:: By Glenys Salas, Ai Media Group ::
Business Intelligence (BI) and its mathematical cousin Business Analytics (BA) are the hottest words on the street these days. Along with the buzz, a veritable garden of self-service business intelligence software applications have blossomed overnight, promising to end our addiction to boring spreadsheets in favor of interactive canvases that “bring data to life.”
Burning with report envy because Facebook keeps showing you ads with hipster business managers presenting Picasso-like visualizations? Vaguely suspicious that it’s all just hype from overfunded start-ups? Too busy to try out half a dozen different software applications? Here’s a helpful guide to our three favorite Business Intelligence software applications.
You didn’t misread that. Although it’s never been a particularly sexy software, Microsoft Office’s Excel application is the oldest business intelligence tool still in use today. The advantages of Excel include nearly universal adoption, flexibility, a rich portfolio of mathematical functions, and all the basic charts and graphs.
Since Excel has been on the market for so many years, there is also a wealth of information available in online forums around the Web. In 75 percent of cases, Excel is the simplest and fastest tool to mine data and find answers. If most of your efforts involve one-off calculations or the odd chart for your boss, you can pretty much stop reading this article now.
If you’re still with me you may be facing one or more of the following circumstances where Excel fails as an optimal business intelligence tool:
· Lots of data coming in from external sources or different databases
· Existing internal reporting tools are too broad to be efficient or helpful
· Complex budgets that have to be managed daily or weekly
· Massive Excel Spreadsheet with VLOOKUP/SUMIF/COUNTIF/Reference Tabs/Multiple Pivot Tables
· You are constantly re-selecting the source data for charts in your current reports
· You are spending too much time managing a spreadsheet and no time doing your job
There are dozens of business intelligence options out there that can help you out with these issues, but to save you time we’ve selected our two favorites for more complex needs.
Watson analytics is by far the easiest and most enjoyable business intelligence software. It’s a “freemium” model that allows you to upload up a maximum 500MB of data, 100k rows, or 50k columns without paying a penny. The software works best for data with lots of columns containing qualitative information. Some examples include survey data, customer relationship management (CRM( databases, or customer marketing lists. There is almost no set-up time and the software even makes recommendations for the best charts and graphs.
Since Watson Analytics is powered by some of the same text processing software that beat human beings at Jeopardy, make sure your data points and column headers are in natural language with no abbreviations to get the best performance. The main drawback is that the data you upload has to be contained in a single table.
Power Bi is Microsoft’s contribution to the business intelligence software game. Your online account is free for up to a gig of storage and the downloadable desktop application is completely free for unlimited amounts of data, although you do lose some of the handy sharing and community features of the online application.
Power Bi is best described as a combination of Excel and Access (or any other structured database software). Data is loaded in as tables, which can be linked in a relationship view, just like any other structured database. You can look at the physical tables in the tables view to vet the data and set up your source data in whatever structured format makes you the happiest. A fully normalized structured database gets you the best bang for interactivity in the visualizations, but may cost you something in performance.
Charts and graphs are as easy as simply dragging the columns of info you need from the Table menu onto the canvas view and clicking around in your options. Use your keys or dimensions for filters and every chart on the canvas adjusts at a click of the mouse. For visualizations beyond the ones that come with the tool, Power Bi has made its software open source so any programmer can build a custom visualization. Updates are rolled out every couple of months, but they are not fully debugged, so save often!
As a light user, the drag-and-drop interface and pop-ups will be more than sufficient to get you something useful and fun to use. If you want to go deep, get ready to learn not one, but two programming languages. The tables and formulas in the data model are written in DAX (very similar to the formula language in Excel) while the queries are written in M Power Query.
Take the time to work your way through the video and content in the “Learn” section of the Power Bi site. If you have some extra time, we also recommend taking the edx.org online Intro to Power Bi class.
Finding the right business intelligence tool is not a simple process and there are literally hundreds out there with varying degrees of complexity. By sharing three tools in continuous use in our office we hope to save a little time and even head you in the right direction along your hunt. It really depends on what criteria are important to you, such as exploration versus analysis and learning curves so first spend some time determining what is important to your needs.
Glenys Salas serves as senior media analyst at Ai Media Group where she interprets and translates large quantities of data into information people can understand.