This article is also available in Serbo-Croat, courtesy of Jovana Milutinovich at Web Hosting Geeks - certainly our first in that language - thanks Jovana!
Why isn't everyone doing Mobile Research? Teresa Lynch asks the question and answers with the help of our panel of mobile mrxperts
20th Jun, 2012
Imagine if you will - I?'m setting up a market research company from scratch. My plan is to provide quality data, fast turn round, good hit rate and accessible results, all at a competitive price. I don't want particularly high set-up costs and I want to be able to start making money right away. I choose mobile. 
Its a no brainer: the people I want to collect data from are not sitting at their desks waiting for my email, they are not sitting at home waiting for my call or my knock on the door, they are already communicating with one or more others on their portable device while they are mobile. Ditto my clients who will be the recipients of the results. Small point here: as my old poetry professor was wont to say ?you are mobile, the device is portable. So by mobile I mean any connected portable device.
So why, when 100 million Americans are walking about with a smart phone in their pocket1, a conservative 25% of household in the western hemisphere do not have a landline2, and projections suggest that by 2013 mobile access to the Internet will have overtaken desktop3, is mobile research being employed by less than a quarter of research agencies4?
We decided to look at one area of mobile research that researchers say might be discouraging clients and agencies from moving wholesale to interviewing on portable devices: abandoned questionnaires/poor completion rates.
Conventional wisdom says at the moment that mobile surveys should be short and simple, so perhaps the problem is that mobile does indeed have abandonment issues. We asked our panel of mobile mrxperts: ‘In mobile data collection, how big a problem is questionnaire abandonment?’ They were fairly sanguine: the majority saying either that they got better levels of completion on mobile or that their completion rates were about the same. Only three respondents said that their completion rates were worse. These three, when answering the subsequent question about which feature caused respondents most problems, gave slightly different answers: ‘not grids - repetition of grids’, ‘difficult layouts’ and ‘mandatory questions’. Five respondents who didn’t feel that abandonment was worse on mobile also mentioned grids, but presumably they would have also implicated grids in failure to complete in any other medium.
However, the worst offence in the opinion of the majority (11) of respondents was poor layout and design:
"I think the layout has a huge impact. The text and input boxes need to be big enough on a non-cluttered screen. The layout would be a priority in my opinion"
"Difficult layout puts respondents off"
...and, of course:
"?layout not adapted to mobile".
After both layout and grids came the inevitable survey length: "From experience we never ask a mobile survey longer than 15 questions and ideally the questionnaire needs to be under 5 minutes?." said one. Another summed up: A mobile device surveys need to be shorter, engaging and fit for purpose.
Thankfully our mrxperts also had some ideas about how to, as the next question asked, "get clients to accept these limitations/your solutions." Sharing seems to be a big part of the answer: "Easy, by sharing the right knowledge and experience"
and
"Our clients and us have to think mobile and understand that mobile surveys are different to standard web surveys, just like those are different from paper questionnaires. We usually work with early adopters who get this difference so it's easy to get them to accept the differences and welcome the new opportunities!"
Clients are also learning from previous mistakes:
"??It's not easy, although clients are learning and the industry is getting better at understanding the benefits of design by device and delivery channel. Ultimately the consumer will decide in how they respond and stay engaged but I think there is an opportunity to learn from last mistakes".?
and
"??It's a matter of informing about limitations and offering alternatives".
The best thing about our quick survey of practitioners was that of the 13 mrxperts who said they had either "?Constant & Direct?" or "Considerable" involvement day to day with mobile surveys, 8 said completion rates were either '?the same' ? (4) or mobile was 'higher' (4); only 2 said it was worse. This corresponds with a statistic I read recently from Globalpark (now QuestBack) that 35% of respondents replied to a mobile survey in the first 2 hours5. A response rate that lots of conventional researchers can only dream of.
Did we collect any more good advice on getting clients to accept limitations and solutions? There is a certain amount of deja vu in some of the comments:
"?Not always easy, the issue with clients is that they want to ask every question rather than the key questions. When posed with the limitations, it does however focus the client?s minds in deciding what they really need to find out".
"Near impossible... whenever there is a new mode (here mobile) people tend to think it's a bottomless well of opinion."
"... there is too much legacy acceptance of poor survey design to radically change behaviours in many instances".
Any one of the above could have been written when we moved from CATI to online or, heaven help us, when we moved from face to face to telephone. My favourite and very pragmatic response came from Navin Williams at MobileMeasure:
"?The issues are more to do with lack of comfort with deviating from existing norms than limitations that Mobile brings with it. So for business as usual research it's much tougher to move client budgets to mobile at the moment. Its early days, those who dare do - the rest are waiting to follow once they see enough cases out there".
So in summary our leading practitioners know the benefits, they are taking it to the stakeholders and now we will just have to wait for that major shift in the market which may just be months away rather than years.
There's plenty more to talk about with our panel - we want to know about geo-location, ethnography and point of sale to name just a few - and with our pundit's help we will pick the discussion up again soon.
I have one last quote for you. We used uSamp's freemium authoring tool: instant.ly and had no complaints about our software from any of our respondents, in fact one pundit (not from uSamp) said: "It is much easier with surveys with a proper mobile layout. This one is excellent".
Teresa Lynch
1comScore press release March 2012
2Eurobarometer - E-Communications Household Survey July 2011
3MobileMarketingWatch Jan 2010
4Grit 2012 Dashboard
5Globalpark conference workshop 2011
This article has now been translated into Serbo-Croat, courtesy of Jovana Milutinovich at Web Hosting Geeks - certainly our first in that language - thanks Jovana!
Comments on this article
Want to share your thoughts...?
NOTE: Please note that this board is moderated, and comments are published at the discretion of the site owner.
| 
