David Day David is President and Global CEO of Lightspeed Research. He joined the company in 2006 as CEO, Europe to oversee the expansion of the European online panels and market research business. Read the full biography here.

Plenty more fish in the sea?

changes in online research and what the future holds for this fast growing industry

7th March, 2011

The sunlit uplands of online research

Latest ESOMAR estimates say that 20% of global quantitative research is now carried out online. In most developed markets it’s cheaper to use online data collection verses traditional methods and in emerging markets online is increasingly being used to research hard-to-reach consumers: the wealthy, the young and business decision-makers. As economies grow and wages rise, researchers are increasingly turning to the internet to understand the wants and needs of these emergent consumers.

As with many other fields of business, the internet is both a creator of new capabilities and destroyer of old. The high dependency in online data collection is accelerating the demise of traditional data capture as the costs of providing the latter escalate as demand dries up.

Some might think that online research specialists live in a bright and cloud free world. They’d be wrong. The internet is not a stable medium: it’s continually changing and changing fast. Where there was once email there are social networks and SMS. Ask your kids how often they email their friends rather than poke them or send them a text - a worry for an industry dependent on email as its primary channel of communication. Then there’s the mobile internet. Just how do we get those huge grids to be visible on a screen two inches across? Remember, in some emerging markets mobile phone penetration well outstrips fixed line internet access and it’s likely to remain that way for a long time. Not only that, but respondents are fussy. They actually expect questions to make sense, be in their mother tongue (Ideally translated by a professional as opposed to a free website) and to not take too long! Researchers, try doing your own surveys first and ask yourselves if you’d want to spend 20 minutes of your life doing that rather than some other fun thing on the web.

The industry is also uneasy over the issue of respondent quality. There are bad respondents out there though it’s actually less than you might imagine (or more if you are very naïve!). Once overtly fraudulent contacts are stripped out in the recruitment process, inattentive responders, speeders and straight-liners should not be more than a few percent. Most reputable panel companies are now putting a lot of energy into reducing the risk of poor quality results. Buyers of research should ensure that their panel provider is one of these companies before they commission a survey – see the end of this article for my suggestions.

It’s a panel but not as we know it

Despite the rapid consumer adoption of the internet, it is already becoming harder to recruit and keep quality panels. There are plenty of activities to engage consumers online and keeping them on a research panel is getting harder. We might have to think ‘surveytainment’, not an elegant word but worth considering when researchers are competing for the most valuable of anyone’s assets – time. To help combat the rising challenges of recruiting and retaining panelists there are two new techniques that the informed research buyer should be aware of.

The first of these is real time sampling or RTS. This technique involves the interception of a potential respondent while they are surfing a predetermined network of sites. At one level this looks attractive from a research perspective as it feels a little like buttonholing a passer by in a shopping mall. But this is where a little knowledge can be helpful. The probability of intercepting such a respondent is a function of how long they spend online so at the very least they are likely to be heavier surfers. Not always important but good to know. Also, experience shows that when using RTS very few respondents are actually new to survey taking and they are often duplicated on other online access panels. RTS has its place but it’s not a panacea. Its primary value is to remove email from the survey invitation process and drive efficiency as a result.

The second technique is survey routing. Routers increase the probability that a respondent will qualify for a survey by having several surveys available to the respondent at any given time. Which survey the respondent is routed to is determined by some short screening questions asked when the respondent enters the router. Routers can significantly increase the chance that a respondent will qualify for a survey. So, when used properly, routers can greatly improve respondent satisfaction as one of the major causes of panel attrition is the failure to qualify for a survey. However, some companies loop respondents back into routers repeatedly and many surveys can be taken in one session. Efficient? Yes, but not always great from a respondent (Think groundhog day) point of view. There are also legitimate research questions about the statistical rigor of such a non-random approach.

Panels are dead! Long live panels!

I’m often asked, and sometimes told, about the imminent demise of online access panels. I think we have been here before. When EPOS data first became available there was a view that consumer panels and retail audits would become a thing of the past. They didn’t for the simple reason that retailers can know everything that is sold through their stores and brands can know a lot about volume, prices and promotions. Through loyalty card data both retailers and brands can get insight into shopper/buyer characteristics and repertoires but only for that store. Panels are still used to get a holistic picture of the purchasing behaviour of consumers across all distribution channels, to understand their attitudes, their exposure to advertising and its effect on purchasing and to gauge their opinion on new product offerings. So, panels will remain at the core of online research, as both the research and online data privacy legislation will demand it, but they will offer richer insights through the addition of online behaviour and advertising tracking as offered by Kantar’s Lightspeed Research today.

And plenty more....?

The market research industry now has a high dependency on online data collection. When done well, online can be more efficient, provide richer data and deliver improved margins. But these benefits are only there if we have respondents. The industry needs to wake up to the fact that respondents are not just a commodity with ‘plenty more where they came from’, they are a valuable resource. Much is made of social media sites as a new source of respondents. However, it would be dangerous to assume that this means that the industry can carry on as usual. My list of what we need to do? Surveys need to be better designed and more engaging. They need to be concise and written in language the respondent can understand. They need to be better targeted to avoid constant screening out – the major cause of respondent dissatisfaction – and they need to be appropriately incented. In short, we need to love our respondents! If we don’t? Well, have you seen the price of cod lately?

David Day

Comments on this article

Want to share your thoughts...?

Want to share your thoughts?

NOTE: Please note that this board is moderated, and comments are published at the discretion of the site owner.