Posted on: April 20, 2017
Hello, and welcome back! In this three-part series, we are putting the spotlight on Microsoft Power BI – what it is, how it works, and how you can use it to your advantage. In our last post, we got the ball rolling by covering the three basic building blocks of Power BI: datasets, dashboards, and reports. This week, we are diving a little bit deeper and taking a look at how we can garner insights from data by asking questions using natural language.
Part 2 – Power BI Q&A That We All Need
If you know what your question is, but you’re not quite sure how to go about finding the answer on your own, Microsoft Power BI’s Q&A box is the perfect tool for you. Sometimes, it’s just faster and easier to get answers from your data by asking questions using your own natural language. For example, you might wonder “What were our total sales last year?” By inputting this query into the Q&A box, Power BI can do the heavy lifting and provide insights and answers from the data within its stores. It’s a fast and interactive experience that you’ll come to love and, eventually, even rely on.
How It Works
Asking a direct question of your computer datasets might seem a little bit far-fetched, but it works. So, how does it know how to answer your questions? The Q&A box is formulated to rely on the name of various columns, tables, and other calculated fields within the underlying dataset. In other words, the names that you assign to specific datasets is really important! Creating a misleading label for a dataset could cause Q&A to overlook critical information when answering questions, or even to provide you with weird answers that don’t make a whole lot of sense, so take care here! In addition to names and titles, Q&A also answers questions based on the way that your dataset is organized so that it covers all bases and provides the most accurate information possible with all of the data that is available to it.
A Helping Hand
Q&A is designed to help you phrase the question because it wants to ensure that it provides accurate answers. It does this in a number of ways. As you type your question into the Q&A box, it will auto-complete words and questions in order to speed things up or help you use the most accurate phraseology when forming the question. Q&A can also correct spelling, auto-suggest replacement terms, and provide a preview of your answer in the form of a visualization.
When you’re using the Q&A box feature, it will automatically select the best visualization that it comes up with, based on the data being displayed. In some instances, data in the underlying dataset will be defined as a specific type/category, and this will give Q&A clues about how it should be displayed. As an example, data that has been categorized as a city would most likely be given in the form of a map. Of course, you don’t have to settle for the visualization that Q&A selects for you. You always have the option to specifically identify which type of visualization you prefer by including it in your question. For example, you can request that the answer be displayed as a pie chart, bar chart, etc.
Q&A has been programmed with a pretty extensive vocabulary, but its list of keywords isn’t exhaustive. If you type a word in the question box and the word or term is grayed out, it means that Power BI doesn’t recognize it or simply doesn’t recognize it within that context. This will help you to reform the question.
Keep in mind, too, that you don’t have to limit yourself to one dataset when asking questions. Feel free to make inquiries that will retrieve information from more than one dataset so that you can get the most robust answers.
Knowing the building blocks of Power BI and how to ask questions and get answers will get you well on your way to taking advantage of all Power BI has to offer, but you’re not done yet. Stop by for the third and final part of our series in order to pick up tips on how to get the most out of Power BI for your business. Until then, the experts at Uncommon Solutions are happy to answer any of your questions.