FEATURE19 August 2009
FEATURE19 August 2009
When telephone pollsters start sounding the death knell for phone polling, it’s time to sit up and take notice. But Jeffrey Henning of Vovici argues that phone research is here to stay – in one form or another.
Whenever the telephone rings and it’s a survey researcher, my wife hands me the phone. She knows that I will be happy to answer their questions. For me it’s about karma. Since I have personally conducted hundreds of telephone surveys myself, and commissioned tens of thousands more, I feel I owe each interviewer who calls me my time. I know I will never end up doing as many telephone surveys as a respondent as I have inflicted on others as a survey author and interviewer.
But the phone has been ringing a lot less. Here are some of the reasons.
• The increase in ‘do not call’ refusals. Many of the 72% of American households on the Do Not Call list refuse to take a survey when the phone rings. In the United States the national Do Not Call registry allows households to remove themselves from unsolicited phone calls from organisations they haven’t done business with. While the list itself does not opt households out of receiving calls from survey researchers, it has created in consumers the expectation that they do not have to take unsolicited calls from market researchers.
• The expense of dialling cellphones. The rise in ‘cord-cutters’ (households with no landline telephone but with only cell phones) has increased the cost of telephone surveys. By law in the US you can only use automatic dialling for landlines, not cellphones. Manual dialling is much more labour-intensive. That annoying pause you get when you answer such a call is because the dialler is now transferring the call to an active agent, after having tried unsuccessfully to get an answer on many prior calls. Without such diallers, agents are spending their time making calls that aren’t answered. And when you do manually call people on their cellphones, given the nuisance factor of such calls, they are less likely to answer your survey than if you had reached them on a landline.
“From the 1950s to the 1990s randomly composing a phone number and calling it provided a true probability sample of the US population. But thanks to those cord-cutters, that Golden Age has ended”
• Random digit dialling is no longer representative. Phone surveys with automated random digit dialling are no longer representative of the US population, as landline telephones now skew to older respondents. From the 1950s to the 1990s randomly composing a phone number and calling it provided a true probability sample of the US population. But, thanks again to those cord cutters, that Golden Age has ended.
• Online surveys eliminate the expense of data entry. Web surveys to house lists are much less expensive than telephone surveys to house lists. The respondent to a web survey is, in effect, donating the data entry cost, as they select the appropriate choices and type in their answers. With a phone survey, you are paying a call centre representative to transcribe each respondent’s replies.
• The web survey is a visual medium. Surveys that require respondents to react to ad concepts, mock magazine covers, TV commercials, videos and other visual stimuli can’t be conducted with telephone surveys alone. As a result, many publishers, movie houses and TV networks have shifted to web-based research.
• The web survey offers the allure of confidentiality. Thanks to the rise of blogs and forums, internet users are already comfortable confiding sensitive personal information to their web browser. Surveys about sexual habits, drug use, health embarrassments, serious medical issues and whistleblowing at work are all less intrusive when done on the web than when done by a prying telephone interviewer.
• Online surveys are preferred by many respondents. When we invite people to take a survey and offer them a choice of survey modes, they almost always choose the web link over a paper survey or telephone survey. The web survey can be done at the respondent’s convenience, rather than at the moment the phone interviewer happened to call.
So usage of phone surveys is in decline. But that doesn’t mean phone surveys do not have a valid role to play in market research. The television didn’t make radio obsolete, even if it did push dramas, soap operas and serials on to the TV set. In the same way, online surveys don’t make phone surveys obsolete so much as change the ways that they can be used successfully. As long as individuals are conducting market research, some studies will best be done by picking up the phone. For example:
• Qualitative research. Sometimes a structured survey is the wrong place to begin a research project. Exploratory telephone interviews, following a broad discussion guide, are often important in probing the changing requirements of customers and prospects. Once these in-depth interviews have provided a better understanding of the issues at hand, then a more formal questionnaire can be developed to identify the relative prevalence of some of the attitudes and issues identified. In this way, a qualitative phone survey is an important first step, regardless of whether subsequent structured surveys are done by the phone, internet or mail.
• Major account research. “The very best way to survey someone is face to face, but that is not economically feasible,” according to Andrew Hayes, vice president of business development for Bernett Research, which does telephone and web surveys as well as focus groups. “Interviewing someone on the phone is the next best way. For that customer satisfaction research where the individual customer has a high lifetime value, especially in financial services and high tech, the personal touch of a phone survey goes a long way.”
• The human touch. A Rice/NYU study, published in the Harvard Business Review in 2002 as ‘How Surveys Influence Customers’, showed that customers who took part in a customer satisfaction survey by telephone were more loyal than those who did not take the survey. Such customers were more than three times as likely to have done new business with the firm, less than half as likely to have defected and more profitable than the control group. How much of the effect was from the survey, the phone contact or both is hard to say, but clearly a telephone survey is ‘high touch’ compared to an internet survey.
“Concern about panel quality keeps many researchers using phone surveys rather than doing general studies on the internet”
• Demographics that are hard to reach via online panels. Certain demographics such as doctors, IT professionals and C-level executives (CEOs, CFOs, CTOs) are expensive to rent online panels for. For these hard-to-reach demographic categories, telephone surveys are cost-competitive. The online panel industry itself has a number of initiatives under way to improve the quality of their panellists. Concern about panel quality keeps many researchers using phone surveys rather than doing general studies on the internet.
• A backstop for the online survey. Again, for certain demographics phone surveys are important. “We have been called by one of the online panel companies where we will have to do the remaining 50 surveys on the phone because they just can’t do it on the web to fill the quota,” reported Hayes. “Even with the millions of registered panellists, there are projects for which they just don’t have the depth of the panel. They need to have a skilled telephone researcher working through the gatekeeper getting it done.”
My hope and expectation is that the phone survey will be around for a long time to come. My very first survey research project was a telephone survey for which I did all the interviews myself, and in the process I learnt a tremendous amount about how surveys actually work in practice. Every researcher can benefit from doing phone interviews themselves, simply for the insight into how questions are received and responded to, and those who do so will write better self-administered questionnaires as a result.
So while the phone might be ringing less often with that survey request, I have no doubt that my wife will be handing me the phone to take a survey decades from now.
Jeffrey Henning is Vovici’s chief strategy officer, and blogs about research and enterprise feedback management at blog.vovici.com.