The Clearly Podcast

Andy:
Hello everyone, and welcome to the Clearly Podcast episode 13: Management of Business Intelligence Projects. Recorded on Thursday, January 7th, 2021, with Shailan, Tom, and myself, Andy. Today, we'll discuss managing BI projects, technological changes, and key factors for a successful BI implementation. We suggest starting BI implementations early, alongside business systems projects. How's everyone doing?

Shailan & Tom:
Good, thank you.

Andy:
I'm doing well. It's sunny in Houston, but we're expecting -7°C next week, which will be quite a shock.

Today's discussion is about managing BI projects specifically. This isn't just another project management conversation but focuses on BI tools and products. Shailan, you and I had a discussion about the tendency to treat BI as an afterthought in ERP or CRM projects, often leading to issues later on.

Shailan:
Absolutely. BI often gets overlooked in ERP or CRM implementations. People think out-of-the-box reports will suffice, but these are usually just transactional reports. They don’t cater to all data sources or reflect the full scope of an organization’s data.

For example, in an ERP implementation we worked on, we had a parallel BI workstream from the start. This included creating datasets, models, and reports from existing systems and integrating them with the new ERP. This approach ensured a smooth transition and familiarized users with the BI tools during the project.

Tom:
I've seen projects where BI was an afterthought, resulting in missing data needed for reports because it wasn’t captured in the source system. It's crucial to consider reporting needs from the beginning.

Andy:
Exactly. I always ask users to begin with the end in mind. Think about the reports needed post-implementation to ensure all necessary data points are captured. This helps avoid surprises later.

Tom:
This approach aligns with program management, where BI is a strand within a larger program, such as an ERP or CRM project.

Andy:
Yes, BI should run parallel to business system implementations, making it easier for users to transition. Training should also be ongoing, not just at the end, to ensure users are familiar with the tools and data.

Shailan:
Having a project team with an executive sponsor is crucial. Defining KPIs, data models, and report functionality early on helps set standards and expectations.

Andy:
Standardization is key. Consistent placement of slicers, buttons, and drill-throughs across reports helps users navigate them easily. Proper testing, including usability and performance, is essential. Testing should involve someone other than the report creator to ensure objectivity.

Tom:
Testing should also include non-functional aspects, like performance, to ensure reports run efficiently.

Andy:
Naming and version control are also important to avoid confusion and maintain clarity.

Tom:
Re-platforming projects, where existing BI tools are replaced with new ones like Power BI, require careful scoping. Simply replicating old reports misses the opportunity for improvement. Business requirements must be re-evaluated to ensure they align with the new tool’s capabilities.

Andy:
Yes, it’s important to educate users on the new tool’s capabilities to avoid recreating old processes.

Shailan:
The diversity of data sources today means ERP is no longer the sole data repository. Cloud sources, social feeds, and other external data must be integrated for comprehensive reporting.

Tom:
This diversity increases the need for project management to handle multiple subject matter experts and data sources.

Andy:
Projects now are more complex, requiring dedicated BI project management. Treat BI projects with the same rigor as software development projects, involving proper teams, structures, and methodologies.

Shailan:
Start BI early in the implementation process. It helps with data migration, quality testing, and user engagement.

Andy:
Most projects today use agile methodologies, allowing for iterative development and continuous improvement.

Tom:
While agile is common, there’s still a place for waterfall approaches, especially for the initial delivery of minimum viable products.

Andy:
Great points. To wrap up, let’s share our top tips. My tip is to treat BI projects seriously, with proper teams and structures.

Tom:
Ensure a strong business lead for BI projects, not just technical management.

Shailan:
Start early, treat BI as a workstream within the overall project, and engage stakeholders throughout the process.

Andy:
Thanks, everyone. We hope you found this discussion useful. Get in touch with us at clearlycloudy.co.uk for the UK and Europe, or clearlysolutions.net for North America. We'll be back on February 1st, 2021, to discuss testing methodologies in Power BI. Stay safe and have a great week.