Brecht Dekeyser

𝐍𝐞𝐰 𝐕𝐢𝐝𝐞𝐨 𝐀𝐥𝐞𝐫𝐭 𝐨𝐧 𝐁𝐡𝐚𝐫𝐚𝐭𝐡 𝐈𝐧𝐬𝐢𝐠𝐡𝐭𝐬 🎥 Have you ever jumped straight into building a Power BI dashboard… …and later realized the layout looks messy or doesn’t meet user needs? That’s where 𝐃𝐚𝐬𝐡𝐛𝐨𝐚𝐫𝐝 𝐖𝐢𝐫𝐞𝐟𝐫𝐚𝐦𝐢𝐧𝐠 comes in! In my latest video, I explain: ✅ What Power BI Dashboard Wireframing is ✅ Types of wireframing approaches (low, medium & high fidelity) ✅ Why wireframing is a game-changer for clean, impactful dashboards 💡 Think of it like sketching a house before construction ,wireframing helps you save time, reduce rework, and deliver dashboards that truly tell a story. 📺 Watch the full video here 👉 https://lnkd.in/gVQ3gQVH If you’re into 𝐏𝐨𝐰𝐞𝐫 𝐁𝐈, 𝐝𝐚𝐬𝐡𝐛𝐨𝐚𝐫𝐝 𝐝𝐞𝐬𝐢𝐠𝐧, 𝐨𝐫 𝐝𝐚𝐭𝐚 𝐬𝐭𝐨𝐫𝐲𝐭𝐞𝐥𝐥𝐢𝐧𝐠 this one’s for you! Let me know your thoughts in the comments 👇 #PowerBI #DashboardDesign #Wireframing #BharathInsights #DataAnalytics

61

3 Comments

During the "Ask the Expert" shift at the European Microsoft Fabric Community Conference, someone asked me: “Can Power BI have two tables marked as 'Date table'?” I’ll admit, I wanted to answer immediately, "No! Are you serious?" But I took a pause and... tested it. And you know what? You actually can do it! 🤯 But why would you need it? Honestly, I don't know 😅 Here is what my distant friend ChatGPT told: "It’s a small but powerful trick — perfect when you need separate timelines like Order Date and Delivery Date, without juggling USERELATIONSHIP() in every measure." I'm still not convinced by this answer. There is much more sense in using USERELATIONSHIP() function, but I might miss something. So my question goes to you: do you use multiple Date tables? If so, what are your use cases? #PowerBI #MicrosoftFabric #FabCon25 #DataModeling #DAX

23

26 Comments

📊 𝐒𝐭𝐚𝐧𝐝𝐚𝐥𝐨𝐧𝐞 𝐂𝐨𝐩𝐢𝐥𝐨𝐭 𝐢𝐧 𝐏𝐨𝐰𝐞𝐫 𝐁𝐈 𝐨𝐧 𝐝𝐞𝐟𝐚𝐮𝐥𝐭!   Starting early September, Microsoft is making a 𝐛𝐢𝐠 𝐬𝐡𝐢𝐟𝐭 for tenants that already have Copilot enabled in Power BI by setting 𝐒𝐭𝐚𝐧𝐝𝐚𝐥𝐨𝐧𝐞 𝐂𝐨𝐩𝐢𝐥𝐨𝐭 𝐎𝐍 𝐛𝐲 𝐝𝐞𝐟𝐚𝐮𝐥𝐭. ✨ In simpler words, 𝐅𝐚𝐬𝐭𝐞𝐫 𝐐𝐮𝐞𝐬𝐭𝐢𝐨𝐧-𝐅𝐚𝐬𝐭𝐞𝐫 𝐀𝐧𝐬𝐰𝐞𝐫! ✔️ Query semantic models without opening reports ✔️ Get explanations in natural language ✔️ Instantly generate visuals & summaries 🔔 In my next post, I’ll cover 𝐰𝐡𝐨’𝐬 𝐢𝐦𝐩𝐚𝐜𝐭𝐞𝐝, 𝐰𝐡𝐚𝐭 𝐚𝐜𝐭𝐢𝐨𝐧𝐬 𝐭𝐨 𝐭𝐚𝐤𝐞, 𝐚𝐧𝐝 𝐚 𝐛𝐨𝐧𝐮𝐬 𝐬𝐮𝐫𝐩𝐫𝐢𝐬𝐞—stay tuned! #PowerBI #Microsoft #Copilot #BusinessIntelligence #DataAnalytics #PowerBICommunity

