Frederic Van Kelecom

Just in: I've been recognized as a Top Business Intelligence Voice! 🏆  Turns out, translating data into decisions isn't just a job—it's an art.  The future of BI is bright. Let's shape it together.  #BusinessIntelligence #DataDriven

25

1 Comment

Avoid these 7 Common Business-related Mistakes On Data Projects https://lnkd.in/eV_WGAbh #kamphuisgroep #caronne # #ArtificialIntelligence #MachineLearning #bigdata #businessintelligence

3

𝗗𝗮𝘁𝗮𝗯𝗿𝗶𝗰𝗸𝘀 𝘃𝘀. 𝗙𝗮𝗯𝗿𝗶𝗰? This is a question that pops up often on my LinkedIn feed. It almost seems like you have to choose one or the other. But do you really need to? At Databricks User Group meetup last Tuesday in Brussels, Ivana Pejeva and Wesley Backelant shared why using them together could actually be the better approach. Combining Microsoft Fabric with Databricks can be a good solution if your team already uses Databricks but heavily relies on Power BI dashboards for visualization. It's actually very simple to use data stored in Databricks in Fabric with Unity Catalog mirroring. Instead of moving or copying data, Fabric mirrors the catalog structure and uses shortcuts to access the underlying data. This means that changes in Databricks show up directly in Fabric. From there, you can use #PowerBI in Direct Lake Mode to build reports directly on the #Azure Databricks items. It’s similar to Import Mode because data is loaded into memory, but Direct Lake mode has the advantage that it is more efficient to refresh. Do you have a preference or do you also combine both?

394

22 Comments

You have your KPI metadata sorted like a pro to scale your Tableau #KPI analytics? Then this is for you: #PowerKPIs Metadata Formatting ❤️ Excited to share a new feature Merlijn Buit and I have been working on this week and that will be available soon. It's easy like that: 1️⃣ Add your metadata (e. g. number formats, decimals, chart types) to the view 2️⃣ Map your metadata fields to the KPI attributes 3️⃣ Save and enjoy! Infotopics | Apps for Tableau Richard Van Wijk Siebe Grijpma

576

25 Comments

Two weeks ago, I attended the Databricks World Tour in Amsterdam. I gained some interesting perspectives on the current opportunities and challenges in data and AI. Some of my key takeaways are: 𝐆𝐞𝐧𝐞𝐫𝐚𝐭𝐢𝐯𝐞 𝐀𝐈 𝐢𝐬 𝐞𝐯𝐞𝐫𝐲𝐰𝐡𝐞𝐫𝐞—𝐛𝐮𝐭 𝐩𝐫𝐨𝐝𝐮𝐜𝐭𝐢𝐳𝐢𝐧𝐠 𝐢𝐭 𝐢𝐬 𝐚 𝐜𝐡𝐚𝐥𝐥𝐞𝐧𝐠𝐞.  Turning PoCs into scalable, production-ready solutions takes more than just great models. 𝗖𝗲𝗻𝘁𝗿𝗮𝗹𝗶𝘇𝗮𝘁𝗶𝗼𝗻 𝘃𝘀 𝗱𝗲𝗰𝗲𝗻𝘁𝗿𝗮𝗹𝗶𝘇𝗮𝘁𝗶𝗼𝗻. Finding the right balance between centralized platforms for governance and decentralized systems for flexibility is key. 𝗔 𝘀𝘁𝗿𝗼𝗻𝗴 𝗱𝗮𝘁𝗮 𝗽𝗹𝗮𝘁𝗳𝗼𝗿𝗺 𝗶𝘀 𝘁𝗵𝗲 𝗳𝗼𝘂𝗻𝗱𝗮𝘁𝗶𝗼𝗻. It's essential to tackle the current challenges and enable innovation, while maintaining control and scalability.   It’s clear that while tools are advancing rapidly, having a strong technological backbone and a clear strategic vision remains essential. Thanks for the great event!

