Uncovering Microsoft Office Labs

Earlier this year, I covered a new tool out of the somewhat mysterious Microsoft Office Labs called Ribbon Hero. Developed with input from Microsoft's XBox group, the Facebook-based social learning game represented a really novel approach to the otherwise dry topic of application training. On Wednesday, I had the opportunity to follow up with Chris Pratley, General Manager of the Office Labs group at Microsoft and uncover a bit more of what they do.

Office Labs isn't like Google Labs. Their products aren't floating around on the Internet for months before Microsoft decides if they're worth keeping. Rather, the group is a response to what Mr. Pratley called the "monolithic" Office releases that make their way out of Microsoft every 2-3 years. Office Labs looks at incremental improvements, innovative features, and all the brainstormed "silly ideas" (another Pratley term) that just might make the next version of Office worth the price of the upgrade.

Mr. Pratley has been with the group for 3 years now and described Office Labs as the people who stick their necks out with crazy ideas which, if they prove useful, are embraced by the Office team as ideas they had all along and, if they prove to be flops, lose Microsoft nothing in terms of credibility. While his tongue was firmly planted in his cheek, the idea of a lab setting where countless ideas can be tried on a small scale with minimal investment (and occasionally that creates gems like Ribbon Hero) in the context of a very large organization like Microsoft has a lot of appeal.

Go to next page to hear about the group's living laboratory »

The large organization that is Microsoft also gives Office Labs a living laboratory in which to test their ideas. They develop many products for specific workgroups to test internally. For example, they might develop something that works well for their legal department or sales force; if it works for Microsoft's sales force, it stands to reason that it would add value to Office for another company's sales force.

Ultimately, as Pratley said, it's a cheap way to try out lots of ideas. To cut through internal politics, resistance to change, and to make sure that ideas that seem great to geeky engineers actually have broad appeal, the group collects incredible sums of data. These data then drive more widespread testing, modification, or termination of the various ideas the group generates.

Much of this testing is not visible to the public. The least useful ideas never make it out of Office Labs; more promising concepts go into wider testing within Microsoft and the best ideas head into beta and release candidates. Many of those ideas that head outside the corporate walls require social components or simply large numbers to test effectively. While Ribbon Hero is actually available in a corporate version without Facebook integration, for example, serious testing of social learning models really needs to happen out in the wild.

The group also used this approach with "Community Clips," or community-generated how-to videos for various Office features (a project a couple years ago before people used the word "community" in every other sentence). The group assessed whether they could actually get people onboard (YouTube-style videos about Microsoft Office probably won't garner the same attention as skateboarding wipeouts). They also needed to develop methods for far stricter content monitoring than was allowed on many social video sharing sites. They had to protect the Microsoft brand, after all (sorry, couldn't resist at least one little dig; this article has been entirely too positive from a borderline open-source and Google fanboi, I'm afraid).

Go to next page to hear about Labs' data-driven approaches and viral marketing research »

Of course, I'm also afraid that I was completely impressed by their data-driven approach to developing, deploying, and evaluating innovation. Ribbon Hero, for example, was somewhat controversial internally with gamers at Microsoft unable to see any fun or draw and Office team members unable to see the value or feasibility of a gaming approach to training. Data from internal and external testing, however, led the Labs team to move forward. And yet, Ribbon Hero answered a couple of important questions with significant marketing and sales implications:

• Why should I bother with an upgrade? The current version already does 90% more than what I need to do.
• So many useful features exist within Office - how can Microsoft get the ideas out to users?

Ribbon Hero has turned out to be fairly successful, despite the internal hurdles. Downloads remain at 20% of those on launch day; similar tools, games, and utilities see a much sharper drop off. The group is looking at ways to make it more like many of the games that already exist on Facebook as well, making the challenges more open-ended and encouraging more direct competition. They are also looking to expand utilization of Office products like OneNote, using Ribbon Hero to drive users to explore beyond Word, Excel, and PowerPoint.

Perhaps more interesting is the group's research on viral marketing and utilization of the tool. Early data from their internal tests of Ribbon Hero suggest that,

• Lots of people run it once
• Some people run it a couple times and then abandon it
• A small number of competitive, game-oriented users actually go for it and work through the various challenges and levels
• Then this middle group appears to pick it back up and use it more, apparently from the viral effect of the more active users sharing their results and encouraging usage.

The group's goal is to get usage to extend beyond internet geeks and to capture bored secretaries or office staff who suddenly have a workplace-appropriate social game that they can use to learn productivity features of new software.

I have to say that if I had to work for Microsoft, I'd want it to be in the Office Labs group. Their data-driven approach to improving Microsoft's cash cow seems to be a winning formula. Now if only there was an Internet Explorer Labs group.