Using GenAI for Methodology Selection and Participant Screening
In the first installment of this series, we explored how generative AI tools (GenAI), including ChatGPT and Claude, can be employed pre-project to come up to speed, refine research objectives, and develop hypotheses.
In this article, you’ll see how GenAI can be used to determine a research methodology and develop an airtight screener.
Deciding on an Approach
Whether you dabble in market research or are a seasoned veteran, GenAI can be a useful collaborator on research methodology. You can input the objectives and hypotheses, and the tool can advise whether a quantitative, qualitative, or mixed methods approach would be most appropriate.
Example prompt:
A B2B Technology Solution Provider would like to understand what their Ideal Customer Profile looks like. Could you advise whether a quantitative, qualitative, mixed methods, or another approach would be more appropriate?
Claude.ai response:
For a B2B technology provider looking to understand their ideal customer profile (ICP), I would recommend a mixed methods approach combining both quantitative and qualitative market research:
-
-
- Quantitative research through surveys can help identify firmographic patterns in your current customer base – industry, size, revenue, etc. This can start building a profile of your best-fit customers.
- Follow up with qualitative research through customer interviews to add color, detail and deeper insights around motivations, values, pain points and needs of your best customers.
- The qualitative findings can help turn the firmographic data into a richer, more nuanced profile of your ideal customers.
- Quantitatively segment your market using the ideal customer profile attributes to quantify and prioritize the most attractive segments for targeting.
- Supplement ongoing quant and qual research through buyer interviews and win/loss analysis to continually refine your ICP.
-
The combination of broad firmographic data from surveys and deeper insights from interviews will yield a more complete view of your ideal customers versus using either method alone. The mixed methods approach provides the needed qualitative context to turn quantitative data into an actionable ideal customer profile.
In addition to recommending an approach, GenAI tools can suggest the ideal project sequence. While qualitative research is often conducted first to explore topics in-depth and uncover themes to then be measured quantitatively, the reverse sequence would be more appropriate for the ICP project described above:
Example prompt:
How should we sequence a mixed method approach to understanding our ICP?
Claude.ai response:
Here is how I would recommend sequencing the mixed methods approach to define the ideal customer profile:
-
-
- Start with quantitative survey of current customers to gather firmographic data and identify demographic patterns.
- Follow up with qualitative interviews with current high-value customers to gain insights into needs, values, and pain points.
- Develop an initial ideal customer profile (ICP) incorporating findings from quantitative and qualitative research.
- Use ICP attributes to quantitatively segment total market and prioritize most attractive segments.
- Conduct qualitative research with prospects in target segments to refine understanding of needs.
- Iterate on ICP with ongoing quantitative segmentation and qualitative research with prospects.
-
The sequence enables starting broad with quantitative data, refining with qualitative insights, then using the ICP to guide additional quant and qual research in an iterative mixed methods approach. The back and forth between the two methods will continue refining the ICP over time.
Even for experienced market researchers, GenAI tools can recommend different techniques and methodologies beyond your specific familiarity and expertise, or simply remind you of those not top-of-mind.
Example prompt:
Given the research interests I’ve described please list the set of research techniques that can be used to investigate the subject.
This prompt returns an extensive list including methods such as journey mapping, sentiment analysis of customer or social media discussions, and immersive ethnography. The tool can further describe each method and steps to execute it, allowing you to home in on tailored ways to answer the original question. The tool can also consider factors like timeline, budget, objectives, and resources when suggesting a research plan.
Other methodology considerations can come into play for a given market research project, and GenAI can help here as well. For example:
Given the chosen methodology, where could this project run into feasibility problems?
The scope of this research needs to be reduced. Where could cuts be made, and what would be sacrificed in the final report because of those cuts?
We’re initially conducting this research in North America but will eventually field it globally. Is there anything about the current research that could present a problem when it goes global?
The tool provides an informed perspective rather than a definitive answer, with comprehensive input to aid your decision-making. With each iteration and additional detail provided, GenAI can refine its methodology recommendations to best fit the project context.
Developing a Screener
Regardless of methodology, a screener is a critical part of any research project – ensuring that respondents are a good fit for the study. GenAI can be leveraged to refine a draft screener by adding or removing criteria based on the target audience and research objectives. You can use GenAI tools to walk through a checklist, helping to improve the quality of the screener at each step.
Example Prompts:
Please review the following screener against the previously provided research objectives and suggest enhancements.
Based on the research objectives, are there other segments that could add insights to the study that are not covered or are screened out by this screener?
Review the screener and identify any questions that are redundant.
How can the screener be streamlined?
Review the screener and identify any questions where the response options are incomplete. Provide additional options to cover all likely respondents.
How long do you think it would take a candidate to complete the screener if delivered as a web survey?
Is the sequence of questions in the screener optimal? How would you recommend improving the flow?
Can you suggest ‘red herring’ options to catch fraudulent participants?
Not all the recommendations will be useful, and as a researcher you can take or leave suggestions as you see fit. Even if none of the specific suggestions add value, it’s almost certain that this kind of interaction with a GenAI tool will result in a better screener.
Reach out to us today at [email protected] to learn more about how GenAI can be used to determine a research methodology and develop an airtight screener.
Read more in our AI for Market Research series
Part 1: Objectives, Onboarding, Methodology
Part 2: Methodology Selection + Participant Screening
Part 3: Data Collection Instruments
Part 4: Quality Control + Preliminary Findings
Part 5: Final Deliverables
Part 6: The Future + Synth Data (Coming Soon)