How to Write a Discussion Guide That Doesn’t Suck (And Actually Gets You Answers That Matter)
How to Write a Discussion Guide That Doesn’t Suck (And Actually Gets You Answers That Matter)
Let’s talk about writing a discussion guide. You know, that thing you think is just a list of interview questions… but if you do it right, it becomes one of the most powerful tools in your research arsenal.
Over the years, we’ve built hundreds—maybe thousands—of these things. And if there’s one thing we’ve learned, it’s this: The best insights don’t come from what you ask. They come from how you ask—and how much space you give someone to tell you the unexpected.
Get the Tools
To help you get started, we’ve included reference guides with embedded comments that explain why each question is structured the way it is. These are based on a fictional company building a conversion kit for towable RVs and include examples for both:
You can use these as a foundation to write your own or if you’re AI curious, Try using Generative AI with these starter prompts.
Skip the Persona—Focus on the Problems
Let’s clear something up: Personas are overrated and dangerously easy to get wrong.
Most personas are built on assumptions—age, job title, income bracket, even hobbies—and then we wrap messaging and product decisions around this fictional “Sarah” who drinks oat milk lattes and lives in a condo.
The problem? Sarah might not exist. Or worse—she might exist, but not care at all about what you’re building.
Instead of demographic fiction, build around real goals, pain points, habits, and the words people actually use. Ask yourself:
- What are they trying to accomplish?
- What gets in their way?
- What do they love or dread?
- What language do they use to describe all of that?
Your discussion guide isn’t for confirming your persona. It’s for discovering real patterns—and often, being surprised.
The Guide’s Purpose
This isn’t a survey. It’s not a quiz. It’s a conversation. You’re not selling. You’re not pitching. You’re learning.
The best guides:
- Avoid bias and assumptions
- Start broad to uncover aspirations and pain points
- Narrow in only once a real need is voiced
- Verify hypotheticals with real-world stories
- Adapt in real time based on the conversation
- Create space for users to feel heard, safe, and engaged
Use the guide to navigate, but let people lead you toward what matters most to them. That’s where the insights hide.
Important: Do not talk about your product, your solution, or even hint at your concept—unless the person you’re speaking with brings it up first. Even then, keep your response high-level, factual, and brief. Acknowledge the question honestly, but then steer the conversation back to their experiences and needs. The more space you give them, the more truth you’ll get.
How to Get Started
Before you start writing questions, do this:
📝 Start by writing down at least 10 pain points and 5 aspirations your ideal customer might have. These can be directly related to the problem you’re solving—or indirectly tied to broader issues in their life or work.
For example, if you’re building a grocery delivery service, a direct pain point might be: “I hate grocery shopping.” An indirect one could be: “I wish I had more time in my day.” Both are valid. Both are useful.
Write each one in the first person, and try to include some emotion. You want to understand not just what someone struggles with—but how it makes them feel.
Next, use that list to organize your assumptions:
- Which ones feel critical?
- Which ones seem like edge cases?
- Which ones do you need to test first?
👉 Then build your discussion guide so that people could give you those answers—but without being led. Your job is to create openings for those pain points and desires to emerge organically in the conversation.
If someone naturally mentions one of your assumptions, you’ve validated it.
If they don’t, it might not be as important as you thought.
Structure your guide from broad to narrow:
- Start with open-ended questions about their life or work context
- Move into what they love and hate about it
- Then slowly funnel toward the domain or problem space you care about
Break complex ideas into smaller, testable prompts. For example, instead of:
“Would you want an electric trailer conversion kit?”
Try:
“What’s the hardest part about towing?”
“Have you considered other ways to make that easier?”
“What would your ideal solution look like?”
Each of these gives you a path to meaningful insight—without planting the idea first.
Structure Matters: Start Broad, Then Narrow
There’s a reliable flow to great interviews, which you’ll see echoed across both our example guides:
- Start broad
“Tell me about your setup.”
“Why did you start RVing?”
→ Get them talking about themselves and their context. - Explore love/hate
“What do you love about this?”
“What’s the worst part?”
→ This reveals motivation and pain. - Dig into real stories
“When did that last happen?”
“What did you do next?”
→ Real events > hypotheticals. - Refine and relate to your domain
“Have you considered buying an EV tow vehicle?”
“What concerns would you have?”
→ Only bring this up after they’ve had space to share.
This structure helps you avoid leading questions while still arriving at the answers you need.
Love/Hate Questions: Your Secret Weapons
Love questions reveal:
- Aspirations
- Language
- Emotional drivers
Hate questions reveal:
- Top-of-mind pain points
- Their problem-solving process
- Triggers that create switching behavior
Ask:
“What’s the worst part of [topic]?”
“When was the last time that happened?”
“What did you do next?”
“Did anything help? Would you do it differently?”
These real examples are rich with emotional texture—and often give you the copywriting language you’ll use later in your landing pages.
Check Your Guide Against Customer Needs
Once you’ve drafted your guide, compare it to your Ideal Customer Profile (again, not a persona).
Ask:
- Are you giving them the space to bring up the things you think are important?
- Can they talk about their pain points or aspirations without being led?
- If they don’t mention it, would you know that it’s not top of mind?
If someone goes through your entire interview and never brings up your key assumption? That’s data. That’s value. That’s why this process exists.
It’s a Guide, Not a Script
Don’t read it like a script. Seriously. You’ll lose people fast.
Instead, think of it as a map. If the conversation flows naturally toward something relevant—go there. If they say something surprising—dig in.
Tangents are where the gold is.
If your first interview changes everything you thought you knew? Amazing. Adjust the guide and try again.
Create Evangelists
The last few questions in your interview shouldn’t just extract insights—they should build relationships.
Try asking:
- “What’s one change you’d love to see in the next 5 years?”
- “Is there anything I should have asked that I didn’t?”
- “Would you be open to reconnecting when we’ve built something based on your feedback?”
If people feel like they had a hand in shaping your product, they’re far more likely to champion it later.
TL;DR Best Practices
- Don’t sell. Don’t pitch. You’re learning.
- Start broad, then narrow.
- Ask “love/hate” questions to get both dreams and frustrations.
- Verify hypotheticals with real stories.
- Don’t bias the conversation.
- Leave space for unexpected insights.
- It’s iterative—refine as you go.
- Always thank them and offer a follow-up loop.
If you follow this process, you’re not just validating an idea—you’re discovering something people actually want. You’ll walk away with words you can steal for your marketing, insights you never saw coming, and a handful of people who are now invested in your success.
And if you are wrong? Better to know now than after the launch.
