In 2009, Greg Nemeth, along with co-founder Arun Gupta, dropped out of college to work full time on a start-up company, WakeMate. The company's eponymous product, a mobile-connected wristband that would track users' sleep patterns and wake them up at an optimal time, had already generated buzz on tech blogs, and Nemeth presold thousands of units. He was confident they would be ready for shipping soon. But a series of technical issues delayed the WakeMate's release for nearly a year.

Right from the beginning, everything that could go wrong went wrong. We were testing samples from a few overseas manufacturers, to see whose quality was best, and we got back things that were just absolutely wrong. One company soldered some of the chips on the circuit boards backward and upside down.

Missing our release date was pretty disheartening. We knew that we were going to have a lot of upset customers on our hands. We thought we'd be able to work through all of our issues pretty quickly, so we could just say that there were some delays and we were working on them. We also promised some new features. That didn't keep people happy. They wanted more transparency. When we finally said, "Look, this is what happened. We bought some parts, and they turned out to be defective. We're going to need to come up with a workaround," people were much more understanding. By the end of the process, we were giving them as much information as the employees had.

One day, you feel like you're going to take over the world; the next day, you feel like you're going to fail miserably. That is definitely compounded when you have a bunch of customers constantly saying, "Where is this thing? Does it even exist?"

I think the debacle with the chargers was the worst part. We had finally gotten over the hump of actually getting the product out into people's hands. Then we received an e-mail from one of our customers: He had seen his WakeMate spark. Our first reaction was to alert all customers as quickly as possible that their chargers might be hazardous. We sent out an e-mail to everyone, with a link that they had to click to confirm that they had read it and were disposing of their chargers properly. For people who didn't respond to that e-mail, we text-messaged them, and then called them, to let them know what was going on. Within three days, we had reached everybody. Customers obviously weren't too happy about the possibility of their units catching on fire, but they appreciated the effort that we made to let them know.

Afterward, we did an autopsy on that customer's device and on a few chargers and saw that there was a faulty soldering joint. The chargers were putting out a lot more voltage than they were supposed to be. The actual units we got were a lot different from the samples we had tested. It was a lesson: Once you buy, you really have to make sure the quality of everything you're getting is what the manufacturer says it is. The best way to do that is to manufacture everything locally, which is what we're doing now.

Some people said, "This thing isn't quite ready for prime time," which was true. But we've been constantly improving it, and people have been extremely helpful and responsive. Most of the reviews and the Twitter buzz now are really positive. We're happy to be at the point where things are stable and the actual functionality of our product can shine through.