Forum Discussion
Power BI technical requirements document
Hi
I am trying to put together an initial technical requirements or prompts for things that would be good to know about at the start of a new power PI project as outlined below.
Ideally, this document would be completed by a product manager with help from other individuals after the project definition document has been completed. The topics outlined below are likely to end up as a quick reference cheat sheet that will be continuously updated throughout the project cycle.
Please can you provide any feedback or links to useful resources?
Initial Technical Requirements:
Project:
List names of users, distribution lists (DL), or Azure Active Directory groups (AAD) that will be involved in the project stages listed below.
- Prototyping
- Developer
- Testers
- End users
- communications
List people who can help with the development
- Product manager
- Data sources
- User testing
- UI Design
- Dax help
- Modelling help
- Publishing/ deployment help
What additional software might you need?
- Power automat
- Power App
- Power Bi report Builder
- Other
List old products that may still be running when this project is done
- Product name
- Estimated dual running period
Data:
List data source needed for the project and characteristics
- Data source name
- Is data source available
- Data source origin:
- Data type (dataset, Excel ....)
- Where is data stored
- authorisation:
- data sensitivity or private level
Data source wrangling and Maintenance
- Data source name
- Has it been Quality assured
- Where is ETL done? (source, power bi, hybrid)
- Refresh frequency and type (manual, automated)
- Import type
- Is data modelling done at source or Power BI
Report Build:
Where will the file be stored
- Folder/ SharePoint path
- Access
- Ownership
Report page outline
- Report name
- main purpose
- Main Visuals
- Type of page navigation and filters
Product version control
- Using SharePoint Check in and out
- Manually renaming files
- Other, specify
Deployment:
Where will the end users consume the product
- Workspace
- App
- Dynamics 365
- Embedded
Workspace details
- Workspace Name
- Purpose
- Is Available
- Used for (holds dataset or reports, hybrid)
- Refresh maintenance type (automated, manual)
- User/Groups with Access and Level
- Used In Deployment Pipeline
- Used Deploy Power BI App
I think it always depends on your particular use case as to how much you need depending on resources (small teams many of the functions you mention here are sometimes just one person) , what skills and tools you have available to you.
additional things i might be thinking about are
test cases created related to test driven development https://www.amplework.com/blog/test-driven-development-and-best-practices/
CI/CD (continious integration/deployment) depending on the type of project you are doing orr acceess to tools
definitely where the artifacts needs to be kept in source control at the very least
a good approach to a development lifecycle will bring out the correct information also at the same time
ie requirements / profile data / data quality / prototype / iterate
interfacting with the data owners who may or may not be on the project.
also i really like data goblins checks lists he has quite a few and are really helpful too
1 Reply
- vanessafvgCommunity Champion
I think it always depends on your particular use case as to how much you need depending on resources (small teams many of the functions you mention here are sometimes just one person) , what skills and tools you have available to you.
additional things i might be thinking about are
test cases created related to test driven development https://www.amplework.com/blog/test-driven-development-and-best-practices/
CI/CD (continious integration/deployment) depending on the type of project you are doing orr acceess to tools
definitely where the artifacts needs to be kept in source control at the very least
a good approach to a development lifecycle will bring out the correct information also at the same time
ie requirements / profile data / data quality / prototype / iterate
interfacting with the data owners who may or may not be on the project.
also i really like data goblins checks lists he has quite a few and are really helpful too