One of the nice benefits of using a cloud based BI service such as Power BI, is the ability to easily share content with users outside the organisation. This is very different than the old school BI solutions that are installed on-premise, and require access to your network. Another Power BI feature is RLS – Row-Level Security, which allows the publisher to share subsets of the data with different users. For example, allow sales representatives in Canada to see only Canadian sales data. In this article, I’ll talk about using both features together- sharing a subset of the data (using RLS) with users outside the organisation. You would think all you need to do is just use both features, however my…

I’ve just spent several hours fighting a piece of software almost as old as I am, in Internet Explorer. (“almost” being 20 years younger than me, which is still pretty old for a browser). When I finally got the issue fixed, I was irate, because I’ve actually had the same battle in the past, but forgot to document it, so here I am, documenting it. I hope it helps my readers, and I’m sure it will help me. Internet Explorer, out of the box on a new machine is extremely annoying, because the default configurations usually do not allow you to do anything. That’s why I usually hurry into downloading Chrome, but sometimes this is not allowed as well (hint:…

The date dimension table, a basic ingredient in any BI dataset, enables us to enrich a date with many attributes such as day of week, workday/weekend/holiday, financial year, and the list goes on, depending on your needs (Here’s an previous article on its importance). This is why, in 99% of my Power BI datasets, there is a date dimension connected to the fact tables. Whenever I need to filter by date or present a filter/slicer to my users, I use the date field in the date dimension, instead of a date field in a fact table. I’m sure this practice isn’t new to anyone who’s been building BI systems. However, there’s one annoying Power BI behaviour which I’ve been wrestling…

The Power BI Desktop September version is loaded with many features I find useful. They aren’t perfect but a very important step in the right direction to becoming a mature BI product which is not only easy to use and cheap, but also full of important features. Here are the main additions in my opinion: Drillthrough to another report page. It’s a basic feature in other products, which has finally arrived to Power BI. It allows you to define a page as one that other pages can drill through to it, passing a filter. A good example is a customer page, where you show details of a specific customer. Then, from a sales page, you can click on a transaction…

Your Power BI reports and dashboards likely rely on an underlying data source being up to date and following some business rules. For instance, It may be a data warehouse schema which gets updated once a day at 3:00 am, collecting data from all 314 branches of your bank. Your users want to be sure that the information they see is correct and up to date, so you want to be able to respond to any data issue as quickly as possible. You want to be sure that all information from all 314 branches has been received and processed at 3:00, and if there’s an issue, you need to know about it ASAP. There are many monitoring systems that can…

Power BI Desktop gets updated once a month, and every month I keep my fingers crossed, hoping for items on my wishlist to be added, mainly the ability to save filters which I cannot understand how it can be missing from one of the market leaders product. Seems like I’ll need to keep my fingers crossed for another month, but it’s not a disappointing release at all. The biggest news as far as I’m concerned are the “what-if” parameters. It’s a simple way to let the user play with parameters and see how his selections change existing data. So, for example we can let our customers see how a sales report would look like if prices go up by X…

When working with Power BI desktop, you can define whatever date format you want to use for each date field in your model. That works well on Power BI desktop, and on the Power BI service (the browser). However, there’s one annoying case where your format is ignored, and I’m here to help with that, after a recent tough struggle. That case is when you use a date field in a slicer, shown as a slider: The date format used in this case has nothing to do with what you may have defined as the format of the field. It will show as dd/mm/yyyy or mm/dd/yyyy and that’s usually OK, but which one will it be? dd/mm or mm/dd ?…