Billobi Contributor

I’d do some additional ETL before doing this visualization. There are two issues that would make date-based beast gymnastics in the card a pain: * Open.Date and Close.Date are likely two different columns in your dataset (assuming each row represents a Case). The x-axis can only look at one column worth of dates, which we…

There's not an option out there to have a mirrored dataset, but you can accomplish nearly the same thing with a transform-less ETL. Just setup magic ETL with an Input Dataset and an Output Dataset. Set the ETL to run anytime the Input Dataset is updated. You can then use your Output Dataset as if it's a copy of the…

rado - I like your simple ETL there, it does the trick. If the regular input balloons in size over time, the transform's run time could eventually degrade. Nice to avoid the potential for duplicates there if possible. I attached another simplified ETL, just like yours but without the remove duplicates. This can be used in…

Recursive ETL is a bit funky when you first do it. Agreed with the above comment to checkout Recursive ETL in the KB, that will get you most of the way there. Two areas to really keep an eye out: Priming the pump. When you run it the first time through, use a naming convention for the output that isn't going to confuse you…

I'm not a big fan of hard coding in static logic to set attributes on given date ranges, as it inserts overhead to maintain and you have to continually apply it forward over time. That said, I have had to do it before. Doing it in ETL is the most cumbersome way to apply attributes to classify custom date ranges. You could…

Hi Manasa, If your QB datasets were refreshing OK weekly in the past, but you're now seeing that error and are unable to create a new QB dataset, then it sounds like something on your account on the QB side may have changed. I'd sit down with one of your QB Admins and have them show you the QB user profile settings that…

Sounds like a tough spot, since the color options are usually geared toward just that particular card, or all cards across the dataset. In your case, that's either too little or way too much. Depending on how many colors or cards you're talking about, I'd consider doing a beast for the dimension/measure that you want…

You're definitely on the right path. When I mock up the \?(.*) command, it snips off the ? character and anything to the right of it from the URL, which I think is your desired outcome. This sort of ETL won't collapse any of your rows or give you counts of the various URL names, you'd need to do any grouping or counting…

When pulling a custom lookup field (often used for users to force an input selection) that is returning a UID, you'll want to try out the __r suffix. Check out this - https://developer.salesforce.com/forums/?id=906F00000008llXIAQ So if Account.country is a custom lookup, you'll need to put the field into your query as…