You are here:
Home |
White
Papers | Finding an Internet Research Supplier
Download PDF
Version
Finding an Internet Research Supplier
By
Michael Richarme
Quietly breath the words "research," "survey," or
"opinion" in your office. Go ahead, I dare you. Within a few minutes,
you'll have several calls from market research suppliers wanting to help you
find out exactly what your customers think about pretty much anything.
To a degree, it's always been that way. Like consultants, researchers have
pretty fine-tuned radars when it comes to finding business people with
questions. And, like consultants, there are lots of variations in the
quality of the work product that you may receive. This situation has been
amplified over the past few years with the rapid acceleration of Internet
surveys as a primary methodology, rapidly replacing snail-mail surveys or
dinnertime-interrupting telephone surveys.
Having been in the full-service research business for the past 26 years, and in
the Internet research business for the past 8 years, Decision Analyst has
worked with clients and research industry associations around the world to
provide solid answers to their marketing research questions. In the
process, we've learned the three major differences between quality Internet
research and fly-by-night operators. Businesses thinking about conducting
Internet research can use these as acid-test questions when talking to research
suppliers.
The first major difference is in programming skill. While
it might sound simple to write a program to ask questions and collect data in
a database, there are research techniques that are important in presenting questions
to respondents that can be challenges to program. These may include automatic
skipping and renumbering for non-relevant questions, response answer randomization,
variable piping to subsequent questions, presentation of stimulus materials,
and the like. Sometimes these features are included in the inexpensive
survey packages floating around on the Internet, but generally they are not.
Closely related to this area is the back end of the survey process. This
would include survey security, which includes making sure that hackers can't
get into the survey or the database, and server capability, which handles huge
volumes of simultaneous responses to surveys. For less-experienced
research firms, these problems often rear their ugly heads right in the middle
of data collection.
The second major difference is in panel management. Where
mail and telephone surveys give researchers better abilities to gather random
samples, which produce more meaningful results, there is no telephone directory
for the Internet that provides this capability. So the leading research
firms have invested millions of dollars in building large panels of willing
respondent, from which pseudo-random samples can be drawn.
The best panels are those with double opt-in members. Those are members
who have agreed to be on the panel and answer surveys (and have provided
demographic information to the researcher so that "representative"
samples can be drawn, and who agree to participate on a specific survey as a
member of that panel. Start-up research suppliers might be tempted to
purchase an email list from a willing Internet Service Provider and claim that
is their panel, without ever contacting any of those members.
A related issue in the panel management area is compensation to the
respondents. Some research suppliers try to get by with inexpensive
trinkets or points in exchange for the respondents information, but this tends
to backfire rather quickly. Established firms who have experience
in managing panels realize that, in order to get high-quality, thoughtful
answers to their research questions, cash should increase with the length of
the survey or the scarcity of the respondent, reflecting the increased
importance of the answers.
The third major difference is in the area of research skills.
Experienced firms know how to ask the right questions, in the right sequence,
with the proper stimuli or scales, to get a meaningful answer from respondents.
Developing the right survey instrument is at the heart of the marketing research
field - if done properly it is brief, direct, and powerful, but if done improperly
it is lengthy, confusing, complicated, and ultimately will produce suspect information.
A lot of research firms competing in the rapidly emerging Internet space have
one or more of these skill areas. Some may be great at panel management
and research, but not understand Internet programming. Some may be great
programmers, but not understand working with those pesky respondents. A
good research firm will have all three of these skill sets in abundance.
Anything less will likely result in poor quality responses that may not answer
the original business question. Even worse, they may provide incorrect
answers that result in a bad business decision being made. It is truly
amazing to see the number of otherwise brilliant businesspeople who slice
marketing research budgets in an attempt to save money, and then turn around
and make new product decisions, advertising placement decisions, pricing
decisions, channel decisions, or other market positioning decisions with faulty
input from low-quality research.
When searching for a great marketing research company that is active in Internet
research, make sure that all three bases are covered, and keep probing until
you are satisfied that the firm understands and incorporates the three major
areas of difference described above.
Copyright © 2005 by Decision Analyst, Inc.
This article may not be copied, published, or used in any way without written
permission of Decision Analyst.
About the Author
Michael Richarme (mrichar@decisionanalyst.com)
is a Senior Vice President at Dallas-Fort Worth based Decision Analyst. He may
be reached at 1-800-262-5974 or 1-817-640-6166.
Additional Resources from Decision Analyst
Related Services
Related White Papers
