Private Online Research Panels
by
Jerry W. Thomas
Over the past decade or so, some companies have set up private online panels as an economical way to conduct surveys and other research projects. A “private
research panel” (sometimes called a “custom panel” or “customer panel”) is one set up by a company solely for its own use. The firm recruits
and maintains the online panel, and uses the panel exclusively for its own research projects. While many private panels are a success, an equal number are deemed
failures. The purpose of this article is to outline some basic guidelines to help you decide if a private panel is right for your company.
Private Panel Versus Online Community
A private panel is sometimes confused with an online community. Typically, a private online panel is made up of thousands of participants, while an online
community tends to be much smaller (i.e., hundreds of members). A private panel is not interactive, but an online community permits its members to interact
with each other. Private panels are used in a controlled, limited fashion, but communities tend to be used intensively and continuously. In general, private
panels are superior to communities as a marketing-research tool. Communities can be valuable in revealing the consumer’s language relative to a product
category or topic, and can be useful as creative stimuli. However, communities are lacking from a research perspective because of respondent conditioning and
interaction biases, and the small size of communities may not permit rigorous sampling controls.
Economic Rationale
A private panel under optimal conditions can greatly reduce the cost of a typical research project, excluding indirect costs and staff time. That $15,000 product
test using an outside research company might only cost $3,000 or $4,000 with a private or custom panel. It is the allure of cost efficiencies that prompts
most companies to look into private panels. A second reason is the panel-member registration data. As panel members sign up to participate, they can be asked
questions about their family, media habits, shopping patterns, product consumption, etc., so that a valuable marketing-planning database is created. Once or twice
a year, you can tabulate and analyze all of this data for marketing-planning purposes.
Types of Research
Private panels are not appropriate for all types of research. If you are trying to survey all U.S. adults, then your private panel is probably useless—since
it most likely contains only your customers, a subset of U.S. adults. Here are some examples of studies commonly done via private panels:
- Product-usage studies
- Customer preferences for new product features
- Customer reactions to proposed advertising
- Customer attitudes about a brand
- Customer reactions to new policies
- Customer thoughts about new services
- Customer satisfaction and loyalty
- Customer evaluation of proposed package designs
- Customer shopping patterns
- Customer media habits
- Customer focus groups or depth interviews
The assumption is that most private panels are largely made up of customers. Private panels generally cannot be used for research studies that require a
representative sample or a noncustomer sample.
Should You or Shouldn’t You?
One of the first questions you must ask is, “Will my company really use the private panel once it is set up?” We have set up a number of private
panels for clients with big research plans that never materialized. They planned to use the private panel frequently, but the burdens and mechanics of proposing
studies to internal audiences, writing questionnaires, pulling samples, tabulating the results, and trying to write reports bogged the whole process down. Therefore,
few studies were actually conducted and these private panels never became economically viable.
Feasibility all depends on the financial details. Every private panel is different, and you’ll just have to sit down and do the math for your particular situation.
As a rule, if you don’t conduct at least one research project per month, a private panel will rarely make economic sense. That is, the expense of setting
up and maintaining the private panel will be greater than the savings on the research studies.
Another consideration is the subject matter of your research. If you want to set up a private panel of golfers, or pilots, or sailboat captains, or motorcycle
owners, a private panel will work like a charm because the respective panel members would be very interested in these topics. It would be easy to recruit
panel members, and participation rates in surveys would be high. Conversely, if you want to set up a panel of insurance owners, or water-utility customers,
or banking customers (comparatively low-interest categories), it will be much more difficult to recruit panel members, and response rates to surveys will
be low. High-interest product categories are the best candidates for private panels.
Lastly, the lower the incidence of your product category, the greater the potential savings from a private panel is likely to be. For example, if your
product category happened to be experimental aircraft, the percent of the population (i.e., incidence) of those who build experimental aircraft is far less than
one percent of the U.S. adult population. If you had to screen large probability samples of U.S. adults to find experimental aircraft builders, every study would
cost a small fortune. But, if you owned a private panel of these aircraft enthusiasts, studies could be conducted at modest costs.