19

Summer day with lots of sun, beach, refreshing drinks and a book which was already long on my reading list. Finally was able to read “Data management at scale” from Piethein Strengholt. Create read if you are interested in data management, data platforms, data governance, data products, etc. Or if your organisation is looking on becoming data driven. Doesn’t matter if you are a small company or a large enterprise the book covers all aspects of setting op a data platform. Got some nice insights like: - start from your organisation capabilities - how to create context boundaries and why they are important - architecture choices to go for centralised or decentralised - importance of master data management - data products vs domain data stores - governance models - etc. But most important, it’s a journey. You will never get their without some trail and error. Let’s put these new insights into practice for some customers that are also making the shift of becoming data driven. #data #datamanagement #datagovernance #dataplatform #azure #azurefabric

71

4 Comments

Don't you hate it when you can't find what you need? Last week at Datafam Europe I demoed Semantic Search for Tableau Server. A way to improve and extend the capabilities on finding relevant content on your Tableau Server / Cloud. This is AI technology I've been able to build with the Tableau platform, part of an ongoing project between Tableau developers and community members named Tableau LangChain led by Joe Constantino. Want to see more of this? Drop me a comment of any problems you'd like to solve with Tableau + AI. I'm excited to find new solution and develop some very in demand skills with the #Tableau platform. If you're interested in seeing how you can combine AI functionality with Tableau, check out my blog post #datafam https://lnkd.in/exFyEeqW

27

2 Comments

Databricks, Amazon, Tony Wills, Patrick Bayne, Well Good Job guys, Databricks and Amazon working together (again). This time, Databricks buys Chips from Amazon. Therefore, we dont need NVIDIA. You saw what happened when Apple slowly got rid of its third party suppliers like Qualcomm, Broadcom, etc. Now you're going to see that for #AI. I said this before. All of you need to start learning Databricks like you used to learn Tableau

5

My submission for the #mavenhospitalchallenge from Maven Analytics. Thank you Marjolein Opsteegh for saying "hup, hup, je kunt het". The basic thought was high level executive means 1 screen, as little filtering as possible. Some way to dive a little deeper, hence the 3 buttons to go a few periods back in time, it filters the page,. A filter on the filterpanel sets the year to 2022, in terms of completed months jan. 2022 was the last. Somebody, let's call him Jasmin Jusufbegovic, once said "I do not understand why people don't do the calculations in Excel and create a Powerpoint instead of all the fuzz in PowerBI". It is not a Powerpoint but I tried to create a nice summary with some insights apart from the dashboard. I believe insights should not be on dashboards. Dashboards are meant to stay for a few years, the world and insights can change a lot in these same years. #powerbi #data #analysis

91

24 Comments

Recently, I finished reading 'Data Management at Scale, 2nd Edition' by Piethein Strengholt. It deconstructs established thoughts, adds new perspectives and guides through the journey from data-driven aspirations to a fully realized data-driven culture. A great read for those looking for a Decentralized approach for managing data at scale using Centralized alignment of organization, governance, technology, and architecture. Personally, I benefitted immensely from the Data Domain and Data Product topics with a focus on practical strategies for modern data engineering pipelines. https://lnkd.in/eFVgWt9V

24

It’s not every day that a tech giant like Databricks highlights the work of one of our experts in an article. Quite proud of our colleague and Databricks specialist Bruno Magri for his impressive work with Volvo. 🔎Check the comments for the full article #databricks #Dataengineering #Bigdata #realtimedata

24

3 Comments

Join Databricks on November 7th for Data + AI World Tour Munich! Discover how leading companies like Fraport AG, Deichmann SE, Bayer AG, and CARIAD SE (A Volkswagen Group Company) are taking control of their data and building custom AI on the Databricks Data Intelligence Platform. Register now!

