The number of annual acquisitions in the U.S. ranges from 1,200 to 1,500 and reaches 5,000 worldwide, for a total deal value of around $2 trillion. Corporate acquisitions affect shareholders, employees, customers, suppliers, and the economy at large by influencing market competition and productivity.

What will shock investors is that 70–75% of acquisitions—presumably done for their benefit—fail, according to our rigorous statistical analysis of no less than 40,000 acquisitions worldwide over the past 40 years. Over that period, most acquisitions miserably failed to achieve their stated objectives of enhancing post-acquisition sales growth, cost savings, or maintaining the buyer’s share price.

We are not alone in making this observation. NYU valuation guru Aswath Damodaran once aptly described corporate acquisitions as “the most value-destructive action a company can take.” A study of contested (multi-bidder) acquisitions found that the stocks of those who failed to buy outperformed buyers by 20–25% in the three years post-acquisition.

Moreover, our data show a “reverse learning curve”—an increase over time in the M&A failure rate. Here are the main factors our statistical model indicates that detract from acquisitions’ success. Overcoming these failure triggers will substantially improve the outcomes of M&As.

The urge to merge

Corporate acquisitions, where an outside business is transplanted into the buyer, resemble human organ transplants in their challenges and risks. Doctors will try any available alternative before resorting to transplants. But that’s not what most CEOs do. Facing a sales slowdown, loss of market share, or worsening of the company’s competitive situation, they yield to worried investors’ pressure and follow the advice of commission-hungry investment bankers and advisors by looking for a big transformative acquisition to save the day.

In our research-based book, we elaborate on attractive alternatives to acquisitions, like the development of internal capacity (patents, brands), partnerships, and joint ventures, and we document that their ROI is often higher than that of acquisitions. But in their urge to merge, executives often buy a strategically misfit target, overpay for it, and fail to integrate it properly. Acquisitions should be the last resort, not the first option.

Value-destructive targets

Our statistical model relied on 43 distinct variables to identify several key target attributes that adversely affect acquisition success. Examples:

• Large targets: The integration of a large target into the buyer is particularly difficult and likely to fail since many employees (buyer and target) must be reassigned, lines of control changed, and complicated operating procedures of the partners unified. Large acquisitions often require the buyer to substantially increase debt, which has to be serviced irrespective of the merger consequences, causing the downfall of many acquisitions.

• Conglomerate acquisitions: Business-unrelated acquisitions now constitute almost 40% of all acquisitions. Those make no economic sense: They provide no synergies because the merger partners operate in different industries, and if an investor wants to diversify their securities portfolio, they can just buy different shares. They don’t need company executives to do it for them and pay a large premium to the target’s shareholders. Indeed, most conglomerate acquisitions fail.

• Operationally weak targets: Successful CEOs are under the illusion that they can repeat their past success by resurrecting failing targets. This very rarely happens. More destructive target attributes are discussed and demonstrated in our book.

Misaligned executive incentives

Many companies pay their CEOs a substantial acquisition bonus just for completing the deal. Executives’ annual compensation also usually increases after acquisitions, since company size is a major determinant of executive pay. And if that’s not enough, we empirically show that serial acquirers serve, on average, four-to-five-year-longer tenure as CEOs than leaders who acquire a few firms. Acquisitions appear in many cases to be tenure insurance for CEOs.

Note what’s fundamentally wrong with these arrangements: the emphasis on a deal’s completion rather than on its success. And with an acquisition failure rate at 70–75%, the difference between completion and success is huge. We also show that the penalty for acquisition failure only results, in many cases, in a slap on the CEO’s wrist. Shifting CEO acquisition incentives from acquisition completion to success is key to improving the consequences of M&As.

The consequences of corporate acquisitions affect peoples’ lives, the state of the economy, and investors’ wealth. The current state of 70–75% acquisition failure is intolerable. Executives must be more steadfast in their diligence and research before signing a deal to start reversing the widespread M&A failure across essentially every industry vertical and sector.

More must-read commentary published by Fortune:

• Former Intel CEO Craig Barrett: Splitting up America’s leading chipmaker is a bad idea
• Americans paid $100B since 2008 to access their own money. I am petitioning the Fed to end this racket
• British fintech founder who moved to U.S.: U.K. tech’s problem isn’t taxes—it’s ambition
• As activist Starboard engages constructively, here’s a potent prescription for Pfizer’s future success under Dr. Bourla’s watch

The opinions expressed in Fortune.com commentary pieces are solely the views of their authors and do not necessarily reflect the opinions and beliefs of Fortune.