14 Chat-GPT Prompts for Faster UX Research

Looppanel is an AI-powered analysis tool purpose built for User Research. Request a Free Demo to discover insights 10x faster.

If you've used ChatGPT, you know that prompting it to get average output is relatively easy. But prompting it to give you exactly what you wanted can take some work, iteration, and time.

Don't worry—in this article, we'll help you skip to the good stuff. We'll walk you through the different use cases for ChatGPT in User Research and arm you with specific prompts for each. We'll cover:

Post this article, you're going to be a wiz at using ChatGPT to make yourself 10x more efficient at work.

Having said If you're looking for a purpose-built tool that will give you the benefits of AI, without the limitations of ChatGPT (cough hallucination cough ), you can check out purpose-built AI tools like Looppanel.

ChatGPT is very versatile and can be used to speed up various types of UX research processes, ranging from analyzing data for a usability test, to helping you generate discussion guides. In this section we’ll cover how you can use ChatGPT for various UX Research methods and use cases.

Here are 5 of the most popular use cases for using ChatGPT in UX Research.

1. How to use ChatGPT for secondary research

Sometimes we don't need new research—the answer already exists (somewhere in the universe). What better way to wow your team than to say—we don't need 3 weeks and $10,000, I can tell you the answer to your question right now.

If the answer exists, ChatGPT (or it's internet-updated cousin Perplexity.ai) can find it for you.

Yes—you should double check anything it tells you, but even still you'll probably save a few hours of rooting around at the bottom of the internet by using these tools for secondary research.

If you need more up-to-date data, we recommend Perplexity > ChatGPT because it's updated real-time with Google data, while ChatGPT is a bit out of date.

2. How to use ChatGPT for creating UX Research Questions

Yes, we all have old templates of discussion guides to lean on, but we all also have those questions we struggle to come up with or phrase in an unbiased manner. My personal nemesis: pricing questions. So tricky to word.

When you're creating your discussion guide or survey questions, ChatGPT can be a really handy sidekick. I use it in 2 main ways:

1. Coming up with a rough draft of a guide or a section of a guide
2. Providing re-phrased versions of a question with less bias

3. How to use ChatGPT for analyzing usability tests, user interviews & focus groups

Once you’ve run your research, you of course have a huge pile of data to get through. We all know it can be stressful and time-consuming to process all of it—not made easier by your stakeholders asking, "Why don't we have the insights yet?" every single day.

ChatGPT can be really helpful in this scenario to:

1. Providing quick call summaries to share out to your team
2. As a research sidekick to help you analyze transcripts and notes faster

We'll dig into the prompts and guidelines to do this well, but one warning upfront: ChatGPT isn't built for user research. This means the tool, while very powerful, comes with 3 major limitations:

1. It can make stuff up
2. You can't trace it's "answers" back to your source data—leaving you asking, where did this insight come from?
3. It's not secure

We have some recommendations on the security piece we'll explain later, but if you want to side-step all these issues with ChatGPT and still get the benefits of AI, you should try a tool built for analyzing user research data.

4. How to use ChatGPT for cleaning & analyzing surveys

Anyone who's run surveys recently knows there are 2 major issues that remain unsolved:

1. Fake and poor quality responses (ironically, AI being responsible for a lot of them)
2. Analyzing open-ends in surveys

ChatGPT can be really handy for both of these.

For fake responses, while it won't catch the more sophisticated bots, you can use ChatGPT to identify speeders (folks who fill out your survey too quickly), straight liners (give the same response to every question), or those write gibberish / inconsistent responses in your survey.

And unless you want to hand-code 300 open-ended survey responses, NPS verbatims, or app reviews, you can use ChatGPT to give you a summary of the insights emerging.

As mentioned above, the issues of traceability and trust remain with ChatGPT. It'll tell you 60% of your respondents talked about hating customer support, but you won't be able to verify or cross-check its work without doing it all manually yourself.

That's the reason we built AI-powered open-end analysis on Looppanel. The product analyzes open-ends for you in minutes, but it also gives you citations for where different themes came from. Speed + trust.

To see it in action, you can book a personalized demo of Looppanel here.

4. How to use ChatGPT for reporting

"Yay, we get to make presentations!"—said no one ever.