15

👩 A new customer is starting its journey on Databricks today. They need 👉 👉👉 YOU to join DATAPAO, get onboarded and start YOUR #Databricks journey. 🚀 Databricks will be the default data platform in five years, and you will be a Databricks champion by then. Shouldn't you start today? #databricks #azure

25

2 Comments

Demystifying ARIMA for Time Series Forecasting! Ever wondered how we predict future trends in data, like sales, stock prices, or even temperatures? Let’s talk about ARIMA, a powerful tool in the world of data science. ARIMA stands for AutoRegressive Integrated Moving Average. Sounds complex? Let’s break it down: AutoRegressive (AR): Look at past values to predict future ones. For instance, if sales increased steadily over the last few months, they might continue to do so. Integrated (I): Focus on changes over time. Instead of just looking at raw sales numbers, see how they’ve changed each month to identify trends. Moving Average (MA): Adjust predictions based on past mistakes. If your previous forecasts were always a bit too high, tweak them accordingly. Example: Imagine you’re predicting next month’s temperature: AR: If the temperature increased by 2°F each day for the last few days, it might continue this trend. I: Look at the daily changes (+2°F, +2°F) to understand the pattern. MA: If past predictions were usually 1°F too high, adjust by subtracting 1°F. So, ARIMA helps you make more accurate predictions by learning from the past and correcting errors. #DataScience #ARIMA #Forecasting #MachineLearning #TimeSeriesAnalysis

3

2 Comments

(𝗦𝗽𝗼𝗶𝗹𝗲𝗿 𝗮𝗹𝗲𝗿𝘁) Successfully delivering data projects is not your #1 KPI. That is actually the 3rd most important. The second? Your number of failed MVPs. Now, why do I say this? Let’s use Customer 360 as an example, since often both finance and sales departments struggle with a full view of the customer. So that gets the master data management pillar off to a great start. Yet, more often than not, these fail. The standard approach is to start the technology build out which takes roughly 2 years. In the end, it rarely delivers as promised. A better approach? - Start with a minimal viable product - a specific business value you can deliver within three to six months. - Mature your MVP with a series of monthly iterations. Now here is the differentiator between a generic data culture and an innovative one. Every iteration will have wins and failures - and in those failures lies the path to your next big breakthrough. Be open to burning your fingers when you deploy and iterate so rapidly. Imagine a chef. When they start cooking, they burn their fingers all the time. The sauces don't come out perfectly the first time they've tried it. Have you tried making a sauce, especially a French one? They never come right the first time but eventually they do, and it’s magic. Create a culture in your team where every single delivery is analyzed both for what worked, but, more importantly, for things that didn't - setting the basis for your next iteration. At the start I shared the second and third most important KPI with you. What’s the first one? The number of actions you and your team have taken outside your comfort zone - actions that you haven’t tried before. This is the key to unleash breakthrough innovation.

51

4 Comments

Great to see Qlik investing in GenAI. Self-serve data/analytics/bi has been the narrative for a while now, but skillset hurdles still exist. LLMs on top of data feels like the natural lowering of the bar. Combine increased speed and quality of output, alongside improved data literacy, and this could really be something. Looking forward to seeing the roadmap. #QlikAnswers #AI #GenAI #Qlik #data #bi #analytics #QlikLuminary

17

To anyone in my data/analytics engineering network: Can anyone recommend me an article/blog/guide/video/github project..etc) about how to build a data warehouse using Airflow the same modular way of dbt? meaning how to structure an Airflow project (config, dags, tasks) similar to the dbt modular approach (staging, intermediate, facts & dimensions)? I know that I can build the data warehouse using dbt Core and orchestrate using Airflow, but currently the stack doesn't include dbt Core/Cloud. #data_engineering #analytics_engineering #dbt #apache_airflow

18

23 Comments

"90% of the data in the world has been created in the last 2 years." — Forbes And most of it is time-series data. What’s the #1 mistake I see? People default to a basic line chart to show trends. 📉 But that’s like bringing a butter knife to a sword fight… You’re leaving a TON of insight on the table. In this video, I break down 15 different ways to visualize your time series data, each with a specific use case to supercharge your analysis. 🚀 🔍 What You’ll Learn: 👉 How to build 15 unique time series visualizations. 👉 When and where to use each one. 👉 Free workbook included for hands-on practice. If you want to make your insights unmissable, skip the boring line charts and see what’s really possible. 📊💡 👇 Watch now and level up your data game! #DataVisualization #TimeSeries #Tableau #BusinessIntelligence

20