Tableau Blend vs Join vs Relationship (Data Model) Explained - Part 2 | sqlbelle
Ғылым және технология
Part 2 Tableau Join vs Relationship - In this second part, we discuss the Tableau data model and focus on the differences between join and relationship, and highlight advantages and limitations of each. This also demos typical issues and errors we encounter with joins, like duplicated records or missing rows. At the end of the video, there is a comprehensive summary of blend, join and relationship.
--
This is a two-part series where we explain and demo the differences between Tableau's blend vs join vs relationship - which are all ways of combining multiple files or data sources together.
We will describe each one, walk through demonstrations, and identify distinctive characteristics and potential limitations and/or drawbacks.
Part 1 focuses on Tableau Blend. We will look at how to create blends, underlying way it works, some quirks, and main reasons for using blend.
Part 2 focuses on Joins vs Blends. It also discusses the new Data Model introduced in v2020.2. A comprehensive summary is provided towards the end of the Part 2 video.
Related videos:
Blend vs Join vs Relationship (Data Model) - Comprehensive Comparison - Part 1
• Tableau Blend vs Join ...
Blend vs Join vs Relationship (Data Model) - Comprehensive Comparison - Part 2
• Tableau Blend vs Join ...
Database Joins Explained
• SQL Joins Explained wi...
Level of Detail (LOD) Expressions
• Tableau LOD (Level of ...
Data Interpreter Tricks
• Clean data in Tableau ...
Пікірлер: 227
Thank you for the detailed explanation. The information given are very clear and easy to understand and remember. You narrate it like a story. Awesome skill. I have watched all your videos. Do upload videos on advance concepts as well. Thanks again..
@sqlbelle
3 жыл бұрын
Thank you so much for your comment Usha! I am happy to hear you found it useful! 🙂
@mychaelvo7597
10 ай бұрын
Totally agree. Awesome skill!!!!!!!!!!!!!!!!!!!!!!!!
21:55 is pure genious! "Use relationships if you can, join if you must, blend if you absolutely must."
@sqlbelle
3 жыл бұрын
Thanks Franco! Lots of new features and improvements allow us to work around the blends. And given it's limitations, I find it's best to leave it as last option 🙂
@pradeepp9126
3 жыл бұрын
@@sqlbelle Seriously, You have a special gift for narrating things crisp and clear, I don't feel it like a tutorial it's like a journey us through all depths and heights, keep rocking!!! Thank god I found your channel :-P
I hvnt seen anyone in my life explaining things so vividly and in such a easy way.. Hatss off sqlbelle. Keep posting👍🏻👍🏻
@sqlbelle
3 жыл бұрын
This is such a kind note, thank you Vishesh!
As a recent convert from Power BI, the video segment on relationships was invaluable! I finally had an "Oh, I get it!" moment. My mind started seeing what I had been familiar with in Power BI, but in Tableau language. Thank you so much! Love the videos! I might have to get your advice at the end framed and hung in my office. :)
@4evaism
Жыл бұрын
I share the sentiment. This is exactly the Aha moment of tableau
Wow, what a superb explanation. This is just what i was looking for. When i started with tableau, I always happen to find duplicates when i have different data sources, so my always go to action was to fix duplicates at database level, create a view / table and use it as a single table , for this i need data engineers to move all my data sources into one database( could be Redshift or snowflake) which took a lot of time, now by using relationships , i can use them on the go without any issues. Secondly, i can answer different business questions by changing my primary and secondary table in the data viz itself, No need to go back to data source.
@sqlbelle
3 жыл бұрын
Thank you Aditya, I am happy to hear it was helpful. I think relationships are a game changer, the same way LODs were a game changer when it was first introduced. I am excited to see all the new features and improvements that are planned for the product for the coming year. Happy New Year! 🙂 Donabel
Best tutorials I ever watched, the way you present everything a non-tech person will also understand!! thank you very much for bringing this knowledge hub to us.
@sqlbelle
3 жыл бұрын
Thank you Mahesh for the kind comment. I am happy to hear you found this video helpful!
Thank you- I was much confuse about this topic and was browsed # of video in youtube but still was not able to understand, But this time with your way of explaning i can say i completely grabs topic very clearly thank you so much please keep updating further. #LOVE FROM BANGALORE, INDIA#
@sqlbelle
Жыл бұрын
Thank you so much Wasim!
The More , I am watching your video the more I deep dive in tableau expertise. What a champ you are . Love your content , Love your voice and love your explanation . God bless you Madam.
@sqlbelle
3 жыл бұрын
Thank you for your very kind comment, I appreciate it!
