In a recent session, I shared how AI can help researchers move beyond efficiency gains and instead unlock richer, more systematic, and more creative qualitative analysis. The idea is simple:
AI doesn't replace research expertise.
It amplifies it.
And when used well, it helps us become what I call 10x researchers.
Watch the full session here:
AI is often described as a speed tool.
AI is often described as a speed tool.
It summarizes faster.
It transcribes instantly.
It automates workflows.
It transcribes instantly.
It automates workflows.
But speed isn't the real breakthrough. The real breakthrough is depth.
From Faster Research to 10x Research
Today's large language models (like ChatGPT, Claude, Gemini, Grok) are trained on vast amounts of conceptual knowledgeβincluding research methodologies and qualitative frameworks.
That changes the game.
Instead of spending time explaining a framework from scratch or struggling to apply it correctly, we can now:
Instead of spending time explaining a framework from scratch or struggling to apply it correctly, we can now:
- Ask which framework best fits our research goal
- Learn how to apply it in context
- Generate structured prompts
- Test multiple analytical lenses on the same dataset
- Iterate rapidly
This allows us to focus on interpretation and judgment β the parts only humans can do well.
Why Frameworks Matter in Qualitative Analysis
Over the past 30+ years in research and design, one thing has consistently separated strong analysis from surface-level summaries:
Frameworks.
Frameworks direct attention.
They help us decide:
- What should move into the foreground?
- What should fade into the background?
- What patterns are worth digging into?
Think of qualitative analysis like using a metal detector on a beach.
You scan.
You hear a signal.
You dig.
You step back.
You scan again.
Frameworks are the settings on that metal detector. They determine what kind of "signal" you're looking for.
Some examples:
Some examples:
- Jobs To Be Done β Focus on goals and underlying jobs
- Journey Mapping β Focus on time, stages, emotional arcs
- Laddering β Move from behaviors to motivations to values
- Thematic Analysis β Identify recurring patterns
-
Personas β Cluster patterns into human archetypes
Each lens reveals something different.
And AI makes it possible to apply multiple lenses quickly.
Same Data. Different Depth.
Let's take a simple example:
A participant says they "order takeout after work when I feel exhausted. I know it's expensive, but it makes life easier."
Without a Framework:
We might summarize:
- Convenience matters
- Fatigue influences decisions
- Price sensitivity exists but is justified
With a Journey Map Lens:
We uncover:
- Trigger: End-of-day exhaustion
- Overwhelm: Too many food options
- Relief: Order placed
- Guilt: Spending more than intended
- Emotional arc: Fatigue β Relief β Mild guilt
With Jobs To Be Done:
- Functional Job: Get dinner with minimal effort
- Emotional Job: Feel taken care of
- Deeper Value: Restoration, recovery, identity reinforcement
With Laddering:
- Attribute: Ordering food
- Consequence: Saves effort
- Higher-Level Why: Protecting work performance
- Core Value: Competence, self-image, achievement
Same quote, different levels of depth.
Now imagine being able to test all of these approaches in minutes.
The Real Constraints in Traditional Analysis
Why don't we normally apply 4β5 frameworks to every dataset?
Because of:
- Skill limitations (some frameworks require deep expertise)
- Time constraints
- Budget pressures
Applying multiple methods rigorously is expensive. AI dramatically lowers that barrier.
Not because it replaces expertise β but because it understands the structures of these frameworks well enough to generate structured starting points.
You can now:
- Ask AI which sentiment frameworks to combine
- Generate a Personal Construct Psychology prompt
- Apply Means-End Chain analysis
- Create behavioral personas
- Generate goal-based personas
- Build journey maps
- Combine thematic + sentiment + persona modeling
And then iterate.
This isn't automation.
It's augmentation.
It's augmentation.
Diversity of Thought on Demand
One of the most powerful shifts is this:
You're no longer analyzing alone.
When you prompt AI using different frameworks, it's almost like having multiple colleagues:
- One thinking in JTBD terms
- One thinking in laddering
- One thinking in personas
- One thinking in journeys
You can challenge your own bias.
Test alternative interpretations.
Interrogate your data more rigorously.
That diversity of thought leads to stronger insight quality.
Depth Through Natural Inquiry
Perhaps the biggest change is conversational analysis.
Instead of rigid analysis pipelines, you can now:
- Follow your natural chain of thought
- Validate hypotheses in real time
- Zoom in on one ladder
- Expand a persona
- Generate a journey map for just one archetype
- Re-cluster themes based on emotional tone
The analysis becomes iterative, fluid, and intellectually engaging.
That's where depth happens.
That's where depth happens.
Important: AI Is Not the Final Answer
AI-generated output is a starting point. Human judgment remains central:
- Is this interpretation valid?
- Does it align with the raw data?
- Is it overreaching?
- What nuance is missing?
AI helps you go deeper β but you decide where to dig.
Want to Try the Prompts?
All the prompts demonstrated in the session are available here:
You can use them in: ChatGPT, Claude, Gemini and ofcourse DoReveal
No special tools required.
Final Thought
The question isn't: "Can AI do research?"
The better question is: "How much deeper can we go when AI becomes our thinking partner?"
That's where the real transformation lies.