Upon announcing 7,800 job cuts at Nokia (”NOK”) in July last year, Microsoft wrote off the entire investment of its acquired mobile unit. And Skype and Yammer haven’t been particularly rewarding either. But Microsoft is hardly alone. In February, Yahoo (”YHOO”) marked down $230 million of its acquired microblogging site, Tumblr. Along with BrightRoll and all other acquisitions since 2012, Yahoo recorded a $1.2 billion impairment charge in the fourth quarter of 2015. Even Google (”GOOG”) had to fire-sale Motorola at $2.9 billion, which it bought for $12.5 billion back in 2012.

If that’s not grim enough, this abysmal track record is confirmed by nearly all studies: 70% to 90% of M&As fail. Why is this the case?

There are bright spots, to be sure, although they’re few and far between. For example, Google’s Android is a poster child. Skeptics scoffed when Android was bought for a meager $50 million in 2005, seeing no chance that Google could successfully break into the mobile space dominated by Apple (“AAPL”), Nokia, and Blackberry (”BBRY”). In less than three years, however, the first Android phone was marketed by T-Mobile, thousands of new apps were being written, and third-party phone makers such as Samsung and HTC were coming on board. The fact that Android is a free system for anyone who wants to use it turns out to be consistent with Google’s overarching strategy: We don’t charge for usage; we make money from advertisements.

If any safe play exists in M&As, it’s where a parent company lets the newly acquired entity leverage the existing model to turbocharge growth. When Spinbrush was acquired, it gained immediate access to the distribution channels that P&G (”PG”) had nurtured over the years. When VMware (”VMW”) was acquired, it tapped in to a long list of existing EMC (”EMC”) customers. Few changes with respect to strategy or operating models are required on either side. Synergy is immediate and apparent.

This is why when Pepsi-Cola acquired Frito-Lay, it benefited from the direct store delivery logistics system that PepsiCo had honed over the years. In contrast, when Quaker Oats was acquired, managers painfully discovered that its traditional warehouse delivery method had very little in common with that of PepsiCo’s, and consequently, the acquisition failed to meet the financial expectation.

What all of this means is that Microsoft is facing some long odds with its current purchase of LinkedIn.

The deal size is so big that Microsoft, the world’s third largest technology firm, will need to issue new debt to fund the acquisition. Buying LinkedIn is not driven by industry consolidation, and in that sense, there isn’t any redundancy to cut that can realize quick savings. With the acquisition, Microsoft will try to use LinkedIn to create new services and yet-to-be-seen offerings. All of this implies that the sort of synergy it is seeking is neither apparent nor immediate.

To be fair, there are lots of arguments in favor of the merger. Nadella even offered an example of a busy executive walking into a business meeting scheduled on a Microsoft Outlook calendar fully integrated with LinkedIn functionalities. One can receive a notification if the other party went to the same university, enabling a whole new level of human connection. But couldn’t this have been done through some form of strategic alliance or partnership? A $26 billion payout for LinkedIn seems outrageously generous, which is great news for LinkedIn, but not so great for Microsoft. That’s why when the news broke, LinkedIn’s stock price shot up by 47%, and Microsoft’s went down by 2%.

While it will be interesting to see what this integration brings, at least for now, it seems the odds are against this merger.

Howard Yu is professor of strategy and innovation at IMD. In 2015, Yu was featured in Poets & Quants as the World’s Best 40 Under 40 Business School Professors. He received his doctoral degree at Harvard Business School.