OPINION19 June 2013

Always on, always watching

Opinion

A host of passive behaviour-tracking tools and technologies were on display at the Insights Innovation Exchange conference this week. Jeffrey Henning reports live from Philadelphia.

Whatever Americans think of trading off privacy for security through the recently leaked federal surveillance program PRISM, they are willingly trading off privacy for incentives for commercial “surveillance”.

The startup Raw Data discussed its panel of respondents who willingly run an always-on listening app in the background on their smart phones, in exchange for unlimited data plans. For instance, want to find out if radio listeners change the station when a particular commercial comes on? Now you can. (As a result, the company was runner up in the Insights Innovation Exchange, in which six finalists competed.)

Such passive tracking apps are becoming surprisingly common. Research Now showcased their tracking of panelists through a smart phone app passively logging GPS location and app use, as well as tracking online behaviour on panelists’ computers. Observational data collected in this way revealed that 28% of panelists had purchased a vehicle, higher than the 17% reporting such purchases in a 90-day tracker, and 23% intended to purchase a vehicle, compared to 12% in the tracker.

Observations of non-digital behaviour are on the rise too. Shopperception discussed an aisle monitoring system that passively tracks all activity in a store aisle using a Microsoft Kinect sensor. It tracks items being removed from a shelf or put back as well as where shoppers stop and look, producing a heat map of activity and attention.

And Abacus Data discussed their prototype for tracking interactions with digital signage. Their system performs eye tracking across digital menus on multiple screens, records the video output of each monitor as it changes, and reports the results to Abacus servers in real time through the store’s WiFi.

Neither firm offers real-time adjustments to the environment based on the insights gathered, requiring sequential testing of different retail environments. However, both hinted that such abilities were in the future: adjusting the digital menus, in the case of Abacus, and sending messages to digital signs, for Shopperceptions.

With the rise of new methods of observational research, this would seem to naturally portend a decline in directed inquiries such as surveys.

Not so fast. The survey is proving remarkably adaptable, and essential.

In fact, Paul McDonald of Google said that micro-surveys “future proof” the survey, enabling it be easily taken on mobile devices, which will drive the majority of internet traffic in two years. The company announced that it will be soon be rolling out five to seven-question surveys (contradicting my prediction last year). McDonald portrayed a future where Google surveys could reach a billion potential respondents around the world. And as an additional flourish, he railed against traditional brand trackers, declaring them a waste of time and money.

Jon Puleston of GMI argued that brand trackers were wasting respondents’ time, and the rational response for a respondent was in fact to speed through the survey to the end. Jason Anderson of Insights Meta had declared the day before, “A survey is a game. It’s just a really boring one.” And Puleston argued that finally treating trackers as a game was the best way to reward respondents for spending the time to provide honest answers. He discussed techniques for gamifying awareness, usage and attitude collection, and the resultant increase in respondent satisfaction and the quality of answers.

The award winner for the Insights Innovation Challenge was in fact a survey company, RIWI (an acronym for Real-time Interactive Worldwide Intelligence). RIWI supplies a new, randomised, fresh, unincented river of respondents from almost every country in the world. The randomisation comes from leveraging the direct navigation bar to capture millions of people on every web-enabled device who mistype a URL and instead end up at a site offering a RIWI survey. Surveys are typically four to eight questions long. RIWI was the only research firm to predict the Egyptian election, using a mobile survey completed by 10,000 Egyptians.

Jeffrey Henning is president of Researchscape International