Interview 3 real customers, build 2 honest personas
Personas made up in a meeting room are fiction. Pick a product, interview 3 real potential customers about their actual habits and frustrations, then use AI to help find the patterns — and build 2 personas where every claim traces to something a real person said.
Posted by The AI Internships
The work brief
- 01Pick a product/service with customers you can reach (a local shop, an app your friends use, a tuition class).
- 02Use AI to draft your interview questions, then cut anything leading — you want habits, frustrations, and exact words.
- 03Interview 3 real potential customers, 10 minutes each: their current habits, what annoys them, what they’d pay for, exact words.
- 04Use AI to spot patterns and differences in your notes — that’s where personas split — then verify each one against what people actually said.
- 05Build 2 personas: name, context, goals, frustrations, buying triggers — each detail tagged to which interview it came from.
What you’ll produce
Submission standard
Submit the product, your interview notes with real quotes, the 2 personas with interview-traceability, and what surprised you. Also include your AI workflow — tools, best prompts, and what you changed from the AI’s output. Attach a link proving at least one interview really happened (recording, or a photo/screenshot of the chat), with names blurred if needed.
The product and why these 3 people are potential customers
Short answerRequiredNotes from all 3 interviews, with at least 2 direct quotes each
Tag each interview with the person’s first name/initials and when/where it happened; include ≥2 direct quotes each.
Written responseRequiredThe 2 personas — every key detail tagged [I1]/[I2]/[I3] to its source interview
Written responseRequiredWhat the interviews told you that you’d never have guessed
Written responseRequired
You’ll complete these inside your private workspace.
What you must submit as proof
This brief requires evidence an AI can’t fabricate.
Link to proof of at least one real interview (voice-note, recording, or a chat screenshot)
Ask their consent first, then blur every name, face, and phone number before you upload. We need proof the interview happened, not who it was with — a visible unblurred identity is an automatic fail.
Public linkRequired
Submissions without this evidence cannot be submitted.
Protect other people in your proof. Blur faces, names, phone numbers and email addresses before you upload, and refer to anyone you worked with by role or number ("Listener 1", "the stall owner"). Your proof is only ever used to check your work — it is never published, never appears on your certificate, and is never shown in your public portfolio.
How your work is evaluated
The passing benchmark is 70/100.
Real interviews
22%Notes read like 3 distinct real people with specific, messy detail.
Persona grounding
22%Persona claims trace to interview evidence, not stereotypes.
Insight
11%The surprises are genuine discoveries that would change marketing decisions.
Proof of real people
44%There is first-party proof at least one interview happened, and the notes/personas read as 3 distinct real people whose claims trace to what was said. If proof is missing, or everything reads in one AI voice, this criterion is a fail regardless of persona polish.
How we grade your AI usage
Using AI is the point — it’s the skill this certificate proves. You’ll answer three short questions about how you used it: what you asked, what was wrong with its first answer, and what you changed. Specific, honest answers score high. “I pasted the brief and submitted the answer” scores near zero.