance],"House Tully") then "Other Houses"

ELSEIF Contains([Allegiance],"House Tyrell") then "Other Houses"

ELSEIF Contains([Allegiance],"House Umber") then "Other Houses"

ELSE "Other Allegiances"

I used the below calculated fields to capture the counts of deaths in the seasons;

**House **

If Contains([Allegiance],"House") then 1

Else 0

END

**Not House**

If Not Contains([Allegiance],"House") then 1

Else 0

END

With these calculations I created the below summary to check my calculations.

I first tried to get a unique House count by using a Fixed calculation.

I noticed this calculation wouldn't work. It was displaying a total count of 2245 which is the total number of rows.

I also wanted to create a Normalized calculation in order to create Map layers.

I then revised my calculation to have an **If Then** statement inside a **SUM() **formula.

This calculation worked for my data but it would error for my Normalized calculation. I then had to change the Normalized calculation by duplicating the Count format above and replacing the SUM with a MAX formula.

**House Frey MAX**

MAX(IF [Revised House Name] ="House Frey" Then [House] END)

I then changed my normalized calculation to the formula below;

The dashboard I am working on for this blog is located __here__. I am still updating it at this time to add additional map layers.

## Comments