Get certified for free when you join Fabric Data Days 2026 and dive into Fabric, Power BI, SQL, AI, and other essential data skills.
Join nowData Days is here! Join us now for 60+ days of learning, challenges, and connection. Learn more
Hello everyone,
What should be preferred when PBI Datasets are used? Live Connection or DirectQuery?
And how DirectQuery will be better than Live Connection for PBI Datasets? One reason I can think of is we can create new relationships and create calculated columns. Please let me know if there are any other reasons.
Will there be any discrepancies in the performance of the two? Please guide.
Discussion on these will be great.
Thanks.
Solved! Go to Solution.
Hi. It looks like a question to give an opinion so here is mine. In my opinion Live connections is the best option. That connection is faster than the other one and will help a company maintain data integrity. The direct query option I can only see it worthy for a very special situation that need a composite model to incorporate one table for an specific requirement. Otherwise if it's a needed table for more reports it should be added in the first dataset to keep all connections lives and avoid data silos that will be generate with people adding tables in their direct queries.
Regards,
Happy to help!
Hi @rajulshah
DirectQuery
Advantages
Disadvantages
Live Connection
Advantages
Disadvantages
To learn more details,you could read the following article!
DirectQuery, Live Connection or Import Data? Tough Decision!
Wish it is helpful for you!
Best Regards
Lucien
Hi @rajulshah
DirectQuery
Advantages
Disadvantages
Live Connection
Advantages
Disadvantages
To learn more details,you could read the following article!
DirectQuery, Live Connection or Import Data? Tough Decision!
Wish it is helpful for you!
Best Regards
Lucien
Hi. It looks like a question to give an opinion so here is mine. In my opinion Live connections is the best option. That connection is faster than the other one and will help a company maintain data integrity. The direct query option I can only see it worthy for a very special situation that need a composite model to incorporate one table for an specific requirement. Otherwise if it's a needed table for more reports it should be added in the first dataset to keep all connections lives and avoid data silos that will be generate with people adding tables in their direct queries.
Regards,
Happy to help!
Don't miss out on Data Days, June 15 through August 7. Learn Fabric, Power BI, SQL, AI and more.
Check out the May 2026 Power BI update to learn about new features.
| User | Count |
|---|---|
| 16 | |
| 10 | |
| 10 | |
| 8 | |
| 7 |
| User | Count |
|---|---|
| 40 | |
| 36 | |
| 34 | |
| 28 | |
| 20 |