There’s an old saying in journalism: “If your mother tells you she loves you, check it out.” In other words, don’t believe anything you hear until you go to a second source and check your facts before publication.

Even if you think you’re working with reliable sources.

In 1998, I was doing research for my first book, Counting What Counts, which I co-authored with Marc Epstein, an expert in corporate accountability, formerly at Harvard and now at Rice University. As a journalist, one of the things I brought to the co-authorship was first-hand reporting on companies managed by leading executives.

In writing a chapter on how CEOs could better run their boards, I pursued Dennis Kozlowski, the CEO of Tyco International, headquartered not far from where I live in New Hampshire. I picked Kozlowski as a candidate for a governance role model after The Wall Street Journal cited him for his state-of-the-art governance at Tyco.

It was a short drive to Kozlowski’s office, and I arranged to interview him in person. In front of a glowing fireplace—real logs!—Kozlowski talked with pride about governance. He said he valued an active board, and he had in place all the leading governance structures of the time—a “lead director,” annual voting for board members, rotating committee chairs, and so on.

To check Kozlowski’s story, I interviewed Kozlowski’s lead director, an independent member of the board. This director echoed everything Kozlowski said. He was proud to cite the same leading-edge practices. I was pleased. Everything about Tyco seemed to add up.

But of course—and you probably recall the story from here—Kozlowski was accused of looting the company for $135 million. True, he had built Tyco into a giant conglomerate, delighting shareholders with a tenfold increase in the stock price. But he apparently he was using the corporation as a personal bank account. He made himself the poster child of the greedy CEO.

Of his many indiscretions, two became legendary: He had the company pay half the $2.1 million bill for an extravagant party for his wife in Sardinia. Second, the company paid for a decadent $27 million apartment in New York that featured a $6,000 gold-and-burgundy shower curtain.

No matter that Kozlowski had been cited by reliable sources as a leader. No matter that I interviewed him to find he had in place leading-edge governance structures. No matter that an independent director confirmed the story. No matter that public documents filed with the Securities and Exchange Commission (i.e., the proxy) provided a record of checks and balances few other companies had. My reporting missed entirely the dark backstory—and I wasn’t the only one duped.

So the lesson is that research is never perfect, especially if you’re an outsider, and especially if you have funds and time for only so much digging. That’s why you can’t be too careful when gathering content for your book. Checking directly with primary sources, cross-checking with independent sources, fact checking afterwards to correct errors—these all help minimize the chance of leaving a fictional time bomb in your nonfiction book.

The difficulty of getting a story that holds up when all the dirty laundry is exposed is legendary in journalism. Errors are always going to happen. But ultimately the lesson is simple, if not foolproof: If your mother tells you to go back and triple check your sources, do what she says.

[Revised January 2020. Originally published March 5, 2015]