UPDATE: Here is the link to the presentation on channel 9
If you haven’t realized that Microsoft really likes Yammer now, you need to come out from under your rock. Yammer integration in the next-gen O365 is, we’ll say, thorough. That’s great and all, but how does an organization use that to its advantage analytically? Richard diZerega, a Microsoft Architect, told us exactly how to do just that. The good news is that this is really not too difficult. To be fair though, it’s easy because Richard took the time to build a bunch of tools that are available for free.
So here we go. This is a simple three-step process. First, you export the data out of Yammer, then massage the data (this is the longest step), and finally, you import it into analytic and display tools. The tools used were: Power BI in Excel, Power Query (formerly Data Explorer), PowerPivot, Power View, and Power Map. With all that power, it must be easy, right?
The export is only available to network admins, so that’s your first hurdle. Once you’re an admin, you go into your Yammer settings and there is an export option. Easy. That export will give you .csv files with admins, files, groups, messages, networks, pages, topics, and users. What that export does not get you are: mentions, following, sharing, detailed dates, or user demographics. These are all pulled out using those tools Richard made, available here. So the cleanup process starts with viewing the data in Excel. There you can take out columns you don’t need, remove duplicates, and remove blank rows. Power Query, an add-on for Excel, helps with the cleanup. If your organization is big, these files can be rather extensive so cleaning them up goes a long way in speeding up the rest of the process. Do this for all the lists you want to use. Then with PowerPivot, you can load the lists and literally draw lines between columns that are associated across the lists. Things like user ID should be mapped this way so that the analytics can track individual people across dates, messages, interactions, etc. Once you have everything connected the way you want, you can start looking at it in very interesting ways. O365 has a Power BI Q&A function that lets you ask questions against the data you mined. It will then display that data in the best visual based on the requested data. If you ask for geographic data, it will show you a map with that data laid out, if you ask for a particular fact it will display that value, and so on. The catch here is that it does this only one question at a time. If you want to make a presentation with multiple visuals you need to use Power View. This gives you those same data views but allows you to compile them to show at meetings, or print and hang on your wall if you’re into data that way. Power Map is similar, but with displays on maps. Both of these viewing tools allow you to make videos. You can set up a view and press play to see an animation of the data over time. This was really cool to see. Graphs change to show trending content and maps change to show topics across geographic areas. As a marketing professional, I was very impressed here and immediately saw the business value and use cases for these functions. Richard also made two custom viewing apps, Nucleus and Yamosphere. These are just plain fun to play with and deliver seriously useful insights. Look for them in the Office Store soon (they’ll be free).
If you want more detailed information and haven’t clicked the link up there, do that.