'Passive data' has great potential on its own: more when combined with survey research, according to our panel of MRXperts.
6th December 2012
New generations of mobile devices - and their spread to an ever-increasing portion of the population - offer a raft of new opportunities to the survey researcher, from multimedia presentation and survey apps to on-the-spot feedback and diary-style, camera-assisted responses. Some of the leading researchers in the field, however, are just as excited about devices’ ability to pass on data ‘passively’.
Once you have a mobile user’s agreement, a phone or tablet can legitimately and constantly feed back information about what’s being done with it, and where its user is. Although this throws up questions over privacy - the definition of getting agreement and the anonymisation and storage / sharing of data are thorny issues - the potential advantages for researchers are huge. We perhaps hear less about them than we do about the potential for ad targeting, but be in no doubt, researchers are going to be making more and more use of passive data.
This isn’t just a hunch on my part. Asked this month what were the most exciting advantages of mobile phones for researchers, our panel of mobile MRXperts, not unexpectedly, picked ‘Immediate, in the moment feedback at the point of purchase/experience’ as the no.1 - we’ll come back to that in a later article. But second equal, and far ahead of any other answer, come two other aspects: the ability to get passive usage data from the phone itself, and the ability to locate users using GPS features. And it’s worth pointing out that most of our panel are not specialists in automated data collection, but survey researchers.
These two answers between them accounted for a fraction under half of all the panel’s selections (see below) - and yet the potential is evidently under-exploited at present:
‘There appears to be a pretty big appetite for passive measured data by large MR agencies and other digital firms but very few today are doing it on the scale they would like.’
True, there have been a lot of announcements about passive metering and feedback technology in DRNO’s news lately - a few about the privacy issues involved, but rather more about the potential for feeding useful information to researchers. But this is the leading edge, and very often it’s not full service research companies that are involved. Passive data is being used more by networks and a handful of tech-oriented data providers.
GPS / location data is not as frequent a mention in the news, but seems to me one of the most exciting developments in mobile tech, as yet largely unexplored - pioneering companies are no more than scratching the surface of what the technology makes possible.
Nielsen is something of an early adopter of the first of these - the company ‘has been running passive gathering of smartphone usage for some years’, says Telecom Practice VP Edward Kershaw, through its programme of opted-in panels of Android and iOS owners. ‘It delivers a wealth of data, with the ability for detailed analytics to understand smartphone usage’. [NB: Edward will be discussing passive data and the potential for combining it with survey findings, in a forthcoming article in this section]. Others feel that while the technology is there, researchers remain reticent, partly but only partly because of issues surrounding data protection and consumer privacy.
Several panellists said they see the biggest potential in combining data direct from the device with survey responses:
‘ I think the smartest firms are those marrying up this type of data with other data - web-tracking, self- reported, and so on.’
I suspect there are few researchers who wouldn’t find this appealing for one reason or another. Mobile hybrid methods might take a variety of forms:
combining passive data with online and off-line behaviour and habits to offer ‘a holistic view of the consumer’, as one panellist put it
using automatically-generated usage or location data to serve as a fault-free and unintrusive screening process - pick up the right people for your survey, when they’re in the right place, and without having to take them through some or all of the otherwise-necessary introductory questions or waste the time of both parties discovering they don’t qualify
running surveys about online behaviour and using passive data to help gross up the results
or indeed using passive data to check the validity and quality of survey responses. Does Respondent A really use that app as much as he claims? Is respondent B anywhere near the shop in question, or is she sitting at home?
I’m sure this is just a start - one of the exciting things about the sector is that nobody’s really able to see where it’ll take us - there’s scope for a lot of lateral thinking and a few fortunes to be made. And yet researchers don’t need to take risks to explore these possibilities - apologies for the cliché but it’s more dangerous not to do so. If we get into arguments over ‘surveys vs passive data’, the latter will win out in many cases, with clients looking for cheaper and faster data all the time - but ‘think what you could do if you had both’ ought to be an easy sell. Hybrid methodologies have always been around, whether it’s combining quant and qual, survey work and desk research, or server access logs and online panels - and they’ve generally been strong contenders.
One last thought on the benefits of tapping one respondent for both survey and passive data - harking back to my days doing b2b interviews. If you ask a respondent when you first speak to them if it’s OK to identify them in reporting back, you get a lot of refusals and you risk getting rather guarded responses. Once you’ve spent some time talking to them they’re much more likely to accept the research for what it is, you for a human being, and the feedback request as a genuine one, and if you ask at the end very few say No. Similarly, I would imagine that if the only thing you ever ask a mobile user is ‘can we get data from your phone?’, you’ll only get OKs from a certain type of person; if you combine it with a more personal approach such as a survey, you’ll get a much broader spectrum and much better data.
As well as your views on the above, I’d love to hear some examples of hybrid approaches in action, so please get in touch using the comments form below.
Hi Antoine, thanks for your comment. You're right, there is of course a huge difference between self-completion and having a chat with an interviewer, and to get a rapport as discussed at the end obviously I'm talking about interviewer-based surveys, whereas hybrid approaches will come in many forms / combinations.
We will be going into more detail about the key issues here, in future articles, and talking about both interviewer-based and automated surveys during 2013.
Nick Thomas, MrWeb
The 'GPS' tracking is a tricky one and I think that there should be a clear distinction between self responded surveys and mobile surveys being performed by a pollster. It also makes a lot of difference in the way to propose Mobile Survey.
Any chance the study is covering this issue in more detail?
Antoine, Datafield
Want to share your thoughts?
NOTE: Please note that this board is moderated, and comments are published at the discretion of the site owner.