19

𝗡𝗮𝘁𝗶𝘃𝗲 𝗩𝗮𝗿𝗶𝗮𝗻𝗰𝗲 𝗖𝗼𝗹𝘂𝗺𝗻 𝗖𝗵𝗮𝗿𝘁 𝗶𝗻 𝗣𝗼𝘄𝗲𝗿 𝗕𝗜 𝚗̶𝚘̶𝚝̶ 𝗣𝗢𝗦𝗦𝗜𝗕𝗟𝗘 If you check the Power BI Community forum for Column Variance chart implementation, you'll be told to use 𝗰𝘂𝘀𝘁𝗼𝗺 𝘃𝗶𝘀𝘂𝗮𝗹𝘀 and that it's 𝗻𝗼𝘁 possible natively. Here's the thing though - it 𝗮𝗰𝘁𝘂𝗮𝗹𝗹𝘆 𝗶𝘀 𝗽𝗼𝘀𝘀𝗶𝗯𝗹𝗲. Using careful visual context management and some clever tricks, you can create dynamic variance column charts using 𝗼𝗻𝗹𝘆 𝗻𝗮𝘁𝗶𝘃𝗲 𝘃𝗶𝘀𝘂𝗮𝗹𝘀. No custom components needed. 🎬 Video tutorial dropping next week. #PowerBI #DataVisualization #PowerBITips #BusinessIntelligence #DataAnalytics #MicrosoftFabric #DataReporting #Analytics #BI #DataScience #PowerBICommunity

48

8 Comments

Explorations of Copilot for Power BI continue... https://lnkd.in/ecvw3xzQ A lot of people have asked me privately: Has Copilot for Power BI gotten better since I evaluated it back in Autumn 2024? In general, yes. Especially for the experience where you ask data questions about reports and models. There are a lot of additions since September 2024 that have improved this experience and made it more reliable and useful. I'm particularly content with it being an entry point for "Explorations" in the service, and the new UI that explains how Copilot arrived at a result. It also seems to more reliably use the right fields; I notice better results in models that have better setup, overall. I think there's still definitely some rough patches that need ironing, but it's nice to see progress. I think that report summaries and generate reports haven't yet improved much, though, if I'm honest. In a future video I'll re-evaluate each of these experiences, but it's still changing too frequently for this evaluation to be worth my time, right now. In general these videos focus on semantic model improvements to get more from Copilot and Fabric Data Agents (videos coming later on that). I'll also explain how to set up Copilot from scratch once it becomes more widely available with lower SKUs like F2. In general I think people are over-focusing on semantic model improvements, and under-prioritizing user enablement and training for these tools, though. Yes, having a "good" semantic model is key, but if your data culture and adoption isn't in order, then that's probably what you should prioritize. #PowerBI #MicrosoftFabric #Copilot #DeepRockGalactic

62

4 Comments

I've been using AI in my Power BI workflow and it’s saving me a lot of time. Sharing a few tricks that might help you too: 1️⃣ Recurring measures → Don't sweat it. Copy your .tmdl, drop it into Claude, tell it the measures you need, and paste them right back. No more rinse-and-repeat coding. 2️⃣ Wireframing dashboards → Instead of starting from a blank canvas, I just describe the objectives and ask for a visual mock-up. Pro tip: don’t mention “Power BI” — you’ll get more creative layouts. Imagine how easy this makes it collaborate visually with various teams. 3️⃣ About pages that don’t bore → Feed your BIM notes or rough context into AI, have it draft clean HTML, and pop it into the HTML content visual. Now you have a well documented Powerbi page. link your measures in there as well 4️⃣ Field parameters / Switch statements → Let AI build the DAX for you. Pro tip - inform it to increment order by +10 so you’ve got room to expand later. ✨ Bonus use case: I also throw complex DAX back at it and ask for a step-by-step explanation. Sometimes it even suggests simpler alternatives I hadn’t considered. If you'd like me to expand on any of the above just reach out. Would highly also recommend looking at Kurt Buhler content, especially around mcps 💯 Alternatively, how are you weaving AI into your Power BI dev flow? Would love to hear your hacks. #PowerBI #AI #DataAnalytics #Pro

48

6 Comments