Honestly, I didn't even know about external tools like Bravo. Downloaded immediately. What a time saver. A whole video on your favorite external tools and how you utilize them in action would be great.
I don't know why you make these videos but they are the best out there and they have helped me greatly. And you make them available for free. Thank you for your hard work.
That is true, but my approach would be to not model based on the exceptions. Rather model for the 80%+ of use cases and create alternative solutions for the exceptions.
This video is pure gold! Thank you so much for sharing master. Definitely when someone ask me for optimizations, this is the only resource I'll need to recommend.
Hey Bas, Thanks again for the video, very informative. I just wanted to point out : Step 3: it would be much preferred to push all of the calculations, group by functions to the DataMart in case the user's have one. Step 10: the incremental refresh kind of will help with the import mode since you only import your data once then in the workspace it will create partitions of that data that will be stored and not changed and depending on how you want the data to be refreshed and when you know for example that your data after 2 weeks will not change you can set it up and your data will be stored in the workspace. Thanks again for the video !! PS: when will you do a video about the power bi extension you use and how you use them, it will be great !!
Thanks Bas, excellent tips. I love Tip 3 about switching off the hidden date/time tables being created in the background - this is a new one to me. Keeping in mind cardinality & where data types could be switched to numeric are also great reminders. I will be following the advice given here.
Hi Bas, Excellent video again from you. I have a doubt, at 6:45 when you say, “use a filter measure instead” what do you mean? I didn’t understand this part alone. Thank you for the awesome video again
What do you think about using bidirectional filtering when using a unified star schema in power BI? Would love a video about the combination of the USS and practises in Power BI
Hi Bas, first I'd like to thank you for your content as it has helped me inmensely on my Power BI journey. I was wondering if you had any resources/strategies on how to go from the flat table to a simple star schema as you describe on step 4. Thanks again.
@@HowtoPowerBI it is amazing the amount of things you can accomplish with Power BI (albeit inefficiently) without having a grasp of the basics, much appreciated.
This was great thanks. I still battle to understand some basic things, like when you you create relationships (is it necessary for a type star model?) versus when you you do joins/merges? For example if you do a merge in PQ then that's a join (?), but if you create relationships using the model view and use a field from another table in a visual is that a join too?
Cheers for the video! Do you have any more tips for those who can't use external tools like bravo due to security reasons? Recently started at a large company and need to optimise some reports that are very large due to scope creep, is it just a case of looking at all the visuals and checking to see what columns aren't used etc?
Some more optimizations that I would recommend: 1. Use Binary.Buffer, List.Buffer and Table.Buffer judiciously. They significantly speed up Power Query refreshes. 2. Operate on List compared to Tables in Power Query. List operations are much faster. After transformation on list, the data can easily be transformed into Table using power query function Table.FromList 3. Perform transformation in DAX as much as possible. They are much faster on large tables compared to Power Query 4. Use incremental refresh where older data is not modified. This will speed up data refreshes
@@HowtoPowerBI I think, I should qualify my comment further. Where query folding is possible e.g. data sources that support SQL queries, its probably better to move the transformation to the source of the query. In other cases e.g. REST API requests, it's better to let the power query to construct the tables with the responses and let DAX do the transformations
Power BI Helper seems to have last updated in 2021. I assume that means it won't work with reports saved in PBIP format? Will the other tools you mentioned do so? PBIP has been amazing for version control.
Love your channel.....just off topic a bit, I have a question regarding the forecasting model. In previous version PB displayed the Forecast, Upper bound and Lower bound totals when "show as table". Those totals are not displaying with the latest version. Is this a bug?
I really have question in mind like I see many dashboards which solely have visualization no content to it while others have insights . Which one is the best practice and what is an ideal canvas setting for dashboard size
Do you have any experience with the Aggregated Tables feature and is it a good way to improve semantic model performance? What are your thoughts on an having enterprise level 'Golden Model' - one with multiple dimension and fact tables for cross fact reporting - vs composite modelling?