Pretty well defined... Only after this video I'm clear with this topic.. Thank you🌹❤
@sqlbelle
Жыл бұрын
Thank you @midunsathiyamoorthy593
Thanks sqlbelle your efforts are not just supporting individual lives to change but helping them to come out of poverty line .I think most of the people don't afford to get even paid online course but you are just providing free support .Thanks god bless you and your family . We hope if you could post tableau projects for entry level jobs .
The best relationship explanation I have ever seen! Thank you so much!
You are a light in my dark path of data analysis
@sqlbelle
3 жыл бұрын
Thank you Jafet. I am happy if the videos are helping in your data analysis 🙂
Wow! This is so clearly explained. I take a bow for your knowledge, teaching and intent to share. Without a shred of doubt, your videos are easiest to understand and most exhaustive. Thank you so much :)
@sqlbelle
2 жыл бұрын
Thank you Neeraj!
This is extremely helpful tutorial together with part 1 reg blending :) Thank you so much! I need to check your other videos. Please keep up the good work!
@sqlbelle
2 жыл бұрын
Thank you Jakub! I appreciate your comment :)
Thx Donabel for keeping your word and tackling this essential subject in Tabkeau 2020.2. I know it's a holiday season yet you still find time to make hard easy for your followers. Was hoping you can clarify the Cardinalliy and integrity features but was happy to hear you saying you will in a separate video. I like your way of keeping us tuned up to your channel 😊 Love the last quote in the Video!
@sqlbelle
3 жыл бұрын
Thank you for your comment Abderrahim! Happy to hear it was helpful! Yes I think it might be good to tackle the performance options separately so I can show the differences when you choose one over another. Hoping to have that in the next few videos. Happy New Year! Donabel
Another great presentation on this topic . you made it simple and easier to understand when to use. I like this "Use relationships if you can, join if you must, blend if you absolutely must" . I am going to watch all your videos.. learn more. Expecting your next video and topic. Thank you so much!
@sqlbelle
3 жыл бұрын
Thank you so much for your very kind and supportive feedback, Guna. Much appreciated!
Superb explanation. The only 'make sense' explanation I found on youtube
A big thank you to you to explaining this complex relationships concept! Very clearly explained!!
@sqlbelle
2 жыл бұрын
Thank you Sinchana!
You are a very talented teacher with great pace. Thank you.
this is absoulately crystal clear explaination covering all aspects..a big thank you !
@sqlbelle
Жыл бұрын
Thank you Swati!
Best video with detailed explanation. This helped me a lot to understand the concepts
@sqlbelle
3 жыл бұрын
Thank you for your comment Tarun, glad it helped!
Another crystal clear explanation from Ms. Belle. 👏
@sqlbelle
3 жыл бұрын
Thank you for your kind comment, Nicko! I am glad you found the video helpful!
Thanks for your well structured explanation , Expecting More Videos...
@sqlbelle
Жыл бұрын
Thank you Rajesh!
Best video on relationships and joins i hav come across...thanks a lot
@sqlbelle
3 жыл бұрын
Thank you Rajashekar!
Before watching this video, I had multiple doubts and confusion on this topic. Now, I am very confidence on it and easily can decide at my work environment. Thank you so much..
@sqlbelle
2 жыл бұрын
So glad to hear, thank you Arjun!
she is absolutely great teacher and blessing for tableau students.
@sqlbelle
2 жыл бұрын
Thank you so much for your kind comment Irshad!
You are bested Tableau Teacher. Thank you so so much for amazing videos. I am so grateful and thankful to you, you are doing amazing job ❤️😇
@sqlbelle
2 жыл бұрын
Thank you Kriti! I very much appreciate your note! :)
Dear, I must say your videos are so easy to understand and covers even littlest of the details... Has anyone observed how cover pic of part 1 of this video highlights 'Blend' as that part covers only blend but part 2 highlights join and relationship words, as they are included in video... It makes so easy to pick a video for re-watch.. well done and lots of encouragement to you!!!
Fantastic explanation and very well articulated!! I'm lucky people like you are out there are making life easier for people like me :)
@sqlbelle
2 жыл бұрын
Thank you SP, I am glad you found it helpful.
It helped me a lot ☺️ Thanks a lot for making this video and explaining it so beautifully
You are a great educator. The way you explain the things is just amazing. Please try uploading a video on covering all these in a scenario based case study where we have multiple data sets and we have to use relationships + (Joins or blend) and some common mistakes we do while building the viz.