What many stakeholders don't realize is that even after you've got your insights, packaging them in a way your team can digest is its own leg of the race.

ChatGPT is really handy here to help you rephrase insights, simplify language or even create report outlines.

Okay, so we know what we can use ChatGPT for, but how might we use it?

Here are 14 Chat-GPT User Research prompts you can use for the different stages of the research process. Use these Chat-GPT UX Research prompts as guidelines on how to structure your prompt. Definitely customize them per your needs and add your team / company / product's context to get the best results.

When giving tasks to ChatGPT, remember to think of it like an intern on your team. It doesn't have all the context you do, but with the right instruction, it can do a pretty good job.

For complex planning and analysis especially, you'll want to break down the task or data into smaller steps. For example, instead of asking ChatGPT to write an entire essay, you want to ask it to right one section at a time—it'll do a much better job and you'll be able to edit your prompts along the way.

And without further ado—here are our tried-and-tested ChatGPT prompts for UX Research!

UX Research Planning

1. Use Case: Read and analyse related secondary research

Provide a brief summary of these articles/research reports: [insert links]

2. Use Case: Find the right UX research method

I need to do a UX research project for my product X, a [describe the product]. The research goal is to find out [describe the goal]. The target audience is [add user demographics, numbers, and relevant details]. What research methods would be right for this project? Give me a broad research plan for each option.

3. Use Case: Create a research plan to test a new feature

‍Create a user research plan to explore the feasibility of a new feature for our product X. X is a [Insert details]. The new feature Y does [details]. We want to understand our users’ possible needs, their openness to Y, and any potential concerns they might have. Please include sections in the plan for user interviews, surveys, and usability testing, within a Z-month timeline (customize this section based on your needs!)

4. Use Case: Generate potential user personas

‍We are in the early stages of developing a new product X, a [describe the product]. X can [describe product features and tasks]. Please generate a list of potential user personas in detail, that we should consider in our design and development process.

Data collection (+ using ChatGPT for UX research questions)

5. Use Case: Create user interview scripts

‍Create a user interview script of 15-20 questions as a UX researcher, to gather detailed insights from users of X app that [insert product details]. Include warm-up questions about the user’s day and work life. Covers key topics such as user needs, preferences, and experiences with similar products. Ask about A, B, and C features. Use simple language. 

6. Use Case: Design feedback surveys

‍Create a short 5-minute user feedback survey for users of product X, which does [insert details]. X provides features that include [add features] for users who are  [describe users]. The survey should cover aspects such as ease of use, satisfaction with features, areas for improvement, and overall user experience. Please include a mix of question types, such as multiple choice, Likert scale, and open-ended questions.

7. Use Case: Generate test scenarios for usability testing

‍I’m doing usability testing for X product, a [describe the product]. X does [list the features] and is designed for [describe the users]. Please generate test scenarios that cover key user tasks and interactions with X.

8. Use Case: Create a script for usability testing

‍Create a usability testing script for product X, a [describe the product]. X’s features include [list the features] and it is designed for [describe the users]. The script should guide the user through [describe tasks] with X, and provide instructions for the tester on what to observe and record. Please include sections in the script for the introduction, task instructions, and closing remarks.

9. Use Case: Create a script a focus group discussion

‍We are conducting a focus group discussion for X product, a [describe the product].  X does [list the features] and is designed for [describe the users]. Please draft a discussion guide for a focus group on the latest update that allows Y. The guide should cover key topics such as user needs, preferences, and experiences with X and the update Y.

Data analysis

10. Use Case: To generate insights

‍Here are the results of a product feedback survey conducted on product X’s latest feature Y. The product X does [list the features] and is designed for [describe the users]. Summarize the survey findings as 5-6 key insights.
[Paste survey results]

11. To identify patterns and trends

‍Read the 5 condensed interview transcripts pasted below and identify the common patterns and trends in user experiences based on them. The product X does [list the features] and is designed for [describe the users]. Summarize the patterns and trends into bullet points. Also, share summaries of user responses to each question as a tabular column.
[Paste interview transcriptions]

Note: for interview transcripts beyond a certain length, you may run into context limits. This means the system can't take all the data in at once. You'll need to break your transcripts into chunks and paste them into ChatGPT. For any new prompts you provide, remind ChatGPT to refer to historical content you've already provided.