10 Steps to Optimize Your Data Model are a pragmatic way, before transferring your Data Model from Test to Production environment. A couple of the 10 steps are also explained in the course material for Microsoft Power BI Data Analyst (PL-300) certification. 👨🎓👩🎓
2 more suggestions from my side: Preference for Custom column in Power Query instead of Calculated columns in the model. Also, disable Power Query query load for unwanted tables.
So I understand the concerns that are introduced if you use bidrectional filters as in Step 6, but you didn't explain how to solve them in a more performant manner. Can you expand?
Hi Bas hope you read this. Can you give us tutorial how to get auto refresh visual report in Power Bi service with any data sources (from Excel, Dataflow, Sql Server, etc)
One more reason to prefer calculated (DAX) columns vs custom (Power Query) column for strings concatenation: DAX allows to use emoji 😎😎 while Power Query seems to be limited
Hi I have doubts about the drop down without single select slicer is this possible to when we select any category in the slicer then the slicer pop up does not disappear automatically when we have to click any other place then the slicer pop up disappear. Any solution i want when we click on the category in slicer then the slicer pop up automatically disappears.
I have a question of Power BI, but not related to this video. when I use slicer, I don't want to show data in the line chart when I don't select any option in the slicer, and that need to be the default status when i open the report, how do.I achieve this?
99% of those who build reports with Power BI use the "Click, drag and drop until it works" approach, so all these very valuable hints are just ignored. Still Bas is so good to be able to condense in a few minutes plenty of essential hints.
I have always struggled with step number 1 to remove unused columns and tables because we use golden datasets with many live connection reports created by our team and outside users for their own analysis. Any tips for this?
Use Measure Killer tool, using this you can analyse live connection reports and identify unused measures and columns. This tool will help us to delete these unused objects in dataset automatically. We don’t need to do it manually
MSMGQM (Model, Script, Manage, Query, Generate, Monitor) Power BI External Tool HOW to install this external tool! can you please provide the link for this tool ?
‼️ urgent ‼️ Hi Bas, I would like to know, what are the add on certification should i take in order to learn power bi eco system, Certification such as DP 900, AZ 900.. there is a certification drive going on in company at free of cost… Let me know
Not convinced on Snowflake being worse! Some metrics to back up your argument would be nice. Compression benefits can outweigh the impact of additional joins.
There are no compression benefits with snowflake. Only the distinct values matter in column store. No of rows doesn’t matter much. But joins do matters, and a lot.
Hii sir iam aspiring data analyst, currently I am pursuing btech final year i want to do internship under your guidance . Please help me to do with which was unpaid internship to get real world experience and upgrading my skills.please give me a chance to work as intern.
Honestly, I didn't even know about external tools like Bravo. Downloaded immediately. What a time saver. A whole video on your favorite external tools and how you utilize them in action would be great.
same here - especially it is from Marco Russo who allready published some solid DAX literature.
Yes external tools video would be huge
Requesting same …
I am also requesting to be video on external tools
Can we also see which measures are used or which are not used in report with Bravo tool?
I don't know why you make these videos but they are the best out there and they have helped me greatly. And you make them available for free. Thank you for your hard work.
That is soo nice 😊 thank you 🙏 really appreciate it
Aggregation is good, but there is always someone who would want to drill down to more details! :D
I wanted to give a 'haha' react for this
Yes especially finance teams :)
@@lhoussineelomari5689yes just to export into excel
That is true, but my approach would be to not model based on the exceptions. Rather model for the 80%+ of use cases and create alternative solutions for the exceptions.
This is where knowing the audience and getting solid requirements is huge.
This video is pure gold! Thank you so much for sharing master.
Definitely when someone ask me for optimizations, this is the only resource I'll need to recommend.
Hey Bas,
Thanks again for the video, very informative.
I just wanted to point out :
Step 3: it would be much preferred to push all of the calculations, group by functions to the DataMart in case the user's have one.
Step 10: the incremental refresh kind of will help with the import mode since you only import your data once then in the workspace it will create partitions of that data that will be stored and not changed and depending on how you want the data to be refreshed and when you know for example that your data after 2 weeks will not change you can set it up and your data will be stored in the workspace.
Thanks again for the video !!
PS: when will you do a video about the power bi extension you use and how you use them, it will be great !!
