Oh, what’s that? GPT-4? Yeah, GPT-3 is just SOOOOOO January 2023.
AI is getting smarter, and while we are still patiently waiting for the robot overlords to take over society, do we think it is at a place where it could completely replace the UX researcher?
We mean, it’s not exactly easy working with some of these Product Managers…
@tldv.io We love constructive criticism #productmanagement #product #tech #productmanager #product #userfeedback #corporatehumor #startup
♬ original sound - tldv.io - AI Meeting Recorder
While we’re not trying to do anybody out of a job here, with this latest innovation, is it time to start phasing out the role itself and instead making it into an “AI prompt writer for UX research”?
It’s no secret that UX research is vital to any organization creating products. It helps identify user behavior, pinpoint problems and solutions, and most importantly: it grounds us with data-based insights when it comes to our product design decisions.
However, GPT-3, and now GPT-4, have been making waves for their ability to generate seemingly coherent writing using natural language processing (NLP). Could this be something that will replace the need for a UX researcher altogether?
There’s only one way to find out. Battle Royale time!
What makes up UX Research?
The UX researcher has traditionally influenced many key areas regarding Products. These are:
The Rounds
We’ll break each section down, compare the pros cons, and declare a winner.
At the end of the battle, we will tally the scores, count the teeth lying on the battlefield and choose a winner.
Will the UX researcher need to start looking for a new job? (We hear that AI is a particularly up-and-coming industry, lol.) Or is AI only ideal for geeking out in the lab and not quite ready for prime time?
Oh, AND, we’ll even throw in some free ChatGPT Prompts for Product Managers to each section to show how it can be done. Yep, that’s right. No thinking required. Just some ChatGPT Prompts for UX Researchers at your fingertips.
Let’s dive in and see who makes it out alive.
Define The Persona
Personas are your people. They are your customers created in a living, breathing, albeit 2D, profile. They dictate the way you design, market, and even product management.
😍 UX Researcher Pros: UX Researchers are trained to craft a story and persona out of data given to them. This allows for an accurate representation of a person based on facts. Equally, in real-life, UX researchers can bring deep expertise to the table when defining a persona by using their experience with user research methods such as surveys and interviews. This means that by bringing together all types of data and insight, the built personas are rich and filled with real-world experience.
😵 UX Researcher Cons: UX researchers may not always be able to connect the dots accurately. They can be jaded by their experiences, distractions, and lived experiences. While being objective is entirely impossible for both, it’s much more likely that a human UX researcher will be more prone to bias. Also, collecting the data for these detailed personas will likely take over two weeks to build a true picture.
😍 GPT-4 Pros: GPT-4 can utilize its AI algorithms to help generate a consistent, accurate persona based on what it has learned from the data given. This allows the computer to craft a persona that can be as close to reality as possible without requiring much manual labor or time input. The GPT-4 engine could probably build a pretty decent persona in under 10 minutes.
😵 GPT-4 Cons: GPT-4 cannot yet read between the lines and understand what might be missing from the input data. It can’t make any creative decisions on its own, either. Additionally, it can only generate based on what it knows from pre-existing training and data input. This means the information given in the first place can be limited. There are also issues with the quality of a computer-generated persona and how realistic it could be. Finally, all AI is typically external to a company. If security threats exist on the external AI database, sensitive company information could be lifted. Eeeek!
🏆 The Winner: Speed is great, but security and accuracy are eye-wateringly more important. Less speed, more security! The human wins.
🤖 BONUS AI Prompt: Create a UX persona for me based on {INDUSTRY}
Competitor Research
There’s nothing like a good old-fashioned competitor analysis to get the UX machine rolling. A competitive analysis allows companies to understand what other people are doing in the market and how they can differentiate themselves or improve on existing solutions.
😍 UX Researcher Pros: The UX researcher is well practiced in looking through the industry and creating detailed insights into market trends, what the competition is offering, and even how they’re marketing their product. This is invaluable information that can be used to inform product design decisions.
UX Researcher Cons: The UX researcher may miss out on some nuances or have trouble understanding a complex market or product. They also may struggle to keep up in a fast-changing industry, with some of their insights becoming stale quickly without their knowledge.
😍 GPT-4 Pros: GPT-4 can be fed large amounts of data to make sense of the market landscape and quickly generate insights into what competitors are doing, their pricing models, and even predictions for future trends. This allows companies to react quickly and stay ahead of the competition.
GPT-4 Cons: GPT-4 is still limited by its access to training data and may miss out on important information that would otherwise be uncovered through human research. It also can’t make creative decisions or think for itself and needs to be closely monitored to ensure the insights’ accuracy.
