This blog is a preview of my session at Sirus Connects on October 9, 2026, where we explore how to choose the right Power BI storage mode for your use case, comparing Import, DirectQuery and Direct Lake across performance, scalability, flexibility and governance.
Speaker:
"Import is fast." "DirectQuery is flexible." "Direct Lake is the future." I hear these one-liners all the time, and none of them are wrong, exactly. But after twenty years of designing data platforms, I've learned that the moment you treat a storage mode as a label instead of a decision, you're setting yourself up for trouble down the road.
Over the past twenty years, I’ve helped organizations design and implement data platforms within the Microsoft ecosystem, and if there’s one decision that gets underestimated time and again, it’s the choice of storage mode. Import, DirectQuery, Direct Lake: on paper, they look like a technical checkbox you tick early in a project and move on from. In practice, that one choice quietly shapes almost everything that comes after it: how fast your reports feel, how fresh your data is, how easy your model is to govern, and ultimately, how much it costs to keep the whole thing running.
Here’s the thing I see over and over again in the field: teams often pick a storage mode based on habit, on what worked “last time,” or simply because it’s the default option, not because they’ve actually weighed it against their specific use case.
And I get it. The technical definitions of each mode are well documented, but the practical trade-offs, the ones that actually determine whether your solution will scale gracefully or become a maintenance headache, rarely get the attention they deserve.
There is no universally “best” storage mode. There’s only the right one for your data volumes, your refresh requirements, your governance needs, and the experience you want your users to have.
Direct Lake is changing the equation
And the landscape is shifting again. With Microsoft Fabric and Direct Lake, the rules of the game are changing in ways that genuinely excite me.
Direct Lake blurs the line between the flexibility of DirectQuery and the performance of Import, and it opens up architectural possibilities that simply didn’t exist a few years ago.
But with new possibilities come new questions: When does Direct Lake actually make sense? What does it mean for your existing semantic models? And how do you future-proof an architecture in a space that keeps evolving?
Why streaming isn't batch
This is where streaming differs from batch. In batch, if a job fails you run it again. The data was always there, patient and waiting.
In streaming that safety net is gone. Your source might not buffer indefinitely. Your window to recover can be measured in hours, not days. The stakes are different, and the solutions have to be too.
From technical definitions to real-world decisions
That’s exactly what I’ll be unpacking in my session. I won’t just walk you through the technical definitions. You can find those in the documentation.
Instead, I want to give you a practical decision-making lens: how to weigh performance, scalability, flexibility, and governance against each other, and how to make a well-informed choice instead of a default one.
Expect realistic scenarios, honest trade-offs, and a few insights that might just challenge how you’ve been approaching storage modes until now, including a clear-eyed look at where Direct Lake genuinely changes things, and where it doesn’t (yet).
Make your next storage mode choice a deliberate one
Whether you’re a BI developer knee-deep in model design, a data engineer building the pipelines underneath, an architect drawing the bigger picture, or a decision maker trying to keep costs and complexity under control, I promise you’ll walk away with something concrete you can apply to your next project. Not just theory, but decisions you can actually defend in your next architecture review.
I’d love to talk through all of this with you live. Come find me at Sirus Connects on October 9!
Want to follow this live?
Join this session at Sirus Connects on October 9, 2026, from 11h20 to 12h05. Register and secure your spot.
Can’t make it? Reach out to us, we are happy to share our experience and help you choose the right Power BI storage mode for your data, performance and governance needs.
