Blend Power BI datasets with Excel files? INSANE AMAZING! 

Thanks for the comment! So many people have been waiting for this! 👊

It was definitely magical!

Woot! 👊

haha love it 👊

Nice! I am interested in how that pans out.

Appreciate the kind words! 👊

Very much agree 👊

Merry Christmas!

Thank you so muich 👊

Love it! 👊

Same problem here. I double checked my powerbi version and I already have the latest version of powerbi desktop. But I still can't add excel data after loading a powerbi dataset

I have the same problem. Any solution to this?

@@ivanmezev any updates? I have the same issue as well

@@lk2481 My PowerBi Dataset is coming from SQL Server Analysis Services and this is why not working. It seems it works for all other types but for Analysis Services.

Appreciate the feedback and enthusiasm Chayakiran! I've heard that feedback before. I know it's not the same thing, but you can still hide tables from report view. That at least helps the field pane not as crowded.

YES! Excited to see what folks do with this! 👊

Indeed!

Correct. It is a preview feature that needs to be turned on in the December 2020 release.

same problem! where is this feature to enable?! thanks!

@@raulvera705 Go to File --> Options and Settings --> Options and a window will open. On the left hand side look for and click on "Preview Features".

@@DPFierce Thanks a lot! :-)

You saved my life. Thank you

Agreed! 👊

That is correct. You are just referencing the other model. There are some operations you can do like adding another column after it. But doing append/merge operations won't work.

He did actually, ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-y4S-k34qmnw.html

He explained it long back

Yeah probably need to revisit that based on this feature. I tried addressing it in this video - I specifically brought it up because it is confusing. You are not alone Matthias :)

Awesome video as usual but at 3:45 are the magic words to the effect of "direct query is required" followed a few seconds later by "you could have done an import." Confusing as these statements are opposites. Which one is true?

@@GuyInACube Hey Adam, when/if you do an update, can you also talk about "data sovereignty" and how it relates to the different connection methods? I did some research, and I feel I have a good handle on it, but I'm still unsure whether any local/on-premises data is ever stored "somewhere in the cloud" with a live or DQ connection after a report has been published. A client of ours is very strict with how their data is accessed and handled, and I'm unsure whether Power BI Report Server is the only solution for them (which is missing some key PBI features), but want to know for certain before going ahead with a recommendation. Please and thank you!

Dataflows are a completely different thing. From a data model perspective, dataflows is just a data source. Same as Excel. There is still a place for dataflows

haha appreciate that Steve 👊

I don't know that this will work with Power BI Report Server. Will need to stay tuned on that. In general Preview features aren't supported in Power BI Report Server, so maybe we'll hear more when it goes GA.

Dataflows are a completely different thing. Someone else asked me the same thing. In the context of a company calendar. They asked it it would just be better to have a dataset with just the calendar, or have it in a dataflow and reference it. As with all things, it depends, but in this example, i'd go with a dataflow instead. Having it as a dataset and referencing that with Composite models adds unneeded complexity to something simple. It's always about picking the right tool for the job. Just because this is the new shiny doesn't mean it's the right shiny to use. Dataflows still has it's place.

No, that's not possible.

Scheduled refresh limits didn't change with this feature. Pro is 8 times a day and Premium / PPU is 48 times a day.

Love it! Thanks Ankur!

You might be right!! I believe this is still not possible, I think we are still not close to the day where we can connect to the multiple Cubes in a PBIX :(

SAP HANA isn't supported. Not sure there are plans to do that. Appreciate the feedback though.

got it. Need to enable "DirectQuery for Power BI datasets and Analysis Services" in Preview features (under Options and Settings)

I tried it and right now, it didn't work. I read on the blog that i'll work with the next major SQL Server Analysis Service release. I wonder when that will happen!?

Yeah not supported right now :(

Hi, I got the same problem. I haven't solved it yet. Have you solved it? Anyone?

