Does the new look still mean a calm evening?
Separate aesthetic approval from the promise the brand must convey.
“exceptional craft, genuine connection, and welcoming coffeehouses.” (opens the source)
The question
Harbor Brands and its premium tea brand Luma are fictional. Luma stands for a calm evening ritual. Two random, separate cells of 200 category buyers see either the current concept or a new AI-generated concept. Both cells answer the same questions.
| Respondents | Associate with a calm evening | Like the appearance | |
|---|---|---|---|
| Current | 200 | 140 | 118 |
| AI concept | 200 | 88 | 160 |
Association and liking are separate questions. A respondent can answer yes to both; these columns are not disjoint groups.
A concept must achieve at least 70% calm-evening association before a limited product test. This invented gate does not establish sales or a statistical difference.
“Choose your next action and explain the evidence behind it.”
Not saved by this site. Your browser may restore text when you navigate back.
A strong answer, step by step
Open each step in turn. This is one strong answer, not the only one.
Step 1: State the intended meaning
What would you do here? Open the move you’d make.
Judge each concept against the calm-evening promise.
The strong move. The meaning is the decision standard, not a design preference.
Choose whichever image most people like.
Tempting, but it misses something. Liking does not measure whether the concept communicates Luma’s intended occasion.
See how a strong candidate says it
“I would start with the calm-evening promise and the people who value it. Liking an image is useful, but it does not show that the brand still means the right thing.”
Why it works: The meaning is the decision standard, not a design preference.
Step 2: Compare the two responses
What would you do here? Open the move you’d make.
Revise the concept that loses the intended meaning.
The strong move. The attractive concept fails the stated meaning gate.
Launch the AI concept because liking rises 21 points.
Tempting, but it misses something. The AI concept misses the 70% association gate by 26 percentage points.
See how a strong candidate says it
“The current concept has 70% calm association and 59% liking. The AI concept has 44% association and 80% liking. I would revise the AI concept before a product test.”
Why it works: The attractive concept fails the stated meaning gate.
Step 3: Use the revised evidence
A revised concept is tested in a new, comparable cell of 200 category buyers. It records 148 calm-evening associations and 146 approvals of its appearance.
What would you do here? Open the move you’d make.
Test the revised concept with the product.
The strong move. The revised evidence supports a different next step.
Reject all AI concepts because the first one failed.
Tempting, but it misses something. The revision meets the stated gate; the first result does not determine every later version.
See how a strong candidate says it
“The revision has 148/200 = 74% calm association and 146/200 = 73% liking. It now passes the meaning gate, so I would take it into the limited product test.”
Why it works: The revised evidence supports a different next step.
Step 4: Name what remains unknown
What would you do here? Open the move you’d make.
Check product experience and competitive meaning next.
The strong move. A concept response is one piece of brand evidence.
Forecast a 74% purchase rate.
Tempting, but it misses something. The 74% result measures an association, not a buying decision.
See how a strong candidate says it
“I would test the revised expression with the actual product and compare it with competing brands. These responses do not establish purchase, repeat use or a unique position.”
Why it works: A concept response is one piece of brand evidence.
The answer, in short
- Decision
- Advance the revision to a limited product test.
- Because
- It meets the meaning gate without relying on aesthetic approval alone.
- What would change it
- Weak product fit or confusion with competitors could require another revision.
Common traps
- Treating liking as the brand promise.
- Converting an association into a sales forecast.
Check your answer
- I named the intended occasion.
- I kept liking separate from meaning.
- I used the revised cell.
- I named the limits of the evidence.
Last reviewed Nov 10, 2026