The Winner: It’s a draw! The great thing about this battle is that both have their use cases and strengths. Depending on the research required, either can be a valuable asset to the team. They often work hand in hand when gathering insights and understanding a market more thoroughly. So while AI may be winning in some aspects, humans remain supreme regarding the “feel” of a competitor’s brand, product, and positioning.
🤖 BONUS AI Prompts:
Who are the top competitors in the industry?
What are the strengths and weaknesses of their websites and apps?
What is their social media presence like?
What are your competitors’ SWOT (Strengths, Weaknesses, Opportunities, Threats)?
What are common pain points among their customers?
How do their UX offerings compare to industry standards?
Are there any emerging UX design and technology trends that your competitors are adapting to?
Conducting User Interviews
User interviews are a great way to gain feedback and insights directly from the users. It can be very illuminating to have an open dialogue with them and ask questions that would otherwise be difficult to answer through analytics or research alone.
UX Researcher Pros: A UX researcher has the benefit of being able to build rapport with participants, ask follow-up questions, and have meaningful conversations that don’t feel overly robotic. They can also interpret nonverbal cues and body language, which helps get honest answers and understand people’s needs more clearly.
UX Researcher Cons: Depending on the user’s location, it can be time-consuming for a researcher to travel or arrange participants for remote interviews. Interviews require active engagement and can be difficult to set up without the right resources.
GPT-4 Pros: GPT-4 can quickly generate questions for user interviews, saving time on research and allowing companies to get user feedback faster than ever before. GPT-4 is also beginning to understand user sentiment (Note, this is different to nuance and non-verbal communication) through NLP, which helps better understand customer needs.
GPT-4 Cons: GPT-4 is limited by its access to training data and may miss out on important insights that would otherwise be uncovered through human conversation. It also cannot, at the time of writing, interpret nonverbal cues or detect subtle nuances in conversations, making it less reliable when conducting user interviews. Essentially, AI really can’t read between the lines. While it can be great for picking up customer feedback, it doesn’t ask WHY often enough.
The Winner: Have you ever tried to speak to a chatbot and then had to have a conversation with a customer service representative? Sometimes, having conversations with a human is easier and more efficient. The same applies here – user interviews are best conducted by humans. GPT-4 can be used to formulate questions, but it is much better to rely on skilled UX researchers for the actual interviews. Ultimately, human interviewers tend to produce better quality results than AI-powered chatbots or robots.
However, one way of using both in harmony is by utilizing tl;dv’s magic AI Note Summarizer. With a single click of a button, it can help to summarize a specific moment. This allows UX researchers to focus on the user rather than taking notes, ensuring that the important moments are captured!
ALSO! tl;dv prevents the dreaded “UX Researcher Shadow Effect“. You know the one… That thing where two UX researchers sit on the same call, but one just sits there silently taking notes. To be honest, it’s kind of creepy and with tl;dv transcribing, tagging, and logging it all, you can ditch that third wheel. Result.
@tldv.io At least they haven’t noticed the dandruff. Magic Search available now for FREE #tldv #chatgpt #ai #tech #startup #onlinemeeting #meetings
♬ original sound - tldv.io - AI Meeting Recorder
🤖 BONUS AI Prompts:
What are the biggest challenges that users face?
How would they like to see these challenges addressed?
What features do they find most valuable in the product or service?
What impact has this product/service had on their lives?
What other products or services do they use and why?
Are there any areas where they feel the product/service could be improved?
Are there certain features or solutions that your competitors offer that you don’t yet provide but should consider adding?
Analyzing other data sources
Another way to gain insights into customer needs is by examining other data sources such as surveys, polls, and customer feedback. This can help understand user sentiment and uncover areas for improvement or emerging trends.
😍 UX Researcher Pros: UX researchers are well-equipped to analyze survey responses and qualitative data from customer feedback. They can develop meaningful questions that address the topics that matter most based on the project context. Additionally, a UX researcher has the skills to interpret structured and unstructured data, making identifying patterns or correlations between variables easier.
😵 UX Researcher Cons: Depending on the size of the project, a UX researcher may not have enough time to mine all available data sources in a timely fashion. Additionally, manual analysis of data can be tedious and labor-intensive.
😍 GPT-4 Pros: AI can quickly process large amounts of data and analyze the results in real time. It can also identify patterns or correlations between variables that may not be immediately obvious to humans. AI eliminates the need for manual analysis and enables companies to make better decisions based on customer feedback or survey responses.
😵 GPT-4 Cons: AI is limited by its access to its database which may bias the results either positively or negatively. Additionally, at the moment, AI cannot interpret nonverbal cues such as tone or body language, which are important factors when interpreting customer feedback.
🏆 The Winner: In this case, UX researchers and AI have unique advantages and disadvantages. While UX researchers are better equipped for manual analysis, AI can help with data mining and quickly analyzing large datasets. Ultimately, it would be best to use a combination of both to gain the most accurate insights from customer feedback and surveys. Again, it’s another draw!