@sqlbelle
2 жыл бұрын
These are great suggestions, thank you Shashank!
Best Explanation Ever. I love the way u explain.
@sqlbelle
2 жыл бұрын
Thank you Chethan!
Thanks Belle.. you are the best ! very detailed explanation.
@sqlbelle
3 жыл бұрын
Thank you Siva! Hope you found it useful!
This is just Awesome!! Thank you so much!
Very well explained. You are a great teacher!
@sqlbelle
2 жыл бұрын
Thank you Adriana!
A great explanation for these daily decisions!! Thanks
@sqlbelle
Жыл бұрын
Thank you Cristopher!
Every learner expects this kind of explanation. Thank you so much for your time and knowledge mam. Waiting for video on "cardinality and referential Integrity".
@sqlbelle
2 жыл бұрын
Thank you Kranthi! Appreciate your comment :)
thanks again for detailed explanation !!! Eagerly waiting for Part -3 (Performance Options) on this topic :).
@sqlbelle
3 жыл бұрын
Thank you Jayasree :)
Hi Belle..Thank you so much ! very detailed explanation, and it is vary useful, keep posting. :)
@sqlbelle
3 жыл бұрын
Thank you Mahesh! Glad you found the video useful! 🙂
We absolutely enjoyed the video..!!!
@sqlbelle
2 жыл бұрын
That's great to hear, thank you Tarun!
It is the best session I have ever seen, thanks so much for the content & presentation. I have also followed you on Tableau Public. :)
@sqlbelle
Жыл бұрын
Thank you Sunil!
Wow... I cud c d amount of effort that was put in to make this... Just wow... N d explanation is fantastic.. U hf no idea how much u r helping us.. thank you soo much..
@sqlbelle
3 жыл бұрын
So happy to hear Sindhuja. Happy to hear you are finding the videos helpful :)
Superb. Liked the way you explain. Thanks
@sqlbelle
2 жыл бұрын
Thank you Ramees
As always the best teacher ...thanks a lot a lot...very very usefullllll....
@sqlbelle
3 жыл бұрын
Thank you so much VDN, so happy to hear you found it useful! 😊
Very good content ...Now I am clear about the blend , join and relationship ...Thanks mam😊
@sqlbelle
3 жыл бұрын
Thank you Mrinal, I am glad you found the video helpful! 🙂
Thank you for the nice explanation Mam. really loved it and made my concept more clear.
@sqlbelle
3 жыл бұрын
Always appreciate your comments and support, Peter! 😊
very well organised way to describe various concepts. I definitely suggest everyone watch this awesome content....awesome.
@sqlbelle
3 жыл бұрын
Thank you for your kind comment Mehul!
@sqlbelle
3 жыл бұрын
Thank you so much for your kind comment, Mehul! Much appreciated :)
thanks for the explanation, after watching I thought it should be blend over join if possible but you seem to favour join at the end.
@sqlbelle
Жыл бұрын
Thank you Enyacat. Yes - blends and joins are for two different use cases, and where possible, the hierarchy would be 1. Relationship 2. Join 3. Blend
Great Video....Explained very well.Thank you
@sqlbelle
2 жыл бұрын
Thank you Aniket
Most beautiful demonstration I have ever seen. Kudos. It would be really great if access is provided to download these excel datasheets for practice purpose.
Thank you for the detailed explanation.
@sqlbelle
2 жыл бұрын
Thank you Aditya!
super explanation.. Thanks !
Super video. Thanks for your efforts..Very clear..
@sqlbelle
2 жыл бұрын
Thank you Vivek, glad you found it helpful.
What a great explanation
@sqlbelle
2 жыл бұрын
Thank you Paspas!
Vary welll explained as always. Thank you.
@sqlbelle
3 жыл бұрын
Glad it was helpful, Anil! 🙂
Excellent !
@sqlbelle
2 жыл бұрын
Thank you Kiran!
Excellent explanation
@sqlbelle
2 жыл бұрын
Thank you Rasika
This is a masterpiece video
@sqlbelle
2 жыл бұрын
Thank you Atlas!
Amazing explanation ❤️
@sqlbelle
3 жыл бұрын
Thank you Kiran! 🙂
awesome explanation
@sqlbelle
2 жыл бұрын
Thank you Diya!
Just Amazing!
@sqlbelle
3 жыл бұрын
Thank you Ivo!
this video deserve tons of likes and subscriptions...
@sqlbelle
3 жыл бұрын
Thank you so much Mehul!
