GrantSmith Coach

It sounds like you want a custom app that you can do and have control over the different sliders on your app to then trigger an event to call your ML model API. You can't call an API from directly within a card. You can read up more on custom apps here:

What do you mean Admin access? Workbench is a windows application so anyone who has admin access to the machine has admin access to Workbench. If you're talking about authentication, no, only one Domo user can be authenticated with workbench and Domo

What issues are you running into with the JSON connectors?

CASE WHEN MAX(`Attendance Type`) OVER (PARTITION BY `Event UNID`) = 'Video' THEN 'Yes' ELSE 'No' END You can use a window function to return a value across an entire bucket (in your case the event). MAX will return Video if a video record exists (it comes alphabetically after Phone). Wrapping in CASE statement to return…

CONCAT(SUM(CASE WHEN `` = 'Forward to HM' THEN 1 END), ':', SUM(CASE WHEN `` = 'Hire' THEN 1 END)) You can use a CASE statement to conditionally SUM your records together and then use a CONCAT to put everything together in your ratio format.

Nope, this should be just fine.

You can utilize LAG functions to do a rolling average. I've done a writeup on this here:

If you don't have a column already named the same as your beast mode or another beast mode named the same I'd recommend logging a ticket with support. I've occasionally seen this happen and needs Domo Support assistance.

If you're pulling in Daily data and don't need to keep more than 3 years of data partitioning would be easiest for you to utilize.

The append endpoint doesn’t exist hence the 404 try using data?updateMethod=APPEND

404 means you’re trying to access something that doesn’t exist. Does your UTL correctly include the proper dataset id?

you can use AND to combine logcal expressions together if both should be true or OR if one needs to be true. Alternatively you can add an additional WHEN clause to your case statement to do something diffferent: CASE WHEN `primary class` = 'supermarkets and grocery stores' THEN SUM(CASE WHEN `Split answer` = 'Female' THEN…

You'll want to log a ticket with Domo Support as they own the connector and is likely something in their code which needs to be updated.

CASE WHEN DATEDIFF(CURDATE(), `Date Field`) <= `Days Variable` THEN 'Include' ELSE 'Exclude' END

DATE will work. There’s also DATE_TRUNC

You can surround your beast mode in a case statement and check if the date is before today then return the date. This will return null for future dates Put the year in the series on your graph so it’ll be separate lines for each year.

I’d recommend using a variable to allow users to input the number of days and use that in a beast mode to filter the card

I’d recommend logging a ticket with Domo support as they own the connector and can help diagnose your issue in greater detail

Currently, variables are only single-select. I'd recommend adding an idea to the idea exchange to get variables to allow for multi-select.

Do the string values look the same when you compare them? Have you done a TRIM on the IDs to remove any trailing spaces? Are they different lengths between the two databases?

This is great @AndreaLovesData!

The dataset alias is the name that's predefined in the DDX under the dropdown. Typically it'll be something like 'dataset0', 'dataset1' etc.

You can add your group names to the partition list PARTITION BY `Year`, `Other Group`

You can get the total for each year with a window function: SUM(SUM(`Duration`)) OVER (PARTITION BY `Year`) You can get a running total with another window function: SUM(SUM(`Hours to Complete`)) OVER (PARTITION BY `Year` ORDER BY `Month`) NOTE: Your Month needs to be an actual date or a numerical representation to be…

You can do this within a Magic ETL. If you do something like this: This will do a cartesian join between the list of years and your dataset and then filter the records where the year is in between your original start and end years. Add formula is adding a new column called Join Column with a value of 1 for the join. I've…

CASE WHEN `field` = '[]' THEN 0 ELSE (LENGTH(`field`) - LENGTH(REPLACE(`field`, ',', ''))) + 1 Some string manipulation options to assist. If it's an empty array list, return 0, otherwise determine the length of the string and subtract the length of the string with the commas removed (count the number of commas). Add one…

You can do a running total with a beast mode with a conditional add instead of using the running total on the card. CASE WHEN `Date` < CURRENT_DATE() THEN SUM(SUM(`Sales`)) OVER (PARTITION BY YEAR(`Date`) ORDER BY `Date`) END This will exclude any future dates today and beyond so you only have a line for the months…

It is, you can use a FIXED function in a beast mode to do this. Here's some documentation on it: - See Use Case #5

What is the formula you're using? You could do something like: CASE WHEN `field` IS NULL THEN 1 ELSE 0 END Then aggregate your dataset and SUM that field to get a count of the records.

This isn't an option currently, but I'd recommend this idea to the Idea Exchange.