🤖 BONUS AI Prompts:
What types of customer feedback do customers provide?
Are there any common themes or trends in customer responses?
Which features are users most satisfied with?
What areas need improvement, according to user feedback?
How does user sentiment vary between different demographics or regions?
Are there any unexpected correlations between variables that may be useful for further exploration?
Make a UX Plan
Once the insights have been gathered, it’s time to create a UX plan. UX plans should consider user needs, behaviors, and preferences and how to best utilize technology solutions to achieve them.
😍 UX Researcher Pros: A UX researcher can bring an invaluable human touch when developing a UX plan. They are better equipped to understand customer perspectives and develop creative solutions that consider the nuances of user behavior. Furthermore, they are more likely to be able to spot potential issues before they happen since they understand human behavior on a deeper level than AI-powered systems.
😵 UX Researcher Cons: Manual data analysis can be time-consuming and tedious, limiting their ability to keep pace with the rapid advancements in technology. Additionally, UX researchers may not have the technical skills necessary to develop a plan that takes advantage of all available technologies. It can also take a couple of weeks to create a plan.
😍 GPT-4 Pros: AI can quickly process data and identify patterns or correlations between variables which can be useful for developing an effective UX plan. It also eliminates the need for manual analysis and enables companies to make decisions based on accurate insights. It’s so quick. In fact, it can devise a plan in literally a minute. Yes, A MINUTE!
😵 GPT-4 Cons: You know the benefit of a plan in a single minute? Yeah, it’s one of those scenarios where you get what you put in. AI cannot fully read or replicate all the factors needed when creating a user-centric experience. In fact, it cannot interpret anything that will give you anything of value. There is a real risk here of leaning totally on an AI, putting confidence and trust in it, and literally spitting out a “wrong” plan.
🏆 The Winner: I mean, it’s pretty obvious, but DON’T TRUST THE MACHINE! UX Researcher all the way, baby.
Do you really want to use dud prompts?
Advise MVP Prioritization
Once the UX plan has been created, it’s time to prioritize features and create a Minimum Viable Product (MVP). MVPs are the most basic version of a product or service that can be released to gather user feedback. This feedback can then be used to improve the product or service before releasing a full version.
😍 UX Researcher Pros: A skilled UX researcher is well-equipped to prioritize user needs and develop an MVP based on those needs. They understand how users interact with products, services, and websites, which makes them better equipped for this task than AI-powered systems. Furthermore, they are more likely to spot potential issues or areas of improvement before launching the MVP.
😵 UX Researcher Cons: UX researchers may not know the latest technologies or tools to develop MVPs. Furthermore, they may find developing innovative solutions based on user feedback difficult if they experience any kind of human “block”.
😍 GPT-4 Pros: AI-powered systems can quickly process and analyze large datasets to identify areas that need improvement and prioritize features for an MVP. It also eliminates the need for manual analysis, saving time and resources.
😵 GPT-4 Cons: AI lacks the human factor when interpreting data which means it cannot understand user feelings or preferences – something a UX researcher could do. Additionally, AI-powered systems are only as good as their inputs, meaning that any inaccurate data will lead to unreliable results.
🏆 The Winner: Again, this decision is obvious. UX Researcher wins in this category too! They are the experts in understanding user feedback and know what design elements need to be improved or changed based on that feedback. AI-powered systems may be able to crunch numbers quickly, but they lack the human touch that is so important when it comes to creating a successful MVP.
🤖 BONUS AI Prompts:
What are some key metrics that can be used to evaluate the effectiveness of MVP features?
Explain the process for prioritizing MVP features based on user feedback?
How can UX researchers work with cross-functional teams to ensure that user experience is prioritized in MVP development?
Can you provide some best practices for incorporating user feedback into MVP prioritization?
What trade offs can be involved in prioritizing features for an MVP, and how to make informed decisions when balancing user needs, technical feasibility, and business goals?
🏆😍 The Ultimate Winner?
It’s a real-life UX researcher, of course!
No matter how advanced AI becomes, there will always be an element of uncertainty when making decisions involving complex human behavior and preferences.
A skilled UX researcher can provide invaluable insight into user needs and preferences, making them the more efficient choice for developing an effective UX plan and prioritizing features for an MVP.
While AI has its advantages when analyzing data, it still cannot replace the human brain and its ability to interpret data more effectively.
So, if you want an effective UX plan or a successful MVP, stick to using a UX Researcher!
They are the experts in user feedback, and they can make sure your product or service is developed with the user in mind.
However, even the best UX researcher can benefit from GPT-based helper, and tl;dv uses GPT-3 to help transcribe, take notes and produce highly accurate content. So, go on and give it a try!
