The Clearly Podcast

Andy: Hello everyone and welcome to the Clearly Podcast, Episode 15 on Self-Service Business Intelligence (BI). I'm Andy Clark, joined by Shailan Chudasama and Tom Gough. Today, we’re discussing the evolution of BI tools over the past 15 years—from IT-created reports to user-created ones. We'll cover how to deploy self-service BI, considerations for success, and common pitfalls. Let's dive in. Hey guys, how are you doing?

Tom: Yeah, very good. Managed to avoid the floods so far.

Andy: Glad to hear it. So, today’s topic is self-service BI. This approach allows users to build and modify reports without relying on IT. We're moving from IT-built reports to a model where users can create their own. We’ll discuss how to define, implement, and manage self-service BI. Shailan, how do you define self-service BI?

Shailan: Self-service BI lets end users create and extend reports using a common data set and proper tools. Initially, it involved extending existing reports, but now users can create custom reports from scratch.

Andy: How does your approach differ when building a self-service model versus a traditional model, Tom?

Tom: For self-service, focus on making everything human-readable. Use clear, descriptive names for fields and tables to make it easier for users. Simplification is key. Users should be able to drag fields without encountering errors, which can be discouraging.

Andy: Absolutely. Simplification and making the model robust are crucial. How do you think about the technical aspects of self-service BI, Shailan?

Shailan: Technically, self-service BI involves providing a common data set that users can browse to create reports. Tools like Power BI, Excel, or SQL Server Analysis Services can be used. It’s about enabling users with the right tools to explore and analyze data.

Andy: Tom, you mentioned that self-service is more of a strategy than just technology. Can you elaborate?

Tom: Self-service BI should be part of a broader strategy, empowering users to work with data directly. However, it’s important to manage and control this to prevent issues like report overload.

Andy: How do you approach training users for self-service BI, Shailan?

Shailan: Involve users early in the project. Train them as the models are being developed, so they’re familiar with the tools from the start. This reduces training time later and ensures they’re comfortable using the system.

Tom: Post-go-live, sit with users to help them build reports. This hands-on approach helps upskill users and demonstrates the system's capabilities, fostering long-term engagement.

Andy: Regular follow-ups and small training sessions work well. Some users will take to it and start building reports, while others may prefer to consume pre-built ones. It's important to cater to both types.

Shailan: Ongoing maintenance is also crucial. Monitor report usage and clean up unused reports to avoid clutter. Have a designated gatekeeper to manage the common workspace and ensure it stays organized.

Tom: Yes, and training materials should be reusable. Create videos or guides that users can refer to for refresher training or when onboarding new team members.

Andy: Great points. Finally, let’s wrap up with our top tips for self-service BI. Shailan, you first.

Shailan: Involve users early in the project and keep models simple and user-friendly.

Tom: Self-service should be part of your strategy, not the entire strategy. It should complement professionally built reports.

Andy: Train users regularly, building up complexity over time. Use reusable training materials to keep skills sharp.

Shailan: Absolutely. Self-service should be friendly and manageable.

Andy: Thanks, everyone. Next week, we'll be joined by Reza Rad to discuss data governance. Until then, stay safe and have a great week.

For more information, visit clearlycloudy.co.uk for the UK and Europe, and clearlysolutions.net for North America.