The advent of generative AI has injected high-voltage electricity straight into the market research ecosystem. Established software providers are infusing their products with this elixir, while new startups storm the field.

Market research software vendors are building Generative Pre-trained Transformers (GPT) capabilities into their products, and companies like Shapiro+Raj are equipping their in-house proprietary tools with these magic wands to be able to shred through transcripts and extract relevant quotes.

Available to All

In this article, we’ll steer clear of the special-purpose and in-house tools and focus on sharing general-purpose generative AI that anyone can use to up their insights game.

Before you dive in, it’s important to check your company’s policies with respect to these tools, as every company has different policies on the use of generative AI.

Throughout this series, we’ll trek through the AI frontier, but let’s start where it all started – ChatGPT.

ChatGPT for Market Researchers

First, it’s important to note that if you’re going to use ChatGPT you must go into the settings and turn off history and training to protect the confidentiality of your data. If you’re using ChatGPT Plus, you can also opt your data out of being used for training (you can use the form linked here). Without these precautions, any data you input into ChatGPT could end up being revealed to other people who use the AI.

ChatGPT can save time and help at each stage of the research process by improving the quality of surveys and discussion guides, improving alignment with objectives, and supplying inspiration from initial criteria. While it may not as well as an experienced researcher, it can make sure you cover your bases by critiquing and ideating at each step, freeing you up to do more high-value work.

Setting up the Research Project

The best and easiest time to course-correct is before the project begins. Once a project picks up momentum, changes become much more difficult. ChatGPT can be an eager assistant in the pre-project stage to ensure everything is on target.

Onboarding

ChatGPT can act as your personal learning assistant to get up to speed when encountering unfamiliar concepts. Rather than searching and sifting, you can have a genuine (albeit artificial) conversation with the bot, asking to drill down on some topics while moving on with others.

ChatGPT completed its training in 2021, so its knowledge isn’t up-to-the-minute. You can augment ChatGPT with Bing Chat, Google Bard, or Perplexity.ai, which are more current (but less eloquent).

Example:

Ask ChatGPT: What is GPT?

Follow up: Explain transformer neural networks.

ChatGPT is also particularly good at creating a personalized tutorial for you to learn a topic within your research situation:

Example:

Ask ChatGPT: How can GPT influence CRM outcomes?

GPT:

2, Smarter Lead Scoring: GPT can analyze customer interactions, emails, and social media data to better understand customer preferences and behavior. This analysis can lead to smarter lead scoring, helping sales teams prioritize and focus on high-value prospects.

3. Sentiment Analysis: GPT can perform sentiment analysis on customer interactions and feedback, enabling companies to quickly find and address potential issues and opportunities for improvement.

The response included 11 ways that GPT can enhance CRM capabilities. While some are repetitive or not relevant, the output will get you thinking about all the possibilities available.

Research Objectives

You can use ChatGPT to help round out research objectives and ensure the project generates comprehensive insights.

Example:

Ask ChatGPT: My stakeholder wants to test messaging with small businesses about how they are planning on providing GPT functionality in their CRM product. Please list out some potential research objectives for this study.

ChatGPT:

Message Recall: Measure the level of message recall among small business owners to determine how well they remember key points about the GPT-powered features after exposure to the messaging.

Differentiation Perception: Determine whether the messaging effectively highlights the unique selling points of our CRM product’s GPT integration compared to competitors’ offerings.

Barriers Identification: Identify any potential misunderstandings, misconceptions, or reservations that arise from the messaging about GPT-powered functionality in the CRM product.

In total, ChatGPT output 15 potential research objectives. Several are obvious, while others are out of scope, or not aligned with the stakeholder’s needs. However, “differentiation perception” and “message recall” might be worth adding to the project objectives.

A list like this might also trigger you to think of whole new objectives of your own.

Formulating Hypotheses

Once you’ve identified your research objectives, you can use ChatGPT to help brainstorm some hypotheses to test.

Example:

Ask ChatGPT: Based on the following research objectives, suggest some hypotheses for the research to test.

ChatGPT:
Hypothesis 1: Different messaging approaches will result in varying levels of perceived effectiveness in conveying the benefits of GPT-powered functionality in the CRM product to small businesses.

Hypothesis 2: Small business owners exposed to messaging with clear and comprehensible language will demonstrate higher message recall compared to those exposed to complex or ambiguous messaging about GPT integration.

In this case, the hypotheses are incredibly basic and boil down to “good messaging will work better than bad messaging.” It’s important to remember that you’re having a conversation with ChatGPT. If the first response misses the mark, ask for a more refined answer.

Ask ChatGPT: These are quite basic. Use your knowledge of GPT and CRM and think of some testable hypotheses for the research.

ChatGPT:
Hypothesis 1: Small businesses exposed to messaging emphasizing the time-saving capabilities of GPT-powered automation in the CRM product will demonstrate a higher likelihood of adopting the technology compared to those exposed to messaging focused on other benefits.

Hypothesis 2: The integration of GPT-powered natural language processing in the CRM product will lead to a significant improvement in small businesses’ ability to analyze and understand customer sentiment, resulting in higher customer satisfaction scores.

Better. As a smart researcher you might look at the first hypothesis and think, “What does the client think is the most important attribute being presented? If it’s time savings, we should test whether those messages rate higher.”

Conclusion: Expanding Possibilities with AI

In this article, we explored how generative AI can help at the outset of a research project by broadening perspective and possibilities. ChatGPT can rapidly generate potential objectives, hypotheses, and use cases – helping to ensure the research covers all bases. While not every AI-generated idea will be relevant, having a wide pool of options lets you select the most insightful ones for your specific research needs. By conversing with ChatGPT to learn about and ideate on unfamiliar subject matter, researchers can overcome limited viewpoints shaped by their own experiences.

The future of insight is one of partnership – with AI expanding what’s possible, and researchers guiding what’s valuable.

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)