@sqlbelle
3 жыл бұрын
Thank you Mehul! This is such a nice comment, thank you :) I do hope more can watch it :)
Best tutorial 👌👍
@sqlbelle
2 жыл бұрын
Thank you Sayak!
Quite informative video. We understand that blending is useful to play with data from different data sources. However the cross data base Joins or Relationships can solve the purpose but comes with a few limitations. Can you please make a video on to deep dive in knowing the difference between Cross database Joins/Relationships vs data blending?
Mam we want more advanced concept like this plz keep uploading..
@sqlbelle
3 жыл бұрын
Thank you for your comment Piyush! 🙂
Great video as always! I'd like to know, Donabel, what would be the best way to convert a complex dashboards from blending to relationships?
@sqlbelle
2 жыл бұрын
Thank you Tanya! I don't think there's an easy button though to convert from blending to relationships because it works different ways (for example, you will need one data source instead of multiple). You may need to re-create some of them, but it might be a good thing as it can also allow you to take advantage of other newer features in Tableau.
Happy New Year Donabel!!!
@sqlbelle
3 жыл бұрын
Happy New Year Eddy!!! 😀
You explained the concept very simply. Pls make video to explain performance options in relationship
@sqlbelle
3 жыл бұрын
Thank you Abhishek! That is a great suggestion. I will definitely try to include that in one of the next videos.
@abhishekarora1
3 жыл бұрын
@@sqlbelle I will look forward to it...
You're amazing👍🏻
@sqlbelle
3 жыл бұрын
Thank you! Hope you found the video helpful!
Great video! thank you so much!! Also, can I apply filters when I creat join calculations? For example I only want to show the customers that purchased 3 items or more, is there a way to express that in the join calculations? (i know we can do this by creating filters in the worksheets later on,but what if I want to filter it out on the data source level)
@sqlbelle
2 жыл бұрын
Thanks Jiangzheng, that's a great question. If the conditions are more than just simple conditions, I am inclined to recommend doing a "cleaning phase" first before connecting them in Tableau, if possible. For example, you can use some ETL tools on the database side, or for example client tools like Tableau Prep or Power BI.
Wow… this is really nice. It helps a lot. Can you provide same datasets (3 excels) which you being used in this tutorial is really appreciated.
Thanks. Have you dived deep into relationships? So e.g. why does it matter which table you first drag into Tableau when setting up your "relationship" model? What if there are multiple fact tables and you need to have the same dimension tables twice in your model? Have you experienced any strange behavior so far? Would highly appreciate more on this as the documentation and videos (at least I found so far) are not addressing these question in detail enough for me. best regards and take care
@sqlbelle
3 жыл бұрын
Thank you for your note Daniel, and these are great points. I am not able to address some of the why's of Tableau as I am not privy to those, but I can understand why they may have chosen to go this route. Perhaps there are different ways to tackle this. In my projects I would handle drill across in fact tables using multiple views from the database end. I would be wary of having a different tool automatically handle this especially if the dimensions are used in different kinds of fact tables (transactional vs periodic for example). I would opt to do the joins within the database itself and expose as a view for Tableau because it provides more control, and will allow me to address nuances of my specific datasets outside of Tableau. I will take advantage of Tableau's features like Joins, Blends, Relationships on smaller data sets, but wherever I can I would data cleaning and modeling to the data source. Often my data sources are powerful/resource-heavy enough to handle the cleaning/modeling with more ease than Tableau. I am not sure if this is helpful, but just wanted to share with my own thoughts and strategies.
Enlightened :)
@sqlbelle
2 жыл бұрын
Thank you Kirti!
@sqlbell. For practice purposes is possible to attach the exact data sets you use for each video?
Omg finally a good, really well done example of why relationships are so important in terms of making analytics more accessible. Follow up question on your parting advice. How do we create relationships between published data sources? So let’s say, I have customers and orders and products all nicely related and I publish the extract to be refreshed daily. My colleague is analyzing delivery estimates which are a function of order total and customer membership tier. They don’t want to also create the same relationships that I already setup, and they just went to add some more to it… is blending just the best option in that scenario?
excellent
@sqlbelle
2 жыл бұрын
Thank you Anil!
This is a awesome video. Could you please let me know like how to handle Measures while using relationships. Tableau by default applying left join when we pull measures from more than 1 table. which we need to filter out the data later. i am looking for a way to fix this issue without using filters. It will be great for me if you can share few tips related to this.
cleared all my confusions thank you! It is impossible to learn by just reading official online documentation and even the official tutorials don’t explain the concepts clearly, I wonder why that’s the case…
@sqlbelle
2 жыл бұрын
That's great to hear, thank you Jia! Glad to hear you found the video helpful!