Alternatively, if you want to skip jumping-through-hoops to get ChatGPT to work for you, you can try an AI-powered tool built for user research analysis like Looppanel.

Sharing Research Findings

12. Use Case: Rewrite research insights

‍Given the following research data about our product X, a [describe the product], please provide a concise summary of this insight. Use simple, readable language that can be shared in a report. [Paste data + insights]

13. Use Case: Outline a research presentation

‍I’m preparing a presentation to share our user research findings on product X’s new update Y with stakeholders. X does [list the features] and is designed for [describe the users]. Please create an outline for the presentation that covers the research objectives, methodology, key findings, and recommendations. [Paste key findings here.]

14. Use Case: Write an article on research findings

‍I need to write an article that summarizes our user research findings about X with the rest of the organization. Read the findings and insights pasted below, and draft an article that’s engaging, informative, and accessible to a general audience. Avoid UX jargon, keep it action-oriented, and highlight the outcomes of the research.
[Add key findings to share here.]

4 Bonus Prompts using ChatGPT for UX research

1. Use Case: How to use ChatGPT for competitor analysis

‍Prompt: I’m a UX researcher working on a competitor analysis for our [industry/product type]. Can you help me analyze our top [X] competitors? Please provide insights on their product features, user experience design, and any unique selling points they highlight. Also, include an overview of their strengths, weaknesses, and potential gaps in their offerings compared to our product. If possible, suggest opportunities we might leverage to differentiate ourselves.

2. Use Case: Define key business metrics

Prompt: I need to define key business metrics for a [industry/product type]. Can you help me identify the most important metrics to track for understanding user engagement, customer satisfaction, and product performance? Please explain why each metric is important, how it can be measured, and any industry benchmarks or examples that would help set realistic targets.

3. Use Case: Identify your users and their behavior

Prompt: As a UX researcher, I’m trying to better understand the users of our [product/service]. Can you help me define the key user personas and their behaviors? Please consider factors like demographics, psychographics, goals, pain points, and motivations. Additionally, suggest methods for gathering data on these users and how to interpret behavioral patterns to improve our product’s user experience.‍

4. Use Case: Reaching the Target Audience for Insights

Prompt: I’m looking to gather insights from our target audience for [product/service]. Can you guide me on the best strategies to reach and engage this audience? Please provide suggestions on effective recruitment methods, channels for outreach, and how to incentivize participation. Additionally, outline the types of questions or surveys that would yield the most actionable insights for improving our product.

ChatGPT is great, but it's got some serious drawbacks we've referred to above:

1. It makes stuff up which makes it hard to trust its output
2. You can't trace any of its output back to the source data—again contributing to our trust issues with ChatGPT
3. It has limits on how much data you can put in at a time, a.k.a., you have to spend time breaking your transcripts into smaller chunks it can process
4. It's not connected to the rest of your research workflow. so you're hopping between miro boards, zoom recordings, transcripts and ChatGPT (which gets old fast)
5. It takes a while to optimize your prompt for a specific use case.

These limitations can put a big damper on any efficiencies ChatGPT is able to buy you.

For more complex use cases where trust and accuracy are important, we recommend using a dedicated tool built for User Research. These tools can give you the benefits of AI, without the limitations mentioned above.

Let's dig into our top 3 ChatGPT tools for User Research.

1. Looppanel

Looppanel is an AI-powered analysis and repository solution built by user researchers for user research. The product is built with a keen understanding that both speed and quality of insights are important.

Looppanel provides AI features for 10x faster analysis of insights, but it also allows you to:

1. Run your entire analysis to reporting workflow in one place
2. Traces all AI output back to the source so you can verify it
3. Allows you to edit and improve on AI notes, tags, etc.

The proof is in the pudding—our customers love Looppanel for how much faster and easier it makes analysis of User Research data.

Key features: 

• 95% accurate transcripts
• Automatic notes and auto-tagging of data
• Smart search that lets you find answers to any question in seconds

Pricing: Starts at $27 per month, with a free trial available.

Check out Looppanel for Free.

