Count and Percentage of On-Time shipments per Distinct Document

KylerD · 2025-06-25T19:06:02+00:00

There was an error rendering this rich post.

Hi, I'm new to Domo and struggling to figure out what I think should be a pretty simple calculation. I am trying to figure out on-time delivery percentages for sales orders that have shipped. I have a transactional dataset with multiple rows of data per order (or document_number) and I'm trying to avoid having to modify the ETL or making a new dataset. I have a beast mode called SO OTD Tag with that tags each line as 'On-Time', 'Late', 'Past-Due', or 'Open' as shown below:

CASE

  WHEN type = 'SalesOrd' THEN

  CASE

    WHEN actualshipdate IS NULL THEN

    CASE

      WHEN requested_shipdate < CURDATE() THEN 'Past-Due'

      WHEN requested_shipdate >= CURDATE() THEN 'Open'

    END

    WHEN actualshipdate IS NOT NULL THEN

    CASE

      WHEN actualshipdate > requested_shipdate THEN 'Late'

      WHEN actualshipdate <= requested_shipdate THEN 'On-Time'

    END

  END

END

Now that I have tagged all sales orders, I want to group by distinct sales order and count whether the sales order is late or not, then divide this by the total number shipped (On-Time SOs + Late SOs). Again, each sales order has multiple lines within the dataset. Currently, I have a draft of OTD % shown in the below code block:

(COUNT(CASE WHEN `SO OTD Tag` = 'On-Time' THEN 1 END))

/

COUNT(`document_number`)

This calculation doesn't necessarily work because it calculates based on the number of lines, not the number of distinct sales orders (or 'document_number').

Any way I can group by documentnumber and assign an OTD tag for each individual document_number and get an OTD %?

Find more posts tagged with

Accepted answers

All comments

MarkSnodgrass

I'm having a hard time visualizing your data so my initial answer may not be what you are looking for. If you have an order with 4 items in the order and 3 are tagged as On-Time based on your initial beast mode, then that should yield a percentage of 75%.

Given that, I would suggest you alter your 2nd beast mode to to be the following:

(SUM(CASE WHEN `SO OTD Tag` = 'On-Time' THEN 1 ELSE 0END))
/
COUNT(`document_number`)

If you use a table chart and then place the document and this 2nd beast mode in your card, I believe you should get that percentage.

ArborRose

It sounds like you want to calculate on-time deliver percent on sales orders (document /order number), not lines. Your beast mode tags each line as on-time, late, or whatever - rather than each order.

You probably may need something with a FIXED()

CASE 

  WHEN MAX(CASE WHEN `SO OTD Tag` = 'Late' THEN 1 ELSE 0 END) OVER (PARTITION BY `document_number`) = 1 THEN 'Late'

  WHEN MIN(CASE WHEN `SO OTD Tag` = 'On-Time' THEN 1 ELSE 0 END) OVER (PARTITION BY `document_number`) = 1 THEN 'On-Time'

  ELSE 'Other'

END



COUNT(DISTINCT CASE WHEN `Order OTD Status` = 'On-Time' THEN `document_number` END)

/

COUNT(DISTINCT CASE WHEN `Order OTD Status` IN ('On-Time', 'Late') THEN `document_number` END)

As Mark mentioned…it would be helpful to see a few rows of anonymized data.