Mate, your videos are great! They are a real deep dive into PBI for more advanced folks.
Thanks Bas, excellent tips. I love Tip 3 about switching off the hidden date/time tables being created in the background - this is a new one to me. Keeping in mind cardinality & where data types could be switched to numeric are also great reminders. I will be following the advice given here.
Great video, next time people ask me for best practices on Data Modelling I can just share this video 👍
That's the best Power Bi UA-cam Channel. Congrats.
Thank you. I really enjoy your tutorials. They are so informative. Can't wait for your next video
Hi Bas,
Excellent video again from you. I have a doubt, at 6:45 when you say, “use a filter measure instead” what do you mean? I didn’t understand this part alone.
Thank you for the awesome video again
What do you think about using bidirectional filtering when using a unified star schema in power BI? Would love a video about the combination of the USS and practises in Power BI
Excelent work, very well condensed in 10 min. Many users do not pay attention to maintenance and optimization and I think that that is a mistake.
Awesome tips!! First time, I am learning about Bravo. Great Video Bas, thank you!!
Hi Bas, first I'd like to thank you for your content as it has helped me inmensely on my Power BI journey. I was wondering if you had any resources/strategies on how to go from the flat table to a simple star schema as you describe on step 4. Thanks again.
here you go 😊 ua-cam.com/video/w4NHK_jizUw/v-deo.html
@@HowtoPowerBI it is amazing the amount of things you can accomplish with Power BI (albeit inefficiently) without having a grasp of the basics, much appreciated.
Always fashinated on a way you explain and share very helpful informations.
Thanks alot... BRAVO..!!!
This is so informative ! thanks a million!
This was great thanks. I still battle to understand some basic things, like when you you create relationships (is it necessary for a type star model?) versus when you you do joins/merges? For example if you do a merge in PQ then that's a join (?), but if you create relationships using the model view and use a field from another table in a visual is that a join too?
Awesome, simply explained and on point. Big thanks 👍🏻
Great instructive video man, thanks!
Thanks for watching!!! 😀
great video, thanks for sharing all these valuable tips!
Cheers for the video! Do you have any more tips for those who can't use external tools like bravo due to security reasons? Recently started at a large company and need to optimise some reports that are very large due to scope creep, is it just a case of looking at all the visuals and checking to see what columns aren't used etc?
Some more optimizations that I would recommend:
1. Use Binary.Buffer, List.Buffer and Table.Buffer judiciously. They significantly speed up Power Query refreshes.
2. Operate on List compared to Tables in Power Query. List operations are much faster. After transformation on list, the data can easily be transformed into Table using power query function Table.FromList
3. Perform transformation in DAX as much as possible. They are much faster on large tables compared to Power Query
4. Use incremental refresh where older data is not modified. This will speed up data refreshes
thx! you got a controversial standpoint with nr 3 there 😁
@@HowtoPowerBI I think, I should qualify my comment further. Where query folding is possible e.g. data sources that support SQL queries, its probably better to move the transformation to the source of the query. In other cases e.g. REST API requests, it's better to let the power query to construct the tables with the responses and let DAX do the transformations
thank you so much for all these information
Power BI Helper seems to have last updated in 2021. I assume that means it won't work with reports saved in PBIP format? Will the other tools you mentioned do so?
PBIP has been amazing for version control.
Is the measure in the filter pane at 6:45 in case 6 being meant to be used as a better alternative compared to the bidirectional relationship?
also curious on this!
I’ve learned sooo much with this video man!! Thanks!!!
Wow, what a great video. It helps a lot.
Looking forward to see similar video like this
Solid and very well-explained video, congrats!
Love your channel.....just off topic a bit, I have a question regarding the forecasting model. In previous version PB displayed the Forecast, Upper bound and Lower bound totals when "show as table". Those totals are not displaying with the latest version. Is this a bug?
About the last part... Which is better to use as source, excel or access? I'm having issues to auto update access from power bi online
Very informative video. Thank you for creating this one.
What a great video! Thank you for this calrity
I really have question in mind like I see many dashboards which solely have visualization no content to it while others have insights .
