Note: I'm not saying it's a bad thing. Just that I would have been _really_ surprised if he stuck around for much longer after lock-in expired. If he did, _that_ would have been news. This was, more or less, to be expected.
I have friends who are on their 3rd iteration of this cycle. :-D
I wouldn't be surprised if, like many acquisitions, once acquired Yammer wasn't just another cog in the machine and being seen as a non-critical part of Microsoft is more like being a cog in the machine than like being a fast moving startup.
With all the corporate realignment, I'm not sure there is much room for any acquired leadership to really make a big impact at Microsoft, but that is more Monday morning quarterbacking than a super meaningful analysis.
It's always interesting to see how these things turn out.
Vs for instance:
he also gave similar talks externally that I'd recommend.
If one was able to predict that, then a build decision would be superior to a buy. Typically, those Corp Dev spreadsheets people work on so much when buying a company assume a certain growth rate for a number of years, and another one after that, etc. but I don't think people plan for the scenario that the "market dies" after acquisition.
It would be kind of like predicting today that Docker-style containers would be abandoned in 2 years. Or that virtual reality would never pan out (Oculus).
The problem is that when markets are hot, people see the lag as insurmountable and deciding for Build looks overwhelming.
So assume every significant business decision carries pretty significant risk and the best you can do is manage that risk better than the next guy. Companies acquire other companies for a variety of reasons, some better than others. There aren't any hard rules either--a right decision in one scenario might be exactly the wrong decision in another. The build vs buy decision is a much more complicated question than it seems at first.
All of that to say, I'd wager acquisition failure rate isn't all that far apart from general business failure rate or even the failure rate of a company doing a project internally vs buying a company.
This doesn't mean that an acquisition is an automatic failure, just that it needs to drive more sales to become a success- either by direct sales or complementing other products. Know how to create a $200m business? Increase Windows sales by 1%. (steveb).
I them see less engineering, more corp Dev driven decisions when the market is hot. If I knew the market was going to cool down, I could instead hire engineers and build a more realistic product that solves the specific problems that would survive the market transition.
I think why ESNs havent been able to make a mark is because they arent designed for individuals or teams but they are designed for the enterprise. This approach of keeping an enterprise in the centre while designing the system doesnt address what an individual needs.
In my experience, individuals work for a team & not enterprise & that according to me is a fundamental reason why the likes of yammer havent been as powerful as ideally they should be.
Infact this problem was the start point for us while designing our tool teamgum - to build a reddit like community within an enterprise.