uxGPT: Write user research questions with prompts and uxGPT Research Questions
Patrick Neeman
UX Leader at Workday. Author of uxGPT: Mastering AI Assistants. Ex-Microsoft. Working on great things with Open AI since late 2022. Data influenced since 1995.
How many of us have gone into discovery interviews unprepared for what we were going to ask, or have other participants asked something totally outside the lines??
Yes. All of us.
Now we can start from a better foundation with draft questions generated by ChatGPT.
Writing user research questions is one of the most effective ways I have used the application as a starter method. You can save time by generating well-crafted questions by using OpenAI’s vast amount of training data. Participants can avoid common pitfalls and most biases, resulting in more information to ask better questions.
When you use OpenAI, there are always going to be some biases because the trained models are always biased because it involves human content. That’s correctable with a human in the loop reviewing the draft.
Most of the teams I have managed shared research duties with product managers. This was always something that was time-consuming and lacked alignment. Now, not so much.
These are tips for writing user research question prompts. Most of this is focused on B2B, but you can play with B2C if you change the focus.
Start wide
Begin by declaring the category you want to research or the areas you want to compare across competitors. I recommend starting with a specific category.
It’s a good idea to set the context of your research so that you can refine it with deeper analysis and track your journey. To show the answer, you can specify any context, including the number of results or the display form.
We’ll start with our example, customer relationship management.
Prompt
Create 10 user research questions about customer relationship management systems.
Refine
Focusing on specific features allows you to dig deeper into the user's actual experiences and pain points. This will uncover detailed insights about what works well and what doesn't, which helps with prioritization and pattern matching on specific needs.
This will also help you gather precise feedback rather than broad, generalized opinions by targeting the questions.
Asking very general questions for larger systems is too broad, especially if you have a system in place. Let’s focus on one goal for the feature set — managing multiple accounts. This is a typical use case for customer relationship management.
Prompt
Create 10 user research questions about managing multiple accounts using customer relationship management systems.
Add protopersonas
Now let’s go to another step — adding a protopersona.
Protopersonas ensure that your questions are targeted and relevant, uncovering insights that might be overlooked with a generic approach of looking at all users in the system.
By tailoring questions to specific personas, you can gather more actionable feedback during user interviews because you’re suggesting a specific role when using the application, even in draft state.
For the record, I’m calling them protopersonas intentionally because they don’t take the place of talking to users, just formulating questions you may want to ask so you can refine. You can upload a persona from our list of users that we focus on. I won’t cover this during this tutorial, but it is something that you can play with at this point to inform your questions.
We’ll add a business development representative for this example.
领英推荐
Prompt
Create 10 user research questions about managing multiple accounts as a business development representative using customer relationship management systems.
Refine more
Focusing on a specific protopersona with a specific goal is even better because it ensures insights gathered are deeply relevant to specific tasks the user may need to complete.nbsp;
The more specific you get about the request, the more it helps narrow down the request. Honing in can uncover nuanced and very specific pain points that might otherwise be overlooked.
However sometimes you don’t want to get too specific because you want to explore. I like keeping it open myself because then it might catch something I missed, and it’s a great way to learn how to craft better prompts.
We’ll add a business development representative managing multiple accounts.
Prompt
Create 10 user research questions about managing multiple accounts as a business development representative focusing on generating more qualified leads using customer relationship management systems.
Test the questions
This doesn’t replace users, but it gives a bit of a sniff test to see if the questions make sense. This is crucial because it helps ensure clarity for participants. ?
This can reveal potential ambiguities, leading to improved question design. Additionally, if you edit the draft questions, you also can paste them back in and ask for the answers using a different prompt.
Sample Questions Prompt
Create 10 user research questions about managing multiple accounts as a business development representative focusing on generating more qualified leads using customer relationship management systems. Give three detailed examples of answers to each question.
Sample Answers Prompt
Create three sample answers about managing multiple accounts as a business development representative focusing on generating more qualified leads using customer relationship management systems from the entered content.
Try out this Custom GPT — uxGPT Research Questions
Don’t want to do the work yourself? Not a problem. I’ve done a lot of the legwork for you.
Try this custom GPT at uxGPT Research Questions.
Articles in the uxGPT series
About Patrick Neeman
Patrick Neeman works at Evisort as the Head of User Experience and Research. He has lead user experience teams at many data companies, like Knowable, Icertis, Apptio, and Jobvite and nPario. You can find his blog Usability Counts here, and he also writes on Medium.
He has taught at General Assembly as a user experience instructor.
All his opinions are his alone.
Assistant Manager - User Research | Product Development | Upstox | S&P Global
2 个月Patrick Neeman Fantastic approach! Testing and refining questions with team feedback can make your user research even more effective and relevant.
user experience designer | design educator & mentor | professional coach | AI UX design founder
2 个月These are the Connected tools you can use (for free) from my agent ask/wAI Pro - the researcher...
user experience designer | design educator & mentor | professional coach | AI UX design founder
2 个月user experience designer | design educator & mentor | professional coach | AI UX design founder
2 个月I've added upload PDF analysys to anticipate treating your Personas into your questions process...
user experience designer | design educator & mentor | professional coach | AI UX design founder
2 个月Nice! I've tried my own AI mentoring support agent last year and it seemed very apt to help to define and refine research questions. It is free from the openAI PLUS subscription thingy.... Called it ask/wAI Assistant Pro Agent - the researcher. Check it out on my website and tell me what you think - always very curious if it helps you. at https://kinokast.org/pro-mentor-work-environment