2. Maze

Maze enables rapid unmoderated testing of prototypes and live websites, offering a range of testing methods like usability tests, preference tests, and card sorting.

Why it stands out:‍

• Easy to set up and run unmoderated tests
• Automatically generates insight reports
• AI-powered features for probing users with follow-up questions

Which AI features do you ask?

• Dynamic Follow-Up: AI-generated follow-up questions based on user responses for more insightful feedback
• Contextual Suggestions: Real-time AI tips for improving tests, like adding relevant tasks or questions
• Bias Detection: AI detects potential biases in test design and participant selection
• Automated Analytics: AI generates customizable reports summarizing key metrics like success rates and task completion times

Pricing: Free plan available, paid plans start at $99 monthly

3. Tl;dv

Tl;dv is a meeting assistant designed to record and transcribe meetings on Google Meet and Zoom. It automatically captures video meetings, provides transcriptions, and includes features to boost productivity and extract valuable insights from meetings.

Key features for include:

• Accurate transcriptions with speaker tags in over 20 languages
• Automatic meeting recording
• Time-stamping and highlighting important moments live
• Auto-generated summary of calls

Pricing: Free plan available, paid plans start at $18 monthly

Tools and prompts are great—but realistically, how do you execute your workflow with this whole new AI in the process?

Follow these 6 steps to optimize your research process with ChatGPT:

1. Define Your Research Goals: Before you even open ChatGPT, be clear about what you want to achieve. Are you looking to generate research questions? Create user personas? Analyze user feedback? Having a clear goal will help you craft more effective prompts.
2. Craft Your Prompts: The key to getting useful responses from ChatGPT is in how you frame your prompts. Be specific and provide context. Instead of asking "How do I do user research?", try something like "What are five key questions to ask in a usability test for a food delivery app?"
3. Iterate and Refine: Don't expect perfect results on your first try. Use ChatGPT's responses as a starting point, then refine your prompts based on what you get. It's an iterative process, much like UX design itself.
4. Combine with Traditional Methods: Use ChatGPT to complement your existing research methods, not replace them. For example, you could use it to generate initial ideas for interview questions, then refine these based on your expertise and actual user interactions.
5. Validate and Verify: Always cross-check the information ChatGPT provides. It can make mistakes or provide outdated information. Use it as a springboard for ideas, but always verify important facts or methodologies.‍
6. Analyze and Synthesize: ChatGPT can help you analyze qualitative data by identifying themes or summarizing long-form responses. But remember, the final synthesis and interpretation should always be done by you, the human researcher.

Chat-GPT is new technology. Any new technology has a learning curve where we figure out what it's good, and not-so-good at.

Imagine when people first had cars—they probably figured out cars and dirt roads don't mix too well pretty quickly. Eventually we came up with traffic rules and tests to manage expectations and usage of this new technology.

Same thing with ChatGPT. We have to figure out when to use it, when not to use it, and what expectations to keep in place to use it safely.

What ChatGPT is Good At

ChatGPT is really, really good at processing text data incredibly fast. This means it can organize and analyze transcripts, open-ended survey responses, NPS verbatims, and other unstructured text very efficiently.

ChatGPT is also skilled at creating standard text content. By standard content I mean things that are well-defined in scope and have been done hundreds of times before in the world.

Writing a participant recruiting email? Sure

Re-phrasing an insight? Done

Coming up with interview questions? You got it

All of these tasks have almost standardized templates of how they can be done. That means ChatGPT has probably found them on the internet and been trained on them already.

What Chat-GPT is NOT Good At

While it's useful to know what ChatGPT is good at, it's imperative to know what it's bad at. Keep these limitations in mind any time you use ChatGPT to make sure you don't get burned by the technology:

1. ChatGPT lacks context

The current free version of Chat-GPT via GPT-4 only has knowledge about the world upto September 2021. The Premium and Enterprise version has web browsing capabilities, and is updated on the latest world affairs.

Even if you're on the premium version, ChatGPT or any AI User Research tools for that matter, don’t have access to a lot of context about your work. They don't know what came up in analytics data last week, or that your team cares about X insight in particular.

You will always need to provide this context to ChatGPT and improve its output with the insights you have.

