In theory, cell phones have made it easier than ever to call for help in dangerous situations. But they're still far from perfect. You can't always reach for your phone, for one thing--and even if you could, in certain scenarios doing so might only escalate the danger.

That's why Dave Benoit and Phill Giancarlo founded Wearsafe, a startup that created a service to help people in distress. The Wearsafe Tag, when pressed, begins sending live audio from the wearer's phone to the devices of a chosen group of contacts. Those contacts are instantly placed in a chat room where they can see the wearer's GPS location, hear the live audio, and decide what action should be taken.

Benoit and Giancarlo first had the idea after hearing of the horrific 2007 Petit family home invasion in Cheshire, Connecticut, in which two men broke into the house, cut the phone lines, took the family's cell phones, and raped the mother and her daughters before setting the house ablaze. The mother and girls died, and the father barely escaped with his life. Making the news even more upsetting was the fact that the ordeal lasted seven hours, with the family unable to call for help.

Working in nearby Hartford at the time, Benoit and Giancarlo thought they could come up with a device that could be useful in emergency situations but had a wider usage than the straight-to-911 devices already on the market. The former college buddies, both working in the investment banking world, pooled together $1 million from friends, family, and their own savings, and officially founded Wearsafe in 2012.

Wearsafe launched its service and started shipping the device earlier this month. It sounds like a simple idea, but figuring out which features to include and which to leave out--all the while keeping the device as simple and easy to use as possible--proved to be design problem that challenged many of the founders' assumptions.

Developing the product

When a person presses the button on the Wearsafe tag, his or her phone starts recording audio and sends it to the pre-selected group. That group can change based on the person's location--so if the wearer is on campus, a different set of friends will be alerted than if they were near their home.

Benoit and Giancarlo say they chose to include the audio feature because a simple alert and GPS combination often isn't enough, especially if the person isn't in an unusual location. "Maybe I call [my daughter], and it goes to voicemail--now I'm just in a panic, and I have no information about what kind of aid she needs," Benoit says. "Having the audio lets you hear the situation--who's there, what's happening--and make a decision based on that."

In creating the design, the founders worked closely with students at three universities--the University of Connecticut, Hartford's Trinity College, and the University of Hartford--to find out what features were important to them and what they would actually wear. Benoit and Giancarlo originally discussed disguising the device as a piece of jewelry, such as a ring or watch, but learned that if the device didn't match the day's outfit, it'd likely be left at home. Eventually, they decided on a tag that can be clipped to a piece of clothing or an undergarment or looped onto a keychain.

Each user can preprogram the device to set the number of consecutive clicks that triggers an alert. The ridged button can easily be found and pressed without looking.

During development, the founders also gathered feedback from retired Secret Service and CIA agents and Navy SEALS, among others. "They pointed out," Giancarlo says, "that people who have reassurance that help is on the way have a much, much higher probability of a successful outcome in a stressful situation." For that reason, the tag vibrates silently when the alert is sent and again when people start responding in the chat.

Maybe the most important--and surprising--finding was that a device that alerts authorities immediately was far less likely to be widely adopted. In many wearers' minds, the fear of an accidental or unnecessary alert outweighs the potential benefits. That's why the founders chose to have the device alert chosen contacts instead. And informing friends can have other benefits: Students on campus, which is a demographic Wearsafe is especially targeting, can often get aid from a friend in a nearby dorm more quickly than the police.

But what if none of the chosen contacts happen to be near their phones? The system is programmable via an app so that it can automatically go to a second or third contact, such as campus security or the police, if nobody responds in the chat within a certain amount of time.

While the device could be useful in helping to break up a violent crime in the act--since in most cases it would be easier to reach down and press a tag than to dial a phone--its main purpose is as a lifeline before a situation escalates. The founders gave the example of someone feeling uncomfortable on a date or cornered at a party. Pressing the button would let a friend nearby swoop in and conveniently break things up.

That's not to say the product is a foolproof solution--far from it. In some situations, like the recent attacks on joggers in New York, Massachusetts, and Michigan, it might be more beneficial to have a product that sounds a loud alarm and lets an attacker know help is coming. Another downside, especially for joggers, is that you need to have your phone with you for Wearsafe to work. And in cases in which an attack is already under way, wearers would certainly want the system to go straight to the police.

Benoit and Giancarlo considered making the device have options--two clicks for friends, three for police, for example--but decided against it, for now. "We found that if people had uncertainty about what was going to happen when they pressed that button, they were less likely to use it at all," Giancarlo says. "In a moment of panic, you don't want to have one more thing to think about." But those options could come with future updates. "We can program the heck out of this," he says. "We've left a lot of optionality in the device and in its software."

Listening in

Another feature the device offers: The option to record the 60 seconds of audio from before the button was pressed. The thinking is that, for a third party, knowing what led up to the button being pressed could be critical in making a decision.

But it also brings up an uneasy thought: that the device is recording at all times. Giancarlo says this isn't the case. "It's not recording--it's buffering," he says. "And that feature is optional to the user. Not everyone wants it. We recommend that people turn that on when they're in situations that are unusual for them, or when they may be at greater risk."

There's also the issue of whether someone could hack the device's app and remotely listen in. "Now they don't have to hack your phone--they can hack the app's system," says Joseph Steinberg, founder of SecureMySocial and an Inc. cybersecurity columnist. "It's my belief that if someone wants to spy on you through your phone, they could probably do that right now without you having this app. It's possible that adding an app like this increases that risk."

Steinberg also has concerns about the fact that recordings like this could be subpoenaed. Once a recording exists, if there is suspicion that it could contain evidence, it could be up to a court, and not the individual, to decide what to do with it.

In the end, of course, a person has the choice of whether to use the device at all--and while it may not solve every problem, it could function as a lifeline in dangerous situations. Wearsafe's service runs $5 per month through the company's website, and the tag comes free.

While the tag is Wearsafe's flagship device, the founders say they expect to activate the service via smartwatches, fitness trackers, and other devices in the near future. With the recent news about connected temporary tattoos, the founders aren't ruling out the possibility that future iterations could be embedded in clothing or placed closer--even on the skin. Or it could move away from the body altogether, so that a person's voice could be the activator.

"We're agnostic on the device," Giancarlo says. "We're a service company."