Dr. Pete on B2B Research
4/12/2011
Dr. Pete
Dear Dr. Pete:
How can we do B2B research? We have a consumer panel, right?
-H. B. Pencil
Dear H. B.:
You ask a very good question. As we all know, B2B research is about contacting people because of what they do rather than who they are. Traditionally, this has meant calling companies and asking for the relevant person. This approach is quite expensive and time—consuming—after all, haven't we all got better things to do at work than answer questions? Of course when the person asking you what you've been up to recently is your boss, the above doesn't apply.
Anyway—what do you suppose happens to all those working people after 5:30? That's right. They go home and become consumers just like you and me. At this point, they are keen to have someone ask them questions—if only to get away from the kids. So it becomes cheaper and quicker to interview working people outside the home.
There is a theoretical advantage to conducting B2B work outside the workplace. Imagine we wanted to interview secretaries. We could buy a list of companies and telephone the switchboard and say, "Can I speak to a secretary please?" Yes," the operator would reply, "we have hundreds. Whom do you want?" "Uhhh, dunno."
In sampling terms, we would have a non-equal probability sample. Each company has an equal probability of selection. If you are a secretary working for a large company, however, you personally have a low probability of being selected, while if you are the only secretary you have a higher probability of being selected. So the resulting sample is not necessarily representative of all secretaries.
Imagine now that we wait until after 5:30. All the secretaries line up outside work and we sample every nth—now we have an equal probability sample. This is the same as using a consumer panel. All you need to do to get a representative sample of secretaries is select a large representative N and invite only those people within the N who are secretaries. Be careful not just to pull off any old secretary from the panel. Remember, the panel itself is biased in terms of age. We can't set quotas on age and gender of secretaries because we don't know these statistics.
Consequently, for common job types, interviewing off a consumer panel is a better methodology than interviewing at work. Unfortunately, researchers are much more interested in interviewing executive job types like IT Directors and MDs. In this instance we run into practical issues. Such people are rich, time poor and (usually) averse to market research.
Why then would such people join a consumer MR panel? And are those that do join typical of all IT Directors? At this point it starts to make sense to build a specialist panel. We recruit IT Directors, for example, to be in a special panel with some suitable reward, in return for which they undertake surveys for us during their non-working hours. We get the research, they don't get bothered during the work day and they don't get asked about soap powder and other ephemera-simple! All we need to do is find a suitable sample source and the right reward...hey, maybe that's not so simple after all.
The third issue with which we need to contend is when the researcher is interested in getting the views of the company. In this instance, the people themselves are pretty much irrelevant. It is because they occupy a particular position in a particular company that we want to interview them. Here the advantage of the consumer panel becomes its disadvantage.
Staying with the secretaries, let's say now we only want to interview one secretary per company. The company that employs many secretaries has a higher probability of selection and there is a probability of duplication. The company that only employs one has a lower chance of selection. Our sample may not be representative of all companies. We can set quotas on company size, type and maybe location, but these statistics are incredibly difficult to find.
Questions then to ask when thinking about B2B research are:
- Am I interested in a sample of companies or a sample of people?
- Would the job type I'm interested in be more or less likely to join a consumer panel than anyone else?
Good luck!
-Dr. Pete
