Inside Edition found differences of over 10 percent between the triplets they tested. That is not a small gap. If you were off by 10 percent on a DNA test, you could technically be a mouse. Maybe it's unreasonable to expect perfect accuracy from saliva you mailed to a lab. But a lot of people do anyway, and Morgan winds up dealing with their complaints.

"At least once a week, we'll get a call from somebody who took two or three other tests and then ours, and complains about how different they are. Usually it's 5-20 percent off, but we got an email from a guy showing how in one test he was 7 percent Irish, Scottish, and Welsh, then on another he was 33 percent, and then on ours 45 percent, and he wanted to know what was wrong with everyone. We wrote to him that each test is different because of the number and types of genetic markers used, which can skew data, but he wrote back and said that we were con men."

Genetics experts from the University of Texas and the University of North Carolina have gone so far as to say that these companies are preying on people, because they don't truly have the information they need to pinpoint your origins on a map, and that it's not possible to trace unique ancestry that way. As they put it, "That's the beauty of this scam. The companies aren't scamming you. They're not giving you fraudulent information. They are giving you data, real data, and allowing you to scam yourself."

Even though Morgan works for one of these companies, he doesn't buy into the accuracy of the product. How could he? "We were doing our own internal tests when I started, and I took the same test five times in five weeks, and I got different results each time. One of the lab assistants wasn't upset about it. He told me, 'Look at the range there. That's about where your ancestors are from.' Somebody asked him, 'We promise accurate results. How is it accurate if he got different results each time?' And the lab assistant said, 'If you average them all, you have a good idea, right?'"

On one hand, these tests are definitely a con. But on the other hand, the customers are as guilty as the companies. People want to know where they come from so they can brag about being 1/64th Cherokee in internet arguments. No one actually wants to spend hours studying genealogy and pay hundreds of dollars for a dozen different, possibly more accurate tests. "If you get a high percentage, it's a safe bet that you have ancestors from there. I'm talking about a 50-60 percent on your test. Anything lower, and take it with a grain of salt."

"If the results only added up to 99.5 percent, we'd say, 'Let's stick that 0.5 percent under Scandinavian.' Other times, when we ask their family name (for women, their maiden name), and we see what country that last name came from, we'll add it there, because they'll be more proud of that heritage more often than not." As scientific as DNA tests seem, we're still talking about a customer-focused industry. It pays to suck up to the people who pay you.

"One woman sent the packet out with green shamrocks and a green leprechaun hat on it. She was really proud to be Irish. She even said she was excited to see if she was 100 percent Irish. But the test found no Irish blood. It was half Eastern Europe, then a mix of different places in Germany and Italy, and even Greece." Clearly, this woman's family had either lied to her, been lied to themselves, or she was one of those stealth adoptions that happen every so often. "The consensus was that she would send a huge fit if she was shown not to be Irish at all, so we made her 20 percent Irish and highlighted our disclaimer about results not being accurate." Or they could have put down "100 percent" on St. Patrick's Day and still technically be right.

Morgan wasn't at all ashamed or guilty about this. And we can't really blame him. If you've ever waited tables, you know how fucking crazy people can get when their expectations are challenged in any way. Now take the anger of someone not booking a reservation they really wanted, and amplify that by someone learning for the very first time that Great Uncle O'Leary's tales about the Emerald Isles were a pack of lies.

"We actually started it here, because during our first year, a woman found out that all of her parents' stories about her ancestors coming on the Mayflower were lies (She was nearly 100 percent German). [She] came to our physical offices, cracked her way inside, and confronted the staff in person. I wasn't with the company then, but it's still talked about here. They had to call the police because all the security staff in the building couldn't get her out."

It's not unheard of for genetic tests to be altered. New York crime lab workers have sued the police for forcing them to change or ignore results, and The New York Times found that anything related to DNA, from Ancestry results to crime scenes, can be fabricated easily. North Korea disavows that Kim Jong-Un is a quarter Japanese, despite a lot of evidence to the contrary. So Morgan and his co-workers aren't even close to alone in their little DNA-based white lies.

It sounds simple enough. But people fuck it up all the time, because people fuck everything up all the time. It's our prerogative. "One out of four people we need a new vial for, because something went wrong. We'll often get food in there, or they drink something, then spit and not realize it."

And yes, sometimes they "fuck up" literally. "Once a man sent us a vial full of semen, and I don't know HOW he could have gotten that part wrong. Our lab assistants were originally going to rule it out immediately because of how cloudy they thought the 'saliva' was, but one of them decided to see anyway. She opened it up to begin testing, but as soon as she saw the viscosity, she dropped it and said, 'That's semen!' One of us had to call him directly and tell him what he did wrong, and to his credit, he did give a mortified apology."

Another thing that complicates DNA test results: interracial lovemaking. "Sometimes the saliva looks good and we test it, but then the results show something really messed up. We had a few tests where the genetic markers where everywhere, on five continents. It's really rare for that to happen. The percentage points were 10 percent in India, 10 percent in China, 5 percent Native American, 10 percent Sub-Saharan Africa, 20 percent Scandinavia-Norway. You can imagine. I called [some of these customers], I explained what the situation was and went down the results, and what always happened was that they would say yes to half, and no to the other half of the ancestry they knew. When we asked if they knew anyone who fit the other ancestry results, they would say, 'Oh, my husband/wife/fiancee/boyfriend/girlfriend is!'"

It's not uncommon for tests to take way more than one vial to work. There have been cases of users repeatedly sending in vials, as many as eight times. "It's not unheard of. And I've had more than eight. We had someone who tried 14 times before it finally worked. They had bad teeth, and little tooth bits kept going in the saliva and messing up the results. They weren't big pieces, maybe like grains of sand. But it was enough. By the sixth time it happened, she was known as 'Tooth Girl,' and I would get a message saying, 'Tooth Girl struck again.' She never called once, either. Usually, by the third time, they'll ask what went wrong and how to do it better. Maybe they didn't clean good, or maybe something bled. It happens. But she never asked, and kept sending in vials with tiny bits of teeth."

Eventually, Tooth Girl's tests worked out and they were able to send her useful information. Morgan and his co-workers would never fuck with someone for having bad teeth. Not when there are so many better reasons ...

Morgan and his colleagues were caught between a rock and a really-want-to-mess-with-racists place. It would've been fun to throw a "10 percent West African" in there, but then they might have a pissed-off, dangerous person at their office, waving a gun. "Since we couldn't do anything to the results (and we wanted to), what we did was add '< 1 percent' to each African category of ethnicity. That way we weren't lying, and they would both be wondering how much under a percentage point was. We always try to round to the nearest number because we sometimes hear about percentage points, but for them, we leave it open to whether it's a one or a zero."

It's a compromise that's elegant in its passive-aggressive simplicity. And it got a result. "The near-N-bomber wrote to us asking what that meant, and we wrote back that it meant it was under 1 percent. And we were not saying zero. Unless they got another test, that was going to bother them. Maybe they weren't 100 percent Caucasian. I mean, they were, according to the results, but this way it leaves it open, and they'll always be wondering."

Bravo. Bravissimo!

But for every heartbreaking revelation of lies and infidelity, Morgan has stories like this: "I didn't talk to him, but someone who called us had lost his parents and his brother in an accident, was left with no family, and then decided to do a DNA test. It led him down a rabbit hole of finding out that he had been adopted, and that he had a completely different birth family welcoming him in open arms."