The local bar scene has always been a mystery. Step into a pub on any given night and you might get a bro-filled fratfest dominated by Polo-clad Murray Hillers. Another evening could attract a smaller, older crowd, shouting out answers at trivia night. The age, gender, and size of a bar crowd will forever remain determining factors in deciding where to go–and now, thanks to an iPhone and Android app released in June called SceneTap, you can figure it all out before you leave your apartment.
SceneTap provides users a real-time snapshot of their local bar scenes. By installing cameras at participating venues, SceneTap uses “facial detection” technology to give users information on a bar’s male-female ratio, average age, and the total number of patrons. But while consumers may jump at the opportunity to find out which bars have the best chances of yielding a good time, according to cofounder and CEO Cole Harper, it’s the bar owners who are likely to see most value from the system.
SceneTap prides itself on its technology, which the Chicago-based startup says is incredibly accurate. Not to be confused with “facial recognition,” the Minority Report-like tech used to identify people, SceneTap offers “facial detection,” which analyzes facial characteristics and matches them to an anonymous database. “[Our technology] doesn’t care who you are, but what you look like. It’s an algorithm behind the scenes that’s trying to figure out what you look like,” Harper says. “Do you look like a male or female? A 22-year-old or a 25-year-old?”
SceneTap looks at a variety of characteristics to determine gender and age: the nose, the eyes, the jaw structure, mouth and overall face shape, forehead and skeletal structure. “It almost takes your face and creates a grid, matching general facial features to males or females, before determining how old you are,” Harper explains. “In a certain sense, it’s trying to find your look-alike in an anonymous database.”
Based on tests, SceneTap is anywhere from 85% to 98% accuracate when totaling the amount of patrons entering or leaving a bar; it is 85% accurate in determining gender; and is 90% accurate when determining age on a range of +/-6 years, and 80% on a scale of +/-3 years. “Accuracy depends on lighting, the camera angle, the distance between the camera and the door–there’s lot of factors that have to be accounted for,” Harper says. “If you go to a club that has strobe lights, or TVs constantly flickering–the software needs very consistent light. It’s not to say it’s inaccurate, but there are factors that need to be calibrated [at each venue].”
There is real value here, he argues, for bar owners. “It’s a way of measuring the effectiveness of your marketing and advertising,” he says. “If a bar owner says to a promoter, ‘Your job is to get me 100 females in this bar by 9 p.m.’ Well, now the owner has a way to measure that promotion.”
SceneTap believes it has created a viable solution for venue owners. A doorman might be able to track all the patrons entering and leaving a bar, but not keep track of age and gender. A POS (point-of-sale) system might provide good data, but a lopsided picture. “If you own a nightclub, for example, and look at the POS system to backtrack over the night based on names on receipts,” Harper says, “well, if you do that in environment like a nightclub, it’s probably going to be 90% males, because they are the ones most likely buying the drinks.”
Other modern services can provide similar data, but are not as comprehensive a solution, according to SceneTap. GenderZoo and Foursquare, for instance, offer users a snapshot of the total number of patrons and the male-female ratio. “But we’ve done a focus study on three of the bars in our network, specifically evaluating our numbers against Foursquare’s, and what we found was that Foursquare is basically collecting half of 1% of the users in terms of actual check-ins versus the number of people who walk in,” Harper says. “So for every 200 people that walk in, you have just one person that’s using Foursquare. For us, to take the numbers from Foursquare and say this is an accurate depiction of what the scene is like at any given moment? I think that’s pretty tough.”