Which one is the best practice and what is an ideal canvas setting for dashboard size
Is it possible to pop up the details about the cell by clicking each data cell in the table in PowerBI?
Do you have any experience with the Aggregated Tables feature and is it a good way to improve semantic model performance?
What are your thoughts on an having enterprise level 'Golden Model' - one with multiple dimension and fact tables for cross fact reporting - vs composite modelling?
What are the best power bi books you would recommend. All around and also dax.
Thank you guys in advance
the top recommendations on amzn r all good
10 Steps to Optimize Your Data Model are a pragmatic way,
before transferring your Data Model from Test to Production environment.
A couple of the 10 steps are also explained in the course material for Microsoft Power BI Data Analyst (PL-300) certification. 👨🎓👩🎓
Great video!! What do you use for creating scenes and record your screen - like in 06:35 ? OBS?
Thanks! Very informative
really good vdo. thx
2 more suggestions from my side:
Preference for Custom column in Power Query instead of Calculated columns in the model.
Also, disable Power Query query load for unwanted tables.
So I understand the concerns that are introduced if you use bidrectional filters as in Step 6, but you didn't explain how to solve them in a more performant manner. Can you expand?
You add filter to the slicer directly that says if other filter is blank then don’t show . Model is the wrong level to apply it
Could you please make a DAX best practices ?
Hi Bas hope you read this.
Can you give us tutorial how to get auto refresh visual report in Power Bi service with any data sources (from Excel, Dataflow, Sql Server, etc)
One more reason to prefer calculated (DAX) columns vs custom (Power Query) column for strings concatenation: DAX allows to use emoji 😎😎 while Power Query seems to be limited
Hi I have doubts about the drop down without single select slicer is this possible to
when we select any category in the slicer then the slicer pop up does not disappear automatically when we have to click any other place then the slicer pop up disappear. Any solution i want when we click on the category in slicer then the slicer pop up automatically disappears.
This is Gold!!
I have a question of Power BI, but not related to this video. when I use slicer, I don't want to show data in the line chart when I don't select any option in the slicer, and that need to be the default status when i open the report, how do.I achieve this?
also the slicer should control how many lines displayed in the chart
Great content.
files for this session not available on the download page
99% of those who build reports with Power BI use the "Click, drag and drop until it works" approach, so all these very valuable hints are just ignored. Still Bas is so good to be able to condense in a few minutes plenty of essential hints.
I have always struggled with step number 1 to remove unused columns and tables because we use golden datasets with many live connection reports created by our team and outside users for their own analysis. Any tips for this?
Start small, then add columns on demand. As of now, there is no tool to analyze what columns are used in LC by self service reports.
Use Measure Killer tool, using this you can analyse live connection reports and identify unused measures and columns. This tool will help us to delete these unused objects in dataset automatically. We don’t need to do it manually
What do we consider a "big data model"
If you have any column as Date/time and you don't need the time format them as Date only, it reduces the model size a lot!
very concise, nice
MSMGQM (Model, Script, Manage, Query, Generate, Monitor) Power BI External Tool HOW to install this external tool! can you please provide the link for this tool ?
‼️ urgent ‼️
Hi Bas,
I would like to know, what are the add on certification should i take in order to learn power bi eco system,
Certification such as
DP 900, AZ 900..
there is a certification drive going on in company at free of cost…
Let me know
Dp 600 pl 300
impressive
Not convinced on Snowflake being worse! Some metrics to back up your argument would be nice. Compression benefits can outweigh the impact of additional joins.
There are no compression benefits with snowflake. Only the distinct values matter in column store. No of rows doesn’t matter much. But joins do matters, and a lot.
My workplace will not allow an external tool😊
Sir Please make Dashboard as example using this concept. being new difficult to understand
this is a good starting point for beginners ua-cam.com/video/w4NHK_jizUw/v-deo.html
not big fan of calculated column though
As always. Great video with some awesome tips. I have learnt so much from your videos.
Hii sir iam aspiring data analyst, currently I am pursuing btech final year i want to do internship under your guidance . Please help me to do with which was unpaid internship to get real world experience and upgrading my skills.please give me a chance to work as intern.
This is really confusing, please explian to just remove this short.
😂