by Why qualitative UX research is one of the best use cases for AI
A white paper from Knit
There are probably hundreds of use cases for AI. But in their current form, LLMs are not optimized for a lot of the tasks they’re being prompted to do. If you’ve asked ChatGPT to write a blog post from scratch or propose a novel solution to a problem, you understand the limitations of the technology today.
But there are also tasks that LLMs are vastly better suited for than almost any human. These tasks mostly involve reorganizing, summarizing, and surfacing information.
It makes sense when you boil the marketing noise surrounding LLMs down to the technology’s essential definition: they are probability or prediction machines trained on massive datasets.
They’re great at generating a distillation of the information they’ve been trained on in response to a prompt. They are not capable of truly original thought (or outputs).
There are a ton of varied and nuanced tasks that ladder up to finally delivering a user research report, wireframe, or prototype to a stakeholder’s inbox.
Somewhere in that mix, teams are spending an inordinate amount of time sifting through huge datasets looking for patterns, answering questions, surfacing pain points, and otherwise rearranging the information they’ve gathered.
Combing through data is not the most interesting part of a UX researcher’s job. That’s where AI can help.
Where AI drives real efficiencies in the UX research workflow
There are certain tasks that AI can significantly expedite, requiring only last-mile human intervention for quality assurance. But there are other processes it can essentially complete autonomously end-to-end.
Data collection and processing
Instead of spending hours hand-coding and processing data, UX researchers can confidently pass data collection and processing off to AI.
LLMs trained on the proper datasets and layered with the right automated tooling can extract and code information from structured documents (like survey questionnaires).
They can also use natural language processing (NLP) to extract information and insights from unstructured data, including videos, text blurbs, customer reviews, and emails.
Data analysis (including sentiment analysis)
AI is also optimized for pattern recognition. It’s capable of skimming through large datasets and surfacing patterns and trends across survey questions and respondents.
For researchers conducting qualitative research at scale, identifying general sentiment across hundreds or thousands of responses can be a rote, exhausting exercise.
AI can offload this task without minimal to zero need for human intervention.
Survey and questionnaire generation
Staring at a blank page is paralyzing. Use AI to draw up a first draft.
Submit your research objectives, use case, and audience criteria to an LLM and prompt it to return a questionnaire.
Quick caveat: while this draft will be solid, it likely won’t be perfect.
A research platform that leverages AI (like Knit) will produce a questionnaire that uses both your inputted research objectives, use case, and audience criteria – as well as your past research – to produce an editable questionnaire in minutes.
This allows your team to make tweaks in-app.
Report creation
An LLM chatbot like ChatGPT isn’t capable of drawing up topline reports complete with data visualizations.
However, a research platform powered by AI can sift through thousands of data points, identify patterns in qualitative data, and conduct sentiment analysis to generate a slide deck of key findings.
Putting PowerPoint slides together is tedious and time consuming. Offloading the tasks of generating data visualizations, surfacing key takeaways, and bulleting out recommendations to AI gives you more time to focus on the meatier aspects of your function.
Case study: How NASCAR integrated AI into qual research
As part of its strategic initiatives for 2025, NASCAR wanted to introduce an in-season tournament. It was a significant departure from their traditional scheduling.
But before moving on this big bet, NASCAR needed to understand the level of interest and engagement potential among fans and non-fans.
The team partnered with an AI-native research platform – Knit – to gather crucial qualitative and quantitative insights at scale.
The goal was to gauge interest in the in-season tournament, identify key elements that resonate most with respondents, and understand the potential barriers to engagement.
NASCAR also wanted to explore the impact of the tournament on the overall fandom, including interest in attending and watching future NASCAR events in person and online.
Methodology
Each respondent was asked to:
- Provide significant layers of qualification and segmentation, including NASCAR fandom, broadcast/streaming behaviors, and more
- Identify familiarity and perceptions surrounding other major sports tournaments and their impact on the fandom
- Identify awareness, perceptions, and interest surrounding NASCAR’s upcoming in-season tournament and its potential impact on the fandom
With Knit, NASCAR surveyed 950 respondents and had a topline report in hand within 5 days.
The AI-generated report included key takeaways, trends, and insights from thousands of qualitative responses.
NASCAR discovered:
- Which segments had the highest interest in NASCAR’s in-season tournament
- Respondents’ reasoning for and against an in-season tournament
- Insights that informed unique strategies across numerous segmentation criteria including fandom level, age, geographical region and more
- Qualitative suggestions for driving additional interest in NASCAR’s in-season tournament
Without AI, this study would’ve taken weeks to months of research time – and accrued all the costs that come along with such an investment.
Instead, one week after launching their research study, NASCAR’s insights team was able to focus on strategizing the best way to design and market its in-season tournament.
Qualitative research. Fast and at scale.
Use Knit’s AI-native research platform to get deeper, more actionable qualitative insights in days. Our platform handles everything from survey creation and sampling and fielding to analysis and reporting.
It’s the simplest and most efficient platform for getting accurate insights at scale.
Chat with our team to find out why companies from Amazon and Adobe to startups of 10 or less people trust Knit to run research studies of all scopes and sizes.
🚨 Final Call! QUAL360 NA is happening this week!
Join us in Washington D.C. on March 12-13, 2025 to connect with leading industry professionals and explore how qualitative research is shaping the future. Don’t miss out on networking opportunities with leading brands like Colgate-Palmolive, Reddit, Meta, Amazon, TikTok, Samsung, Nike and more.
Secure your spot today! Check ticket availability here.
