In the social sciences areas of academia, qualitative research is a hot topic – which usually means that it turns out to be highly useful in understanding your business’ customers, in B2B or in consumer markets. Moreover, the software tools used by academics are at a stage where they are more or less capable of handling business analytics needs. They remain divorced from existing statistical and analytics tools, but they at least support Excel.
What is qualitative research? It can best be understood as a contrast to traditional statistics, in which researchers focus on numeric data whose models and categories are determined at the start of a particular analysis.
Qualitative research therefore emphasizes non-numeric data such as text, audio and video; seeks to elicit the model as well as the categorization by iterative "coding" or other processing of the "qualitative" data; and often emphasizes letting how the study subject(s) see reality determine how the researcher sees what’s going on.
To put it another way, in statistical and analytic research the business often wants insights that fit within its concept of what a customer is. Qualitative research can let the business understand how the customer sees himself or herself.
Types of Qualitative Analytics
Within qualitative research, there are several key subtypes:
- Grounded theory, perhaps the hottest subtype in academia but the least used in business, in which the researcher seeks to deduce theory from data rather than the reverse. As Wikipedia notes, this typically involves taking text from interviews, focus groups or audio/video, and extracting common themes, which are then refined until patterns emerge that form the basis of a theory of what’s going on.
- Ethnography, in which the people examined are treated as a "culture" in which the researcher immerses himself or herself, letting the viewpoint of the culture determine what is important data and what is not.
- Phenomenology, in which a specific event or "phenomenon" such as a product launch is studied, not in terms of what objectively happened, but in terms of what people believe happened.
- Pragmatic qualitative research in which subtypes 1-3 as well as other related research techniques are applied when best suited to the problem. This type is perhaps the best suited for business.
A key common thread in many qualitative techniques is asking the participant to tell a story. From my own personal experience in survey interviewing, I can attest that this is an enormously powerful technique that yields valuable insights not available in traditional statistical surveying.
For example, it was through asking interviewees to "tell me a story" that I first learned that what was really valuable to SMBs was not 99.999 percent system reliability, but the ability to maintain the system by simple tasks carried out by the branch manager every Saturday evening when he/she closed shop for the weekend.
Qualitative Analytics Software
The Big Three in qualitative research software tools are MAXQDA, ATLAS.ti and NVivo. Differences between the tools are far less important than the fact that these all are complete packages. They all have extensive features for semi-automating the process of eliciting patterns from "content," be it interview transcripts or videos. They all provide neat features for organizing the data that you are analyzing and have ways to ensure user-friendly reporting of results, by export to Excel or in-house capabilities. And all provide support for users. It is interesting that NVivo has a strong relationship with Microsoft, but the differences from the other two packages due to that relationship are less striking than the similarities.
Essentially, these packages are best used to support a pragmatic approach to qualitative analytics by the business. That is, the user should approach each project to "understand the customer" by: figuring out which technique will best capture the insights that the customer isn’t sharing via surveys and on-the-fly data capture (aka Facebook); using the tools to semi-automate the process of gathering the data and eliciting patterns; and manually figuring out how to best report results that relate to an existing customer understanding, including existing customer numerical data that has been statistically analyzed as well as in-house "gut feelings."
Over the last two years, publications such as Sloan Management Review have been filled with stories about the critical importance of the gap between how the executive suite sees the business and how line employees and customers see the business. Looking at social media data statistically only takes you so far, because it is still typically being viewed through executive-suite rose-colored glasses.
An approach that emphasizes satisfying the customer’s needs cannot fully succeed unless it starts to hear the customer’s viewpoint without pre-conceptions – and the best tool we have for that right now is qualitative research.
Of course, at best, qualitative research is still more manually demanding and less quick-to-insight than traditional quantitative analytics. It is best reserved either for "crossing the chasm" revolutionary experiments, "canary in the coal mine" detection of problems ahead or a long-run effort to cement long-term relationships with a relatively stable customer base.
However, as I have noted, the tools and techniques are now mature enough to be applied with some confidence in the quality of the data gathered. Also, note that the three cases I have cited are business-critical ones – and you can’t get much more useful than that.
Qualitative research is ready for business, it’s increasingly being used, and it’s useful. It's time to take a good, hard look.
Wayne Kernochan is the president of Infostructure Associates, an affiliate of Valley View Ventures that aims to identify ways for businesses to leverage information for innovation and competitive advantage. Wayne has been an IT industry analyst for 22 years, focused on analytics, databases, development tools and middleware, and ways to measure their effectiveness, such as TCO, ROI and agility measures. He has worked for firms such as Yankee Group, Aberdeen Group and Illuminata, and has helped craft marketing strategies based on competitive intelligence for vendors ranging from Progress Software to IBM.