happy friday🙂 i have a following challenge, maybe you could help me to resolve it. my goal is a table like on a spreadsheet screenshot below. i'm trying to create a metric that will give me the sum of reward value by saas state, brand and saas date which corresponds to a reward creation date. also attaching the data model - i want to get from saas_state_date to revard value by created_at date via brand_key but eventually all the saas stuff can only reach brand & metrics that refer to brand directly in BY statement. what would you advise here? thank you much!

Hi Masha, happy Friday to you too! 🙂 So you are looking to go “against the grain” on the LDM connections in order to slice the

Reward Value

fact by the

SaaS State

attribute, correct? This can be achieved via Explicit Lifting, utilizing the BY keyword (as you noted). Based on one of the examples from the article (which is for our Classic UI, but is still functional), you’d be looking for a metric a bit like this:

SELECT MAX(( SELECT (( SELECT SUM(Reward Value) BY Brand Key)) BY SaaS State Key))

The article explains in more detail what each part of this metric does, but in short it will allow you to slice Reward Value by

SaaS State

and

SaaS Date

. I’m not sure about comparing the

Created at

date dimension with

SaaS Date

in this example. As I mentioned, you should be able to use the latter to slice this report, but I’d recommend experimenting with the syntax to see if the comparison will work out on your LDM. Let me know if you have any more questions or additional details about the use-case to help us come up with some solution!

thanks much Francisco, trying it now

Hi Masha, I’m sorry to hear that didn’t work out. In this case, it wouldn’t be possible to both reach out to the

Dim Saas States

dimension and still be able to slice the metric by

Dim Display Currency

. It is hard to assist with the implementation of LDMs, because they are each so unique to a customer’s data and needs (and it’s also a bit beyond our scope); That said, something that could help you out here would be to invert the connection point between the

Saas States

and

Brand

datasets. This would require adding a reference key from first into the latter, and would also change how they interact: You would be able to slice the data from

Brand

with the

SaaS States

objects. I’m not sure how that would affect your overall use-case, but it would allow for slicing the facts from

Fact Rewards

with these datasets without needing any fancy

BY

metrics. Lastly, the

((

double brackets don’t actually do anything 😅 They were described like that in our classic docs, but they are not necessary. I tested the metric without them and it worked nicely.