After calling shenanigans on ambient social location apps last week, the reaction I have seen has been mixed. For the most part people agree with me: these apps will be consumer duds. The technology and philosophy behind many of these apps is sound as the concept of implicit social graphs tied to explicit graphs through background location is indeed an interesting idea. Yet, there is one evaluation method where they tend to fail. What problem do they solve?
We will call this the “eye test.” In college basketball, the NCAA Tournament selects teams based on a variety of criteria including pure data but also considers if the team “looks” like it is capable of playing for the championship. On paper, the ambient location apps look great. But to be more than an oddity, they need to solve a problem that a real world user has.
Moneyball and the Eye Test
Framing our analysis is another sports related metaphor. In baseball, the concept of “Moneyball” is to take advantage of market inefficiencies through the use of advanced data. The concept disrupted traditional scouting that often focused on the eye test (does the player “look” like a major leaguer?) to focus specifically on advanced metrics.
In this scenario, the specific item we are looking for is the market inefficiency being addressed. There certainly has to be one or we would not have seen upwards of 12 different startups like Highlight, Glancee and Banjo attempting to tackle this problem.
At SXSW a couple of years ago, the hot startups that dominated the conversation were companies like Foursquare, SCVNGR and Gowalla. They were the first generation of “social location” apps and focused on the check-in model. To a certain extent these are the apps that are informing (sometimes literally through Foursquare’s API) the ambient location apps. What Foursquare and the others have never been good at has been social context. Check-ins plus deals, tips, photos and to-do lists are interesting but only nominally useful. Where are your friends? Are you more likely to check-in to a restaurant or pub just because you happen to be there or if you know that your friends are there?
This is where the line is drawn between reactive and predictive social location. With Foursquare you often get to a location and then check-in. That is reactive. With the ambient location apps you could be planning on going to one place and see that your friends are at another and go there instead. In this scenario, the app “pulled” you to a different place than your original destination.
For Highlight, Glancee, Banjo and Sonar there needs to be more than “disrupting Foursquare” to pass the eye test. Another argument that has been put forth is the notion of creating a new style of social CRM. This is the idea of replacing business cards by creating connections in your phone when you meet somebody. The ambient location will know that you were near somebody, save it as a record and allow you to save that person to your contacts or social profiles like LinkedIn. While this is an interesting concept (nobody is saying these apps lack clever ideas) it comes down to the original argument: who and how many people are going to be using these apps? Creating a new social CRM based on ambient location will not work if only a few people you know are using the app or if one person is using Highlight and the other using Glancee. There is a pleasing degree of serendipity with who can meet with the apps, but serendipity is not going to replace business cards. In this case, the reality of real world interaction with these apps does not pass the eye test either.
Subjective v. Objective Analysis
Outside of the concept of disrupting market efficiencies, what Moneyball really boiled down to was the concept of objective vs. subjective analysis. Subjective analysis had ruled in baseball for a hundred years before advanced statistical measures became vogue.
In Silicon Valley, the mindset is often the other way around. Startups and venture capitalists drool of the idea of data. How can it be created? How can it be monetized? What kind of technology is creating it and is it a leap in innovation? This is antithetical to the subjective test of real world adoption outside the microclimate in which these conversations take place. The eye test in the tech world often has more to do with how pretty a user interface is than whether it can be a useful service millions of people.
The objective analysis of ambient location apps points to services that could potentially have great data sets. Better location and social models based on location awareness mixed with the data created by such interactions theoretically could have a profound affect on user behavior. Brands and retailers could find this information useful as well. Look at what Geoloqi is doing for proof that background location data is a very, very valuable tool.
From the subjective standpoint, where most real world users are going to live, ambient location apps have trouble passing the eye test. The problems that they “solve” are minor and the ambitious goals are less likely to be achieved when growth stalls after the first spike of early adopters.