Robots and the Uses for IVR in Market Research Surveys
Posted by Vaughn Mordecai on Thu, May 20, 2010 @ 04:51 PM
Did you ever used to watch the TV show "Mystery Science Theater 3000"? The show was extremely funny and very creative. The basic premise was to take a bunch of old, corny, black and white science fiction and horror movies, play them in the background on a large screen while a couple of "robots" and a human make fun of them. Sounds like a strange premise, but it works. If you've never seen it, make the jump (after you read this blog) to their website and check it out. Very funny stuff.
The thing that every person knows about robots is that, like Mystery Science Theater 3000, all good robots should have 
the ability to speak. Market Research has speaking robots of their own. If you're involved in doing telephone market research, you've heard of "robo-dialing" or "auto-dialing" and "interactive voice response - IVR". Take a look at this picture...now you know who's making these automated research calls.
It's been a while since I really geeked out and posted an article that was extremely focused on methods of collecting market research data. This one has been rolling around in my brain for some time. For awhile now, I've wondered why more people don't use IVR with more regularity as a method for data collection. I've come to the conclusion that researchers may be less familiar with the process, probably don't completely understand how it operates, and don't really make the connection to its benefits.
IVR or Interactive Voice Response has been around for quite some time. About once a week, I receive a phone call from one of my children's schools telling me that they were absent from a class (most times we knew about it). Candidates in the political arena often use the technology for "Get out the Vote" campaigns. Though it can't be used for research purposes, basic IVR phone technology is built into the structure of most company's telephone systems (ex. when the "phone tree" asks for the extension, your voicemail, etc.)
The 2009 version of Rockhopper's "Research Industry Trends" report shows us that only 11% of companies actually use this market research data collection method. I really wonder why the technology is used so infrequently, especially given that it has many of the same advantages as online research, and overcomes some of the barriers. Here are some advantages to using IVR as a method of collection:
- IVR deals very effectively with the self selected sampling methods comparable to what you see online (inbound calls from "current customers"). This is the IVR use that most people are familiar with.
- Most IVR systems have the ability to make outbound calls. IVR is a capable method for Random Digit Dialed (RDD) or client listed outbound studies.
- Because IVR is automated, the labor associated with the job is minimized. The net result is a cost per interview that is much less expensive than "typical" phone studies and similar to online costs.
- Similar to online research, the turnaround time on IVR studies can be much faster for outbound dialed projects, with many of the same benefits of phone dialing and fewer of the "risks" associated with online research.
- IVR is particularly capable of generating qualified responses on low incidence studies. Most systems can be set up to dial through the records and giving the respondent the choice between being transferred to a telephone interviewer (called a Warm Transfer) or completing the interview with the automated system.
- Many IVR systems store data in databases which are easily plugged in to data or research dashboards.
IVR isn't a magic bullet that works for every study (neither is online research...don't shoot the messenger). Here are some project characteristics where you SHOULD NOT consider IVR as a viable method.
IVR should not used for...
- Cell phone dominant samples. TCPA Requirements prohibit the dialing of cell phones using an autodialer. Most phone data collectors are capable of running sample records against a list that identifies cell phone records to ensure adherence to these guidelines when using IVR methods.
- Long survey lengths. The longer the survey, the greater the probability the respondent will terminate the interview. Surveys less than 10 minutes are encouraged. Do not exceed 15 minutes using this method of collection.
- Surveys that have significant screening processes where the phone must change hands multiple times or you must get through some type of gatekeeper. The complexity of the process will lead to an increase in terminated calls.
- Tough for B2B surveys. The method should mostly be used for residential dialing for the same reason as the previous point.
What are your thoughts. Why don't more people use IVR in their arsenal of market research methods of collecting market research data? Are there uses for IVR that I've missed? What about barriers I've overlooked? I look forward to your responses.