Skip to main content
cancel
Showing results for 
Search instead for 
Did you mean: 

Register now to learn Fabric in free live sessions led by the best Microsoft experts. From Apr 16 to May 9, in English and Spanish.

Reply
cristianpark
Frequent Visitor

Development team working in same dashboard at the same time

Hi folks, I'm seeking for help on this subject since I'm running out of ideas and we need a scalable solution for following problem:

Say we created dashboard "Overall Profit" (it contains tons of measures and bookmarks created on it), we used Power BI Desktop for development and Power BI Server for publishing it. Our users are happy with it and heavily using it. After a while they come with different enhancenments, after requirements analysis we created this three tasks:

1. Data Model Change.  It requires the addition of new tables, handling new relationships, creating a bunch of new measures and updating visuals to use new measures.

2. Create new page "Geo". It requires creating a brand new page with new components, couple of measures and 30 bookmarks for navigation.

3. Create new page "Report". It requires creating a brand new page with new components, couple of measures and 3 bookmarks for navigation.

 

In this scenario, the complexity order would be 3., 2., 1. (from easier to harder to implement) but we would like to have our development team working on them at the same time.

For 3. and 2. we have no big issue since we can use copy and paste of visuals (we're glad the functionality it's finally there in PBI Desktop!) and we can coordinate to create measures and bookmarks somehow easy (in big teams it will be a challenge for sure); for 1. nevertheless, we'll have a big blocker in our way because we're going to need to redo some work at some point, e.g. if 3. and 2. get delivered sooner, we'd need  to redo all the work involved in 1. and as stated here, is not minor work, so it worthed nothing that we made previous work on it if for merging with current version we have to redo almost everything.

 

I'm thinking about these three possible solutions:

A. Exporting to a source control friendly environment. We're using GIT for source control but as you know, it doesn't work good with binary files so I'm wondering if maybe there's a possibility of exporting to a folder structure with xml files or so.
B. Using Power BI Service instead and use workspaces (I don't know much about Service yet but I don't think putting data in Microsoft's cloud will be an option and since 1. is requiring data model changes I don't think I'll work either).

C. I just bumped into templates, and I'm wondering if using them here would be a better approach to merge everything in a single light-weight version of dashboard and then we can take it as central point of development and deliver dashboards faster based on it.

 

Have you run into similar issues? can you please give me some directions on how to solve this please?

Thanks in advance

3 REPLIES 3
v-xicai
Community Support
Community Support

Hi @cristianpark ,

 

I am not sure what desired result would you want, could you please share your sample data and desired output screenshots for further analysis? You can also upload sample pbix to OneDrive and post the link here. Do mask sensitive data before uploading.

 

Please read this post to get your answer quickly: How to Get Your Question Answered Quickly.

 

Best Regards,

Amy

Hi Amy, thanks for your response.


I know this is hard to diggest but actually I'm not looking for help with a particular development on Power BI but help on how to define best practices (and maybe use other tools) for having a large development team working in same file at the same time.

 

So, generally speaking I want to pass from Overall Profit v1

1.png

 

 

 

 

 

 

to Overal Profit v2

2.png     

Having three different developers working in parallel in the same file.

 

Let's review the progress on dashboard over time:

2019-09-18: We have our Overall Profit v1 and our developers create their own copies Overall Profit v1_1, Overall Profit v1_2 and Overall v1_3 (we said that tasks complexity is 3-2-1 from easier to harder to implement).

 

2019-09-19: Developer for task 3. finishes his work on it so now Overall v1_3 can be deployed and let's stop here for a moment.

At this point Overall Profit v1_1 and Overall Profit v1_2 are work in progress versions and they don't have changes made in Overall v1_3 and since the changes include bookmarks is not only a matter of copy-paste objects from one dashboard to other.

 

So we have a decision to make:
Either Other developers should replicate Overall v1_3 changes in their own copies (this can lead to errors)
or they know that Overall v1_3 is already done and as soon as they finish their changes (2. is more likely to be done soon than 3.). If their changes are way too big (we put 30 bookmarks creation in 2. and radical data model, measures and visualizations changes in 3. for making them big changes) developers must almost redo their work at some point in time (the more time passes between developers ending their tasks, the more complicated merging changes it gets).

 

2019-09-23: Developer of task 2. finishes his work. On 2019-09-19 we decided to redo the advances that he had at the moment, so his version has all changes so far.

 

2019-10-10: Developer of task 1. finishes her work (it took her so long because changes were big ones and were impacting several visualizations and tables). Now she's facing a challenge: how should she proceed applying changes?

Should she redo all of her work in latest version (containing changes from 3. and 2.) or should her take her copy and try to merge others work on it, the later with high probability of adding errors to final dashboard since she doesn't know for sure what changes have been done to dashboard.

 

So questions are:

- How to merge from this point?

- Can we use a different technique than the one used by this development team?

- Can we benefit somehow by using source control like GIT or because .pbix is binary file and we have no other exporting option for dashboard, we're out of luck?

 

I hope is clearer now with dashboards versions.

I'm wondering if you ever figured something out from this.  I lead a team of 5 bi engineers and the best thing we can come up with is one person working on one report at a time (if you can only imagine managing the product having everyone working on different things...)

Helpful resources

Announcements
Microsoft Fabric Learn Together

Microsoft Fabric Learn Together

Covering the world! 9:00-10:30 AM Sydney, 4:00-5:30 PM CET (Paris/Berlin), 7:00-8:30 PM Mexico City

PBI_APRIL_CAROUSEL1

Power BI Monthly Update - April 2024

Check out the April 2024 Power BI update to learn about new features.

April Fabric Community Update

Fabric Community Update - April 2024

Find out what's new and trending in the Fabric Community.