Working for a communications company has helped improve my listening and communication skills. You’re probably wondering “so what, isn’t that a given?” I didn’t think so. I thought these tools were pretty well sharpened, after 20 successful years in business.
I earn my living by conducting marketing research. I gather data; provide insight and advice on, well, nearly any product or service.
Part of my role is to provide first-hand, current data to help support Public Relations efforts. This is an exciting part of my job. I get to work with very smart and creative people, across a wide variety of industries and clients. I was accustomed to working with other marketing research folks, so I took our lingo for granted.
I began to notice a variety of references to marketing research that meant something different to each party and sometimes confused me. A few examples: “I want to run a poll, we’d like a methodology for gathering information, or my client wants to do an Omnibus”, among others. The term “Omnibus” popped up more often than others.
As I explained the differences between an Omnibus and “ad hoc/bespoke/custom” research, I started wondering what the real benefits were for each. Given the numerous changes in our industry, I thought it would be a prudent exercise. After all, providing the best research design to clients is essential to a successful study.
Researchers are usually very curious souls. Being one, I began my investigation. My goal was to simply gather facts and make comparisons to the services that Redshift offers since we also offer research—“one question at a time”.
I selected a mix of companies – one that is an industry standard, a couple that are that are offered by large, well-known, companies, and a couple that offer an Omnibus (or several) as part of their product portfolio, but aren’t especially associated with this service.
As I expected, every company focused on two key benefits: incredible cost savings and amazing speed (vs. custom studies).
My next observation was that nearly every company had a paradigm for estimating costs that seemed a bit time-consuming. I thought, well, if I were shopping around and comparing prices and services, it could get a little frustrating. I also thought it would be difficult to estimate the precise number of each type of question. So I created a couple of examples and calculated both the cost per question and the cost per interview/complete.
This is a summary of my observations, considering a general population study among adults in the US 18 years of age or older.
|Omnibus Companies||Redshift Research|
|Estimating costs||A bit complex and time consuming; requires estimating some variables that may or may not end up being real||Very straightforward. In every scenario, we were still the least expensive- for an ad hoc/custom study.
Pleasant surprise? We hope so.
|Timing||Straightforward schedule- usually 2-3 day delivery||We offer the same|
|Sample source||Panel||We use the same|
|Sample representation||Scary- but many don’t discuss this. Those who do, balance the sample using different variables such as: age, race or ethnicity, region, income (this is an odd one). A couple of companies only mention that their sample is “nationally representative”. The samples varied a bit between 1,000 (generally the standard) to 2,000.
Having a well-balanced and representative sample is essential for publishing.
|We offer 1,000 or customize it for you.|
|Number and type of questions||This aspect varied. A few companies had a limit of 30 questions. However, I have to wonder how many questions in total a respondent could answer without experiencing boredom or fatigue.
After all, the premise of an Omnibus is to have more than one client “buy” a number of questions on various topics. Could some topics be conflicting? Could one topic bias your results? I have to believe that the answer to these questions is no. But it might be good to ask.
The study is yours alone. Your questions and your topic- with advice given by senior –level experts in your field.
|Respondent fatigue||To be fair, a few companies did touch on this subject and restricted the number of questions per client. I would still want to know how many topics and questions I was buying into. I don’t know about you, but I would rather have a respondent’s attention on one topic or perhaps a few that were compatible and flowed nicely.|
|Deliverables||Usually an excel file, with data run by basic demos. Anything extra is extra money.||We offer the same.|
My conclusion: why wait for an omnibus with its associated restrictions to come along when you can have a custom-tailored, at a time scale that suits you? Value for your money if far better.