@@nadsonlucio me too any updates

It's currently in preview, so head over to Options and check the box "Direct Query for PBI Datasets and AS" under "Preview features".

I don't know if there is a limit on how many you can pull in. Although, I would be careful with doing that. I can only imagine the performance issues you would hit when doing something like that having them all joined together. The chaining limitation is about nested datasets. So, if I created Dataset B that brings in Dataset A. Then I created Dataset C that brings in dataset B. I would not be able to go another layer deeper and create Dataset D that brings in Dataset C.

@@GuyInACube Got it. Appreciate the response.

Hi I am having same issue, I am using latest version of Power BI. I just can't see the way to connect second datasets once I am connected to one. :)

Hi Gautham, Yes, I had same issue, but now my data sources not grayed out. i am able to use this functionality. using September 2021 edition of Power BI desktop. once you enable preview feature, then it requires a restart of Power BI desktop. looks like previous version didn't give this restart warning, I don't remember. give a try.

@@RKTECH1021 thank you! Yes, it did work now!!

Yeah.. it created a new dataset for me. Even if I just add a single calculated column to the Enterprise data model

Yup. When you use this, it is creating a local dataset with references to the other models in "DirectQuery" mode. Think of it as a wrapping model. So, you will see a dataset that represents that object.

It's currently in preview, so head over to Options and check the box "Direct Query for PBI Datasets and AS" under "Preview features".

@@adamcanfield9682 thx for reaching out. It is toggled on already.

@@adamcanfield9682 Thank you for this. Mine was greyed out as well and this solved it.

The doc calls out a limitation with RLS docs.microsoft.com/en-us/power-bi/connect-data/desktop-directquery-datasets-azure-analysis-services#considerations-and-limitations RLS rules will be applied on the source on which they are defined, but will not be applied to any other datasets in the model. RLS defined in the report will not be applied to remote sources, and RLS set on remote sources will not be applied to other data sources. I'm not sure why your visuals are broken specifically. If you have created relationships, I'd recommend avoiding the relationships and do the work from a DAX measure instead with TreatAs or something similar and see if that helps.

​@@GuyInACube thanks so much for a reply. It is really strange. to test I created a report with 1 very basic visual. it is a table with two columns. "Date" from date table and a measure "Sum(Sales_Account[Price])" from my fact table. I tried to control the variables. 1) I have RLS applied to the model and publish to the service. I start another power bi desktop model and connect to the dataset like you would prior to the new direct query feature. Then I publish to web and everything works great and all users can see the data they are suppose to. I go back to the desktop and click "make changes to model". once it makes a local model I publish the exact same way to the service and then to the app (just like I did prior to clicking "Make changes to model"). The visual breaks for all users. The only people it works for are admins of the workspace? Thoughts?

@@GuyInACube After further investigation I found out this. In "Manage Permissions" under the source power bi data set they have to have "BUILD" permissions to see the data in the other report that is using direct query. Is that how it should be? I don't want my end users to have "BUILD" permission. I only want them to have "APP" permission which from my understand is the same as "READ". Are you experiencing this?

Not much to go on. Not sure why you would be hitting that error. Check the limitations called out in the announcement blog to see if maybe you are hitting something there.

hmmm may be something specific with the model you are trying to work with? Does it happen with every dataset you try to do this with, or just the one?

@@GuyInACube thanks so much for the response. Yes it seems to happen with any dataset that is decently big (341,000 KB)

@@noahhadro8213 Happens with mine too

Just to add onto this. As GIAC suggests, I used another dataset and it worked!

I am also having this same error. I am using large models, with a combination of Dual and Imported tables. The data source is Snowflake. I did edit some tables in Tabular Editor which may have corrupt the model for this function. File sizes of around 200 MB. What are your specs? It would be great if someone is able to pinpoint what causes the issue!

It's currently in preview, so head over to Options and check the box "Direct Query for PBI Datasets and AS" under "Preview features".