BenMakesDataEasy

A little more walk through on a new forecasting script I created that ties together Python + FB Prophet + Power BI : ) Check out the script used in the community area on my YouTube channel. Greatly appreciate likes/subscribes there as well : ) Keep at it everyone! @BenMakesDataEasy-t1/community" target="_blank" rel="nofollow noopener">youtube.com/@BenMakesDataE…

Can you do real ACID writes on Delta from pure Python? of course !! no catalog needed — just pip install deltalake and a path. #rust #deltalake #python #polars #duckdb #datafusion datamonkeysite.com/2026/05/24/how…

DuckDB is one of my favorite data technologies, but I have found it a little difficult to know exactly where to start. So I kicked off a "DuckDB Basics" series. My first article was on using SQL in the CLI. Here is the companion video for that article. youtu.be/yO0PGr9ydQo

We spend a lot of energy making fun of BI analysts and developers. This time it's the users' turn. User-aware calculated columns are now a thing in DAX. The report can personalize itself based on who's viewing it. Translations, dynamic content, no RLS needed. Turns out, not everyone loves the idea of their report knowing who they are. Which raises two questions:  1) Have you ever heard of system logs? 2) What exactly are you doing in that report that you don't want anyone to know about? 😎 If you want the technical details: sqlbi.com/articles/intro…

The Simplest Way to Connect Claude and Power BI

What kind of parameters can you use in DAX user-defined functions? This video introduces parameter types and passing modes in DAX UDFs: learning the difference between VAR and EXPR passing modes is the critical element to learn. A practical demo to understand how parameters work in UDFs and why they matter when creating reusable DAX logic. Watch the full video here: youtube.com/watch?v=q29Lyc…

NEW BLOG POST! Quick question for Fabric Warehouse folks: If clustering reduces CPU by 40% and I/O by 69%, but elapsed time stays the same, is it working? Yes. And the explanation is more interesting than the answer. 100M-row demo with real numbers: data-mozart.com/cluster-by-in-…

Finally: pixel-perfect column widths in Power BI!

I currently run 3 businesses: 1. Alex The Analyst on YouTube I say "business", but this one is my lowest revenue and time investments, but I do make some money from it. I do it because I love it and have always loved it - this takes maybe 5-10 hours per week. 2. Analyst Builder This is my ed-tech platform to teach people Analytics and help them land jobs faster. This takes more time and I have a whole team that supports this platform. This takes 20-30 hours per week. 3. Alex Analytics LLC This is my consulting business where I work with tech companies and startups. I do everything from helping with pricing strategy, data migrations, data analysis, to creating content. This takes 30-40 hours per week. I didn't start out with all 3, they evolved over time. I started with my YouTube channel, and as that grew, I slowly started consulting. I started a small mentorship program and as that grew I created Analyst Builder to serve more people. Things don't happen overnight and they don't always look like what you think they will. I'm a big believer in starting small and growing it until it makes sense to pivot.

Friends, I just witnessed a miracle :) Parquet with 600 M rows in #onelake , #duckdb server in my laptop, #PowerBI in Direct Query mode using this experminetal driver github.com/CurtHagenloche…

Used the excellent #AzureStorage lease API to harden a #Ducklake single writer in #OneLake 😃datamonkeysite.com/2026/05/16/ens…

AI agents help you do your work. Will they also decide when you need a vacation? ;) If you don't get the reference, Plan is a new Fabric feature for Enterprise and Corporate Performance Management, launched a few months ago, when we published the comic on our SQLBI Newsletter. Plan is not AI-based, since it’s Saturday, we just thought it was fun to imagine what would happen if it were. If you prefer boring stuff on a Saturday, here's the documentation: linkly.link/2ixiw

Why do you need an eventhouse in #MicrosoftFabric? What are its main strengths and use cases? How it compares to lakehouse and warehouse? Watch the latest Fabric Tech Talk Fridays episode! I had a loooot of fun chatting with one and only Patrick LeBlanc. youtu.be/8GUlLSS3kEo?is…

New version of the power-bi-agentic-development plugins. - Updates to fabric cli skill for new fabric cli versions - Clarifications for those using Copilot CLI on how to use it, and clarifications on how to use just a single plugin. - Various small useful stuff including reference settings.json, status line script, and more - Switching to a YY-WK release numbering Hoping for some bigger updates next week; kids are still sick. Repo link: github.com/data-goblin/po…

You now have the option to use Storage Tiers in #MicrosoftFabric which could save you costs on data that is not being used often. Read the post for more details. #PowerBI community.fabric.microsoft.com/t5/Fabric-Upda…

DuckDB: Quack Remote Protocol You can use DuckDB between multiple machines out of the box... set up a database on a server and work with or query it remotely from another device. Just started testing and it's so simple... This seems pretty damn amazing what you get from an OSS duckdb.org/2026/05/12/qua…

Love the DuckDB community and super thankful to be spotlighted in this month's community newsletter. I'll be speaking at the @motherduck + DuckDB meet up on June 3rd in San Francisco showing off DuckLake's streaming and ETL capabilities. Agents too. motherduck.com/blog/duckdb-ec…

AI is essential in today's data work. In Fabric, Power BI, and Azure, AI tools accelerate delivery, but without DataOps they create risk. Here are 9 tips to reduce GenAI risk and improve data quality: blog.kerski.tech/ai-in-data-les…

Fabric CLI just keeps getting better. New find command let's you search across all workspaces with the onelake catalog api. Finally a public documented version of the datahub apis. Hopefully as rich in the properties though.

I've been consulting in the data space for about 4 years now. Here are 5 things I've learned that I thought was really interesting: 1. 90% of clients actually do know what they want - just not how to execute it effectively 2. Managing Expectations is half the battle - delivering is the other half 3. Small customers are 5x the work of larger customers and pay half as much 4. Doing good work and getting repeat customers is hard, but the best way to sustain long term growth 5. Project creep happens with almost every project, addressing it and not giving into it should be the norm

𝗦𝘁𝗮𝗿𝘁𝗶𝗻𝗴 𝗶𝗻 𝗮 𝗳𝗲𝘄 𝗛𝗼𝘂𝗿𝘀- 𝟵𝘁𝗵 𝗠𝗮𝘆 𝟮𝟬𝟮𝟲- 𝟳 𝗔𝗠 𝗜𝗦𝗧 - 𝟴𝘁𝗵 𝗠𝗮𝘆 𝟲:𝟯𝟬 𝗣𝗠 𝗣𝗦𝗧 40+ Sessions 𝗝𝗼𝗶𝗻 𝘁𝗵𝗲 𝗚𝗹𝗼𝗯𝗮𝗹 𝗠𝗶𝗰𝗿𝗼𝘀𝗼𝗳𝘁 𝗔𝗻𝗮𝗹𝘆𝘁𝗶𝗰𝘀 𝗖𝗼𝗺𝗺𝘂𝗻𝗶𝘁𝘆 𝗖𝗼𝗻𝗳𝗲𝗿𝗲𝗻𝗰𝗲 biworld.in/events.htm