Best Of

essentially it should work like a recursive dataset. By using the output dataset as an input, you bring the data from the last update into the ETL before it's changed. So you take the commentary and join it to the data instead of your webform. Use a Select Columns or Alter Columns before the join to only include the columns you need to make the join and the commentary.

No problem! The REPLACE() function would work inside of either a Magic ETL or in a Beast Mode inside of a card. The function will strip out the pipes "|" and replace with SPACE and strip out the periods "." and not replace those with anything.

If you want to go the ETL route you'd use an Add Formula tile, name it the same name as your Company Name column and paste the two replaces inside where the formula goes.

The process for a beast mode in a card is pretty similar, inside of Analyzer you would click "Create a Calculated Field" and create the Beast Mode. One thing to note however is that inside the card you would have to give it a different name because in this case you're creating a new field rather than overwriting an existing one as you would have in the ETL

Hi all,

I have a Magic ETL flow going currently where I have three input datasets that I am appending into one dataset. All three datasets have the similar fields and columns, but one of the datasets I need to manually SUM three individual columns across each row to get the "Total" count that exists in the other two datasets as just a single column/field.

• Input Dataset 1: Name |Spend | Inside Leads| Outside Leads | Phone Leads
• Input Dataset 2: Name | Spend |Total Leads
• Input Dataset 3: Name | Cost |Total Leads

• Input Dataset 1 Flow: Input Source -> Select Columns -> Group By -> Add Constant -> Append -> Output Dataset
• Input Dataset 2 Flow: Input Source -> Select Columns -> Add Constant -> Append -> Output Dataset
• Input Dataset 2 Flow: Input Source -> Select Columns -> Add Constant -> Append -> Output Dataset

Output Dataset: Name | Cost | Total Leads

Step 3 "Group by" in my Input Dataset 1 Flow below seems to be where my issue is. If I pull the three individual integer columns into the output dataset, I can aggregate them together just fine with the Step 3 formula below through Beast Mode into a table/chart, but ideally I want the dataset to be fully aligned and pre-aggregated so I dont have an output like the one below:

Output Dataset (do not want): Name | Cost | Total Leads | Phone | Inside | Outside

Any insight is appreciated! Thanks!