Microsoft Store ran an experiment on the Surface Book 2 product page. The treatment showed a hero image with fewer, yet larger product photos

Microsoft Store ran an experiment on the Office 365 Home product page. The treatment raised the purchase calls to action higher by removing the hero image.

Microsoft ran an experiment on their Customer Satisfaction Survey at both support.microsoft.com and answers.microsoft.com (Desktop). The treatment contained two icon labels at the opposite sides of the star rating range (ex: Very Dissatisfied and Very Satisfied) - providing it with additional meaning.

Microsoft Skype ran an experiment for the mobile segment of the Skype application with a treatment having combined icons with corresponding labels. The control only showed icons.

In this experiment, the carousel items were increased from 12 to 16.

In 2012 a Microsoft employee working on Bing had an idea about changing the way the search engine displayed ad headlines. Developing it wouldn’t require much effort—just a few days of an engineer’s time—but it was one of hundreds of ideas proposed, and the program managers deemed it a low priority. So it languished for more than six months, until an engineer, who saw that the cost of writing the code for it would be small, launched a simple online controlled experiment—an A/B test—to assess its impact. Within hours the new headline variation was producing abnormally high revenue, triggering a “too good to be true” alert.

HBR, September–October 2017 Issue, https://hbr.org/2017/09/the-surprising-power-of-online-experiments

Note: This experiment was a solid success and replicated multiple times over a period of months. It worked at Bing and had a profound influence. The only reason why we atributed a 0.25 point (a "Maybe") was because we don't have the exact sample size and conversion data.