Basic Cost Components
So, you ask, what does it cost to set up a private online panel? There are four main cost components:
- Website design and database programming. This creates the platform for recruiting and maintaining the panel. Initial costs
range from $20,000 to $50,000, depending on how fancy the website is.
- Monthly panel maintenance. Someone has to answer emails, phone calls, and letters. Someone has to host the database, update
records, delete nonresponders, monitor the firewall, oversee encryption of sensitive data, etc. These costs usually range from $1,000 to $2,000 per month,
depending on the size of the panel.
- Recruiting costs. You may have a database with one million customer email addresses, and you can quickly recruit thousands
of panel members at a cost close to zero. However, if you have to go out and recruit panel members from scratch, it can easily cost $10 to $20 per new
panel member. After the initial recruiting to launch the panel, you will have to periodically recruit new members as old panel members drop out.
- Software costs. You will need easy-to-use software that will allow you to design, program, execute, and tabulate relatively simple
surveys. If you design complicated surveys, you will probably need professional programmers to help you.
These are the basic costs to set up a private panel. Other potential expenses include firewall software, emailing software, encryption software, database
software, respondent-authentication software, and so forth.
Internal Staff Versus Outsourcing
If you decide to set up a private panel, should you do it yourself with internal staff, or should you outsource it to a research firm with “private panel”
experience? The larger your company, and the greater the number of projects, the more likely it is that you should just staff up and do it yourself. If you
plan to do more than a hundred research projects a year, then building a professional “internal” staff probably make sense. Remember, you will need to
hire some individuals with research-supplier “operations” experience—folks who really understand the nuts and bolts of sampling, questionnaire design,
coding, tabulation, and so on.
If your company will likely do fewer than 100 projects a year, you probably should outsource the private panel to a research company with expertise in private-panel
development and management. The research firm will manage your panel, answer emails and phone calls, provide all the software and programming, design questionnaires,
conduct the surveys, and even write the report for you, if you want that service.
This approach frees you up to be the research consultant within your organization, rather than the research technician. You can operate with minimal staff, since
much of the work is being performed by your panel partner.
Recruiting Tips
Recruiting a private panel can be a challenge. The best way is to collect customer email addresses religiously—at every point of contact with customers.
Then you will have a massive database of customer email addresses from which to recruit your private panel. If you do not have email addresses, then panel
recruiting tends to get expensive. You can insert invitations in packages to invite customers to join the panel. You can mail letters to targeted groups
inviting them to join. You can purchase opt-in email lists. You can send out publicity releases announcing your panel. The best method of recruitment will
vary by product category, and you will have to experiment to identify the most cost-effective methods.
Beware of Threats
Once you have recruited your private panel, you will need to be careful that a few panel members don’t join your private panel multiple times or complete
the same survey more than once. If you have a professional research firm manage your private panel, that firm would automatically take care of these potential
problems. If you manage your own panel, however, you will have to search your database of panel members on a regular basis looking for duplicate panel members.
You may also have to resort to “digital fingerprinting” to identify duplicate respondents. You may have to track response by Internet service providers
to make sure your survey invitations are reaching respondents. You must stay on the alert for potential hackers.
Caveats
A private panel can be a useful tool in your research toolbox, but it is just one tool. You will still need professional research agencies to help your company
with studies that require probability samples (market segmentation, bench-marking, volumetric forecasting, etc.). You will need research suppliers to help with
central-location studies, in-person interviewing, and qualitative research. You will probably still need professional research firms to help with very complex
studies (choice modeling, conjoint). You will also need professional research companies to help with major strategic projects, where objectivity and outside
perspective are crucial. However, for the day-to-day studies among your customers, private panels may be a way to get more research bang for your investment.
Copyright © 2011 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
Jerry W. Thomas (jthomas@decisionanalyst.com) is President/CEO of 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
Online Research Services Services
Online Research Case Histories
Online Research White Papers