Great Explanation, can you create a video on improving the performance of the dashboard connected to a live data source?
Hi, Could you please post a videos about cardinality and referential integrity? Also, could you please post a video explaining very clearly about the diagram that you have shown in the video for relationships(please take an example for each of the element in the diagram). I think your explanation is just out of the world.
@sqlbelle
2 жыл бұрын
Thank you for the great suggestions Sunil!
Can I get those 3 excel sheets to practice the same things on my PC ? That would be really helpful.
Perfect, you are gem,,,
@sqlbelle
3 жыл бұрын
Aww. Thank you Vik! Hope you found the video helpful! :)
Hey, in data relationships, is there a way to connect two tables that are already connected to two different tables? (like it can be done in PowerBI by a simple drag & drop)
@sqlbelle
2 жыл бұрын
Hello Aditya - do you mean to pull all related fields automatically? No this option does not exist at this point in Tableau.
Hi Belle, Is it possible to share the Excel, I wanted to test use case at 17.10, what if I want to add product ID and next to it the Salary earned by Customers? To know, what is the range of salaries of customers who are buying particular products. I believe the tableau will again do a sum(salary) if I have a customer buying a product more than once. In this case, can we use again FIXED LOD?
@sqlbelle
3 жыл бұрын
Thank you Aditya. I will post the data set. I will reply back here (and add to description) once I add the data set
If I can get the path to dataset so I can follow along? Also any video on joins with Calculated fields? thx & much appreciated.
@sqlbelle
3 жыл бұрын
I am slowly putting all the workbooks in my Tableau Public (the latest one on LODs have an accompanying workbook). Once this is ready, I will make a video or announcement to let anyone who might be interested know.
superb ....mam do uploads dataset as well with your videos.
@sqlbelle
2 жыл бұрын
Thank you for the suggestion Nitesh!
I have understood the concept but how to explain in interview if asks when to choose joins and relationships?
Thanks for sharing this but can you make video on Context filter Advantage and disadvantage and how many context filter we can use in tableau ,so that performance is not affected
@sqlbelle
2 жыл бұрын
Thank you for the suggestion Mukul!
great
@sqlbelle
2 жыл бұрын
Thank you Abhay!
Hi, Can you provide the link for excel file that you used here for Joins and Relations and blend?
is it correct that using Join will solve the * issue in blend? Because in join, there is only 1 database (no primary or secondary)?
Can relationships are used to join data from multiple data sources
Great! Where i can find this excel file?
Good explanation of a complex Tableau topic. Please share the dataset if possible Donabel, would like to practice the concept.
@sqlbelle
3 жыл бұрын
Thank you Pooja, I will upload the workbook with data sets to Tableau public. I'll let you know when it's there :)
@mahmoudhamza6765
2 жыл бұрын
@@sqlbelle great video as usual a kind reminder for sharing the dataset if possible
Hi Belle, Please guide me on this question. If possible please do some videos on real time scenario base. 1- I want to share the dashboard to my 3rd party client? So, in which format can ? Thank you
@sqlbelle
3 жыл бұрын
Hi Akhtar - you have a few options: - Tableau Online (if you can publish your data set on the cloud) - Tableau Reader (you can send them the twbx file, and ask them to download Tableau reader nd view it from there) These will depend on the security requirements you have for your dashboard
When you mentioned, "the solution was to create multiple data sources" (at 11:32) was to use data blending option? BTW, love your videos!!!
i had a doubt on below statement can you explain .. its from tableau help page relationships cannot be formed between tables from data sources published to Tableau Server or Tableau Online.
@sqlbelle
2 жыл бұрын
Thanks Sampath. I think you are correct that this is still the case (as per: help.tableau.com/v2020.2/pro/desktop/en-us/relate_tables.htm) It means that if you connect to a Tableau Server data source for your tables, you cannot establish relationship between those tables .. or at least they are not supported yet
@sampathkumarreddy4354
2 жыл бұрын
I don't understand the statement can u clearly explain... If possible please make a video's on Tableau online and server .. please.
@sampathkumarreddy4354
2 жыл бұрын
when we are publishing report (embedded in or published separate)on Tableau online or service ...if we publish separate data source and report are in different folders right. By using published separate data source we cannot form relationship ... Am I correct
Query: At 9:00, you have used fixed LOD to show correct salary in joins example. If we use average of salary or min of salary or max salary without LOD that's also gives same result. Is there any specific reason to use LOD here?
Wow!
Could you provide datasets to practice?