2. Chat-GPT needs supervision

AI supplements human expertise; it doesn't replace it entirely. You cannot feed an entire project’s worth of data to Chat-GPT and expect fully formed final insights (they won't be any good).

Even with the most detailed Chat-GPT User Research prompts, you will have to spend a little more time cross-checking and validating Chat-GPT’s answers.

3. Chat-GPT can’t answer the whys

ChatGPT user research tools are good at spotting patterns in user behavior. However, it struggles to grasp the "why" behind those behaviors. That's because human actions are influenced by a colorful mix of personal preferences, cultural norms, and situational factors. 

So limit its role to just pointing out the patterns and themes in the data you provide. You will have to do the work of making sense of the whys behind the data and how the insight affects your organization. 

4. Chat-GPT’s output accuracy depends on the data you give it. 

Simply put, Chat-GPT prompts for UX Research cannot answer any question outside the syllabus.

If your data is incomplete, inaccurate, or skewed, it might not provide the answers you're hoping for, simply because it doesn't have the full context.

For precise results, add as much detail and context about your product and users to the prompt. The more, the better!

Want to try other AI-powered User Research tools? Check out our review of the 15 Best UX Research Tools in 2024.

To maximize the benefit of ChatGPT in your workflow, follow these tips:

1. Use ChatGPT as a Complementary Tool 

First and foremost, it's crucial to understand that ChatGPT is not a replacement for human expertise and creativity. While it can certainly help streamline and enhance your UX research process, it should be used as a complementary tool, not a substitute for critical thinking and analysis. 

Think of ChatGPT as your trusty sidekick – it's there to support you, but you're still the superhero of your UX research story. 

2. Provide Clear and Specific Prompts 

To get the most out of ChatGPT, you need to provide clear, specific, and well-structured prompts. The more context and detail you can give, the better the AI will be able to generate relevant and useful outputs. 

For example, instead of asking ChatGPT to "generate user interview questions," try something like:

"Create a list of 10 open-ended questions for user interviews focused on understanding the pain points and motivations of millennials using personal finance apps. The questions should cover topics such as budgeting, saving, and investing."

By being specific and providing context, you'll get more targeted and actionable results from ChatGPT. 

3. Validate and Iterate 

While ChatGPT is incredibly powerful, it's not infallible. Always take the time to review and validate the outputs generated by the AI, and be prepared to iterate and refine as needed. 

For instance, if ChatGPT generates a list of user interview questions, review them critically and ask yourself:

- Are these questions aligned with my research objectives?

- Do they cover all the key topics I want to investigate?

- Are they phrased in a way that will elicit valuable insights from participants?

If the answer to any of these questions is "no," don't be afraid to tweak, modify, or even start over. The goal is to use ChatGPT's outputs as a starting point, not an endpoint.

4. Secure Your Data

If you aren't careful, ChatGPT will be using your data to train its models. That means that those oh-so-secret insights and roadmaps your team has been working on for 6 months could be out there on the internet for the rest of the world to see.

To protect your data (as much as possible) while using AI, use one of these methods:

1. Ask OpenAI not to train their models on your data (Click 'Make a Privacy Request')
2. Remove your data from ChatGPT

Just a note: while anyone can submit these requests, it seems like OpenAI treats them differently based on the geography you submit it from. Most likely, OpenAI will respond to and deal with your request if you live in the EU or California. If you live elsewhere, we're honestly not sure if the request works 100%.

If you're concerned about security, leverage an AI tool buil for User Research instead. Tools like Looppanel have more security measures in place to secure your data and ensure it's not used for training LLMs like GPT.

5. Use ChatGPT Ethically and Responsibly 

As with any powerful tool, it's essential to use ChatGPT ethically and responsibly. This means being transparent about your use of AI, protecting participant privacy, and ensuring that any outputs generated by ChatGPT are used in a way that aligns with your organization's values and principles. 

For example, if you use ChatGPT to generate user personas, make sure to:

• Anonymize any real user data used to train the AI
• Clearly label the personas as AI-generated
• Use the personas as a starting point for discussion and ideation, not as a definitive representation of your users

By using ChatGPT ethically and responsibly, you can harness its power for good and avoid any potential pitfalls or backlash. 

