The objective of this type of study is to quantify the revenue potential of a possible new asset and ideally provide some depth of insight to support a go/no decision. In other words, answer the key questions: how much? And why?
To answer the “how much” question, conventional thinking tells us that you need a large, projectable sample, large enough to allow you to produce share estimates with a strong enough level of confidence to satisfy management. Answering the “why” question can be done with a few open-ended questions embedded in a survey. However to get deeper insights, we usually suggest bringing in a moderator to ask probing questions that get below the surface.
In our video Chris and I felt that if timeline and budget allow, it really is best to conduct separate qualitative and quantitative studies, rather than trying to achieve both objectives through a single data collection effort. In an ideal world, we would have sufficient time to design and execute an exploratory qualitative phase, provide the insights from this research back to the client team, identify the key questions that need to be quantified, then design and execute a survey to capture this information from a large, projectable sample. (In some situations, it makes sense to do the quantitative phase first, then use the follow-up qualitative research to dig deeper on questions that come up in the quantitative phase.) But this whole process can take several months.
What if timeline and budgets don’t allow for this? Is it possible to achieve the same objectives if there isn’t time for two separate phases of research? We believe it is, as long as some careful planning is done at the outset.
In a recent project, we decided with the client that their information needs were best met through a quantitative survey with physicians. However, the client was also interested in gaining some deeper insights on a few targeted issues, as follow up to several key questions in the survey. The client wasn’t satisfied with just addressing these by including some follow-up questions in the survey; he wanted a moderator to do some thoughtful probing around these questions. In addition, timing was tight; these qualitative insights needed to be delivered quickly.
Our solution was to flag the physicians who we wanted to interview in greater depth based on their responses to the survey. In this case, we identified three specific selection criteria, which were all based on self-reported behaviors that were of interest to the client. If a physician’s survey responses met any of these pre-determined criteria, they were shown a message at the end of the survey inviting them to participate in the follow-up. (They had pre-agreed to participate in a possible follow-up interview in the initial screening, per BHBIA standards.) They were then taken to a scheduler screen where the respondent selected and confirmed the date and time for the interview.
Whenever a survey respondent was scheduled for a follow-up interview, this information was communicated to our project team and to the moderators we had on stand-by, along with the questions to be explored in the follow-up. In some cases, we were able to conduct the follow-up phone interview within a day of the physician having taken the survey, which meant that their survey responses were still fresh in their memory. All of the telephone interviews were completed quickly, enabling us to include these qualitative insights along with the quantitative survey findings in our final report. The client received the statistically robust quantitative metrics they wanted, along with deep qualitative insights on specific selected issues, without the need for an extended project timeline.
This approach succeeded because the client had very specific targeted objectives for the qualitative component – to shed light on a few key issues. If the qualitative needs are more broad-based – for example, developing a comprehensive understanding of the market landscape or buying process in a treatment area – then clearly this “short cut” approach wouldn’t be appropriate. However, it demonstrates that hybrid qual/quant techniques can work in the right circumstances.