# Entrepreneurship — steal this assignment
_EL3vate 2026 · Day 8 · Part 02 of 15 · build 3dfcaff_

## Try it Tuesday (90 minutes)
90 minutes, no fabrication. Minutes 0–15: each team writes one value proposition and, beneath it, the single riskiest assumption it depends on, in one sentence. Minutes 15–45: teams use the landing-page prompt to generate a one-file site plus two headlines that make genuinely different promises, then paste all three headlines into a shared doc. Minutes 45–75: a hallway test — every student walks to another team, shows the three headlines cold, and records which one the stranger understood first and what they thought the product was. Minutes 75–90: teams report the gap between the headline they loved and the one that actually communicated. The deliverable is the hallway-test notes, which grade whether the team can tell traction from self-flattery.

## The assignment
**Make it (3D print · laser cut).** A physical MVP good enough to put in a customer’s hands. Not a finished product: a form study, a fit test, a packaging mockup. The point is that customer discovery interviews change completely when there is an object on the table.

**Build it (AI chat · no code).** A working landing page for the venture, built in one prompt, with an email capture and three headline variants. Students run traffic from their own networks and report the actual conversion rate, not their opinion of it.

## 4-week plan
- **Wk 1.** Teams commit to one value proposition and write the riskiest assumption underneath it in a single sentence.
- **Wk 2.** Build the landing page live in class and submit the print or cut file at the start of the week so fabrication runs in parallel with the digital test.
- **Wk 3.** Run the test: fifteen real interviews and whatever traffic they can generate. No pivoting yet, just collection.
- **Wk 4.** Prototype in hand, data in hand, teams present what they were wrong about. Grade the quality of the disconfirmation, not the size of the number.

## What this replaces
- **Replaces:** The written business plan, or the polished pitch deck, submitted at the end of the venture course.
- **What is lost:** The comprehensive financial and operational planning a full business plan forces.
- **What is gained:** Students get a real signal from a stranger in one session instead of a graded document nobody outside the class will read; the assignment rewards disconfirmation over presentation polish.

## Where AI is bad at this
Ask a model to build a landing page and it produces confident, conversion-optimized copy for a market it has never observed — asserting benefits, inventing a target-customer objection, and implying demand no interview has established. Its fluency at marketing language makes an untested idea look validated, which is the precise opposite of what customer discovery is for.

## Rubric
| Criterion | Weight | What it assesses |
|---|---|---|
| Riskiest-assumption clarity | 20% | The stated riskiest assumption is genuinely the thing most likely to sink the venture, specific enough to be tested. |
| Quality of disconfirmation | 35% | Grades whether the team sought evidence that would prove them wrong and reported it honestly, rather than counting vanity signals. |
| Signal vs. noise interpretation | 25% | Distinguishes a real preference signal from politeness, curiosity, or the team's own hope. |
| Iteration decision | 20% | The pivot-or-persevere call is justified by the evidence collected, not by attachment to the original idea. |

## Starter prompt
> Build me a single-page landing site for a new product, as one self-contained HTML file I can host anywhere. The product is: [describe it in two sentences]. The customer is: [describe them]. Include a headline, three short benefit statements, an email signup field, and nothing else. Then give me two alternate headlines that make a meaningfully different promise, not just reworded versions of the first. Keep the copy plain and specific; no marketing filler.

## Budget & logistics
- **Instructor prep:** 1 hours
- **Class time:** 90 minutes
- **Per-student cost:** $0 for the Tuesday version; roughly $8–15 per team in print or cut stock **[unverified]** for a physical MVP in the four-week version.
- **Fabrication file due:** First day of Week 2 — submit the print or cut file at the start of the week the landing page goes live. PACE quotes 7–10 business days (up to 14 calendar days), so it clears before the week-4 presentation.
- **Calendar dependency:** The digital test needs no lead time. Submitting the MVP file at the start of week 2 clears the up-to-14-calendar-day turnaround before the week-4 presentation.

## Three sizes
- **One session:** The 90-minute value-proposition and hallway-test session above; no fabrication.
- **4 weeks:** The seeded four-week plan: riskiest assumption, live landing page plus fabrication file, real interviews and traffic, then a what-we-were-wrong-about presentation.
- **One semester:** Run a real venture through a full build-measure-learn loop with multiple pivots, a fabricated MVP iterated twice, and a final report graded on what the team was demonstrably wrong about.

---
_PACE · Shidler College of Business · University of Hawaiʻi at Mānoa · pace.shidler.hawaii.edu/maker_
_All fifteen assignments, the demos and every handout: https://el3vate.vercel.app_