6. Continuously Learn and Experiment 

Finally, remember that incorporating ChatGPT into your UX research process is an ongoing journey, not a one-time destination. As the technology evolves and new use cases emerge, it's important to continuously learn, experiment, and adapt your approach. 

Don't be afraid to try new things, push the boundaries, and see what works best for your team and your users. The more you experiment with ChatGPT, the better you'll become at leveraging its capabilities to drive UX research success. 

Incorporating ChatGPT into your UX research process can be a game-changer, but it's important to do so mindfully and strategically. By using ChatGPT as a complementary tool, providing clear and specific prompts, validating and iterating on outputs, using the technology ethically and responsibly, and continuously learning and experimenting, you can harness the power of AI to take your UX research to new heights. 

Remember, ChatGPT is not a magic wand that will solve all your UX research challenges overnight. It's a powerful tool that, when used effectively, can help you work smarter, faster, and more efficiently.

‍

Frequently Asked Questions (FAQs)

Can we use ChatGPT for UX research?

Despite the flurry it has caused in the job market, Chat-GPT isn’t actually capable of replacing human labor. It’s not smarter than a person yet, and needs a lot of hand-holding, especially for UX research tasks.

That being said, Chat-GPT prompts for UX Research can save you a ton of time and resources, if you provide enough information and the right instructions.

Need to draft survey questions? ChatGPT can help you whip up a list in no time. Struggling to come up with user personas? ChatGPT can generate ideas based on your target audience. It can even help you analyze qualitative data by identifying themes and patterns in user feedback.

How to Use ChatGPT for User Research Methods

You can use ChatGPT for most, if not all, user research methods. ChatGPT can be useful for steps across the User Research process:

1. Planning your research: Coming up with a plan, timeline, and set of questions
2. Analyzing user research: Helping you transcribe and analyze large amounts of user research data efficiently
3. Reporting: Creating report drafts and re-phrasing content to help you communicate your insights effectively

Each of these steps may look a little different across different user research methods. For example, for user interviews, you can use ChatGPT to analyze transcripts. For surveys, you may use ChatGPT to analyze survey open-ends.

Let’s say you’re using chatgpt for guerilla usability testing. In guerilla testing you’re typically going up to your target users in person and asking them to try your product or prototype. If you record the session or take notes, you can input those into ChatGPT or a GPT-powered platform like Looppanel to analyze the findings 10x faster.

For a detailed article on how to use ChatGPT for User Research Methods, check out this piece.

Using ChatGPT for User Research process

You can use ChatGPT throughout the User Research process. Typical steps to use ChatGPT for include:

1. Planning your user research
2. Creating interview questions / discussion guides
3. Analyzing data once your sessions are complete
4. As a side-kick to help you write out your report

How to use ChatGPT for guerrilla usability testing

Guerrilla usability testing is all about quick, informal feedback, and ChatGPT can be a game-changer in this process. First off, it's very useful in coming up with a test script for your guerilla usability tests. How do you approach people? How do you put them at ease?And of course, once you've captured snippets from your sessiosn (likely in different formats)—notes, quotes, transcripts of recordings—you can feed all of that into ChatGPT and ask for its help analyzing the data.

How can UX designers use ChatGPT? Can I use ChatGPT for design?

UX designers can use ChatGPT as a versatile tool to support various aspects of the design process. It can help generate ideas for user flows, provide feedback on wireframes, and suggest improvements to user interfaces. Designers can also use ChatGPT to draft UX copy, create personas, and conduct competitor analysis by summarizing key insights. While ChatGPT can assist with brainstorming and refining design concepts, it's most effective when used as a complementary tool rather than a replacement for hands-on design work.

Can ChatGPT replace UI/UX designers?

ChatGPT cannot replace UI/UX designers, as it lacks the creative intuition, deep user empathy, and nuanced understanding of design principles that human designers bring to the table. While it can assist with specific tasks like generating ideas or providing feedback, the core responsibilities of a UI/UX designer—such as crafting visually appealing interfaces, conducting user research, and iterating on designs based on user feedback—require human insight and expertise that AI cannot fully replicate.
‍

This is part of a new Looppanel blog series on AI and UX Research. Read the rest of the series for more on AI tools for UX research, or 6 Ways AI Makes Qualitative Data Analysis Faster.