sum value from field from multiple rows

meadow_ryan

If I have a field 'amount' and am wanting to create sum the values from that field if there are multiple rows of the same order how cold I go about doing this.

 

So for example I have an order number of 0703915 and there are multiple rows with that order number but one row has an amount of -2000 and the other row has an amount of 2000, how can I sum this together, then only display one row for that order?

PodiumMason

Hey Ryan!

 

The cardbuilder is going to be smart enough to combine rows with the same order number when you to a SUM aggregation on the value column. You shouldn't have to do anything else in the graph to get the rows rolled up in your visualzation, whether it's a table card or a bar graph.

 

To test this, create a table card and put the order number and value on the table. Then make sure the Value is aggregated to SUM.  Once this is done, you should see one row per Order. 

 

Let me know if this works or if I can provide additional detail!

meadow_ryan

I set the 'amount' field to a sum aggregation but that didn't seem to have any effect I still have my multiple rows of orders with the different values....I'm wanting to sum those amounts together per order

PodiumMason

Ryan,

 

The aggregation should have taken care of it. (See Screenshots). It's possible if the Order Number column has some hidden whitespace in one order number row but not the other, that could be causing the aggregation to not work. 

 

Also if you have any other columns on the table (Such as two different product names) the graph will not roll up the rows into one because it's still grouping by unique product. The above solution will only work for order numbers that are the same and a value column. Any other unique columns added will not work. 

 

I did a test, take a look at the order number and totals from before and after the SUM aggregation:

 

Before:

 

 

After:

 

 

meadow_ryan

we do have other columns for these rows where their data is not the same for instance different dates, different invoice numbers

PodiumMason

So in order to roll the rows up into a single row for each order, each column that you have must contain the same data. Any differences in any of the columns and the data will not combine into single rows. 

 

This is because those rows of data are basally different rows of data. 

 

Let me know if you have any additional questions!

meadow_ryan

so for instance in my screenshot above each row must have the same blnum, etc.

PodiumMason

Yes, exactly. That or you'd have to remove any columns that aren't alike.