Feeling or Feature? Auditing Your Own Headlines for System 1
Objective: Given 3 real ad or landing-page headlines, classify each as feeling-first or feature-first, rewrite the feature-first ones, then collect 3 real 'why did you buy' answers to check what customers say versus what likely moved them first.
You're a growth marketer at Duolingo checking whether the app's own paid-search headlines lead with an emotional payoff (confidence, relief from guilt) or a feature (streaks, XP, lesson count) before a new batch of ad variants goes live.
Classify 3 real headlines as feeling-first or feature-first, rewrite the feature-first ones, then collect real customer reasoning to see what they say after the fact.
Before you start
What you'll need
Free path (everything below is enough to finish)
Free, fast for a small comparison table
Free, no account needed for respondents
The process
2 steps
Step 01 of 02
The lesson's rule: 'Sleep through the night again' beats '10-hour battery.' Lead the headline with the feeling the product produces, save the spec for the third paragraph where System 2 goes looking for ammunition.
Pull 3 real headlines (Duolingo's own paid-search ad, plus 2 competitor language-app ads). For each, is the first clause a feeling or a feature? Rewrite any feature-first headline.
Procedure
- Copy 3 real headlines in one category
- Underline the first clause of each and classify it as naming a feeling (relief, pride, confidence) or a feature (streaks, price, spec)
- For any feature-first headline, rewrite it to lead with the feeling the feature produces
- Keep the original feature language in the rewrite, just move it to the second clause
headline | classification | rewrite "500+ million downloads, learn 40+ languages" | Feature-first | "Finally speak Spanish without freezing up, one 5-minute lesson at a time." "Never lose your streak" | Feeling-first (loss-aversion pride) | no rewrite needed
Healthy
The team can point to a specific feeling word in the winning headline, not just 'it converted better.'
Unhealthy
Every headline in the account opens with a number (downloads, price, lesson count) and nobody has tested a feeling-first version.
What this means
A feature-first headline talks to System 2, the part of the brain that's already tired and skeptical by the time it reads an ad; a feeling-first headline talks to System 1, which decides in about 50 milliseconds.
So what do I do about it?
| Symptom | Action | Effort |
|---|---|---|
| Every headline in the account leads with a number or spec | A/B test a feeling-first rewrite of the top 3 spending ads against the current feature-first versions | 30 min |
Step 02 of 02
The lesson's rule: ask customers why they bought, their stated reasons are the logical scaffold System 2 built after System 1 already decided, and that language belongs in retargeting and case studies, not in the first headline.
Collect 3 real 'why did you buy/subscribe' answers (from reviews, a quick survey, or people you know who use the product). Do the stated reasons sound like justifications, and would a feeling-first or feature-first headline have gotten them there first?
Procedure
- Create a 1-question Google Form: 'Why did you start using [product]?'
- Send it to 3 real users or pull 3 real public reviews mentioning a reason
- Log each answer, and note whether it reads as a feature justification (price, spec) or a feeling (relief, pride)
- Compare against the headline classification from Step 1, does the stated reason match what likely got their attention first?
answer 1: "It's cheaper than a tutor and I can do it on the bus." (feature justification, arrived after the fact) answer 2: "I felt embarrassed I couldn't order food in Spanish on my trip, so I just wanted to not feel like that again." (the actual feeling that drove the decision)
Healthy
At least one answer names an emotional trigger (embarrassment, relief, pride) that predates the logical reasons listed afterward.
Unhealthy
The survey only asks 'rate our features 1-5' and never gives the customer room to describe the moment or feeling that triggered the purchase.
What this means
The feature-sounding answers aren't wrong, they're just System 2's paperwork; the feeling named alongside them is the System 1 decision that actually happened first.
So what do I do about it?
| Symptom | Action | Effort |
|---|---|---|
| Case studies and testimonials only quote feature reasons | Ask one follow-up question in every customer interview: 'What were you feeling right before you decided to buy?' | 5 min |
Final deliverable
A 3-headline feeling-vs-feature classification with rewrites, plus 3 real customer 'why I bought' answers mapped against them.
See a reference example
Chewy, headline audit (excerpt) headline: "Free shipping on orders $49+" | Feature-first rewrite: "Never run out of food for your best friend again, free shipping on orders $49+." real answer collected: "I signed up the week after I forgot to reorder and had to drive to a gas station for cat food at 11pm." (the feeling: panic/relief, not the free-shipping feature)
Success criteria
You're done when you can:
- Correctly classifies all 3 headlines and rewrites every feature-first one
- Collects at least 2 real customer answers and identifies the underlying feeling in at least one