Come microservice, come cloud, the line-of-business application is always going to be with us. We need to know how our businesses are working: how much we’re selling, how much we’re buying, how our customers are feeling, and every one of a thousand little markers that show the pulse of business.

And that leaves us with the ever-present question: How do we show that information? That’s where business intelligence tools come in, to let us ask questions and get answers, exploring the ever-growing pool of business data we’re storing in our myriad business systems.

Microsoft’s tool for that business data exploration is Power BI, even if you didn’t realize that.

After all, Power BI began life as a set of extra query tools for Excel, taking tabular data and helping you slice and dice it before delivering well-formatted graphical answers. It could connect to SQL Server’s OLAP tool, acting as a familiar (if not user-friendly) user interface to billions of rows of data.

if you like. Power BI’s machine learning-powered features also help identify interesting data, guide visualizations of trends, and point out outliers in your data.

Power BI brings the desktop and cloud together

Power BI is an interesting mix of software-as-a-service and desktop app. You can use the desktop app to explore your data and build reports; but it’s the combination of app and cloud service that really makes sense.

It’s best to think of the desktop Power BI app as a way to get initial insights, where you find out the right questions to ask. You can turn the resulting report into a live dashboard on the cloud service, connected to live data and giving near-real-time analytics of trends and indicators.

One of Power BI’s more useful features is the sheer number of data sources it can use: from traditional databases and spreadsheets to modern SaaS services like Zendesk and Github. Business data now resides in so many different places that it’s difficult to be sure what’s relevant, so it’s good to be able to query as many of the repositories as possible. You can even connect Power BI to the , using it to display complex statistical information on a range of charts.

, which builds on existing JavaScript skills with added type protections; styles are added via CSS. It’s a good idea to start any new visualization with static data, to see if it renders the way you want. Once it’s working as expected, add data bindings to live data. When you’ve tested your code, use the command-line tools to package it for sharing.

Microsoft offers Power BI on a subscription model, with access to most of its features available to Office 365 enterprise subscribers. A new premium tier adds new ways to share reports, with the option of embedding your data in apps and sharing it with an unlimited number of viewers. Experienced business analysts and data scientists build interactive reports, and users then drill down into data on their smartphones or PCs. All you need to do is publish a report (using Power BI’s mobile view tools) and then embed it as a web view in an app. Apps can refresh their data sets up to 48 times a day.