If you’ve found yourself reading this article, chances are you’re a marketer and not an academic. So why is it important to think about qualitative research design?
The short answer is that how you design a research project directly influences the degree of confidence you (and your stakeholders) will have in the insights produced from it. Fostering a high degree of confidence is crucial to ensuring you can nurture your insights along the journey towards business outcomes.
What is qualitative research design?
Qualitative research design is the process of creating a plan for answering identified research questions. It involves aligning your processes, data sources and research objectives. Carefully thought out research design is essential for all your market research activities (including, of course, your projects in Relative Insight).
Key questions that can help shape your research design include:
- Who am I interested in learning about?
- What data source(s) can I leverage to get answers about this group?
- Is my entire audience represented in the data I have, or is there a need to include additional sources?
- Have I devoted equal effort and attention to the data on all sides of my comparison?
For marketers operating with limited resources and deadlines, it is all about striking a balance between the quality of insights and the effort it takes to generate them. While marketers don’t need to pursue academic standards of research design, the following three principles support good quality insights.
1. Balanced & representative sampling
When you are investigating a particular topic, it often isn’t feasible to capture 100% of the relevant conversation. Instead, you want to make sure that you are designing your research to capture a representative sample that is sufficiently robust to support high-quality insights, but not so large that it overwhelms the available resources of you and your team.
For example, when studying how three different demographic groups discuss a topic, you’ll want a comparable amount of data for each. It doesn’t need to be exact, but try to avoid comparing 100,000 words for one audience against 10,000 for another.
Good sampling also requires understanding the demographics associated with various data sources – Twitter may be a good source for millennials but less so for gen Z and boomers. In this situation, you may need to seek out additional data sources to ensure all audiences are represented.
2. More data doesn’t always equal better insights
One of the most common questions we hear from Relative Insight users as they design qualitative research projects relates to the optimal amount of data. Unfortunately, however, there is no magic number. Deciding how much data is right for your project involves weighing the trade-offs between the volume of data and its quality.
While Relative Insight can analyse language sets as small as 1,000 words, where possible you should strive for a minimum of 10,000. Beyond that, smaller amounts of high-quality data are preferable to a larger amount of low-quality data.
What is the best way to assess the quality of your data?
It’s more of an art than a science but the best starting point is to open your data file! This may seem painfully obvious, but it remains a consistently overlooked aspect of the process. Skimming through and validating that the data meets your needs and is free of elements that might distort your analysis is all that it takes. The old adage “garbage in, garbage out” rings true here.
If you find yourself with questionable data, the data cleaning options in Relative Insight can help strip out duplicates, retweets and spam. In other cases, you may need to seek out additional data that better aligns with your brief.
3. Take an iterative approach to your research
There is no ‘silver bullet’ or qualitative research design formula that guarantees good insights out the other side. The best advice we can give is to start somewhere. If you don’t find the answers to your questions, it is easy to adapt your approach and try again – chances are you’ll probably learn something that helps you refine your analysis in the process.
This iterative approach is particularly useful for agencies working on pitches or proofs-of-concept for clients. In these situations, analysing a limited amount of data to distil headline insights is often all that’s needed. When you win the contract, you can always go back and build a more robust analysis and source additional data.
You can’t find gold if you don’t go looking
Finding insights isn’t that different from mining for gold – data sources are the mines and insights are the gold. Some mines are more bountiful than others, but you never know what you’ll find until you start exploring. Be thoughtful about your qualitative research design to maximise your chances of finding riches, but don’t be deterred if you come up short. Each attempt will help you learn so you can improve your odds next time around.