This ‘self-reflection prompt’ turns superficial responses into stories
Getting useful feedback from people feels like pulling teeth.
The problem pervades market research, where survey respondents are motivated not by genuine self-reflection but cash, coupons, and redeemable "points.”
Even well-intentioned respondents give minimal effort.
We all do this, myself included.
But our tendency to provide bare-minimum attention isn't a flaw in surveys – it's an inevitability you can design around.
I want to show you an example of how you can do this based on a recent survey I ran with 152 U.S. adults.
The survey was short — just three questions — yet it shows how a simple self-reflection prompt can lead to better feedback and more meaningful insights.
The Survey Design
I first asked people to describe the most frustrating part of the first 30 minutes of their day, then to evaluate how well their initial answer captured their experience.
It looked like this:
Q1: What's the most frustrating or annoying (however mildly) part of the first 30 minutes of your day after waking up?
_______________________ .
Q2: How well do you think your previous answer captured your mornings?
☐ Superficial Glimpse: I only scratched the surface.
☐ Partial Picture: I captured some aspects but left out details.
☐ Good Overview: I provided a reasonable overview of my mornings.
☐ Thorough Depiction: I captured most of my morning experience.
The first question is a standard open-ended prompt—exactly the kind that typically yields “autopilot” responses.
The second served an important psychological function: it prompted people to self-evaluate their initial response – thus exposing themselves to how much they'd left unsaid – without me directly asking them to do more work.
It also created a natural opening for the third and final question.
Q3: So, what detail would you add to complete the picture?
_______________________ .
The Results
When I analyzed the results, I found something interesting.
Most respondents wrote significantly more in their follow-up response than their initial one – the average word count jumped from 12.6 words in Q1 to 19.2 words in Q3.
Also, people who selected "Superficial Glimpse" and “Partial Picture” wrote more, on average, than people in other groups.
But word count wasn’t the full story.
The quality of the feedback improved.
For example, one person initially wrote "cloudy vision," which could mean almost anything. But they then explained that it takes about 15 minutes for their eyes to fully focus each morning because of a medical condition called Alpha-gal Syndrome.
Another said "mustering the energy to make something to eat,” before pointing out that the annoying part is not cooking but having to "wash at least one pan or kitchen utensil” before they can start making food.
Not every response was valuable.
Some people just reiterated their first answer.
But many shared deeper, more nuanced experiences after the self-reflection prompt.
Check out the table below to read a few more examples.
A good survey should never ask for deeper answers outright (that never works), but create a psychological pathway that makes self-reflection feel natural.
This is the "inevitability you can design around."
When you stop fighting human nature and start designing for it, better feedback isn't just possible—it's inevitable.
Need help doing that?
Consider booking a Survey Roast.
Send me your survey draft, and for $145, I'll make a 15-minute Loom video with copy-and-paste edits and suggestions to improve your survey data quality.
Steal-This Resources
• Data visualization one (download and edit with Datawrapper account)
• Data visualization two (download and edit with Datawrapper account)
• The survey data
Cheers,
Sam