Five ideas screened down to one product. AI tools used to push the concept into a demo good enough to pitch on stage. My first end-to-end run from idea to go-to-market.
The brief: take a blank idea through to a market-ready plan, using the methodology of a product manager. Three months, full chain.
This course doesn't reward "good PPT obedience." It rewards applying the method to a real problem: scanning the segment, converging via idea screening, then replacing expensive product photography & prototypes with AI tooling so the idea looks like it already exists when we pitch.
We landed on a fall-prevention belt — a wearable hip airbag that releases within 0.3 seconds of detected imbalance and cushions the impact on the floor. Falls are the single biggest cause of disability and lost autonomy for the 65+ population. The need is urgent; the market is growing fast.
The work split into two tracks: (1) Idea Validation & Market Analysis — five ideas scored on market growth × pain urgency; (2) AI-Enabled Concept Development — AI image & video generation for product visuals and usage scenarios, plus a complete 4P plan.
Two axes: market growth × pain urgency. The simplest and the hardest call in the whole course — overshoot and you get vanity, undershoot and you miss the real problem.
Senior falls aren't a "be careful" problem. A single fall is often the moment a person stops being independent. This product isn't solving inconvenience — it's solving the loss of autonomy.
WHO data: about 1/3 of people 65+ fall each year, and 50% of those who fall will fall again within twelve months.
One-year mortality after a hip fracture is roughly 20–30% — one of the most common ways an older adult loses the ability to live alone.
Most existing products detect or alert (passive). Almost no waist-mounted product can actively protect within 0.3s. That's the real gap.
Before touching a hardware spec, we ran the deep-dive. One question — is this product worth building? Three pillars had to clear at the same time.
Can the technology actually deliver the 0.3-second protection promise — reliably, at the user's hip, in everyday motion?
Is the addressable market large enough to support a real product line — not a single-shot pilot?
Can we price this at premium and still hold a margin that funds the next iteration?
Two stacks have to clear at the same time: predict the fall early enough, and deploy the airbag fast enough.
China's elderly population is the largest pool we can realistically reach. We sized it conservatively, then layered on the wearable-fall-detection growth curve.
National Bureau of Statistics, 2024.
Conservatively, the 10% of the 65+ pool who are high-risk and have purchasing power or family support.
The wearable fall-detection & prevention market is projected to grow at a CAGR of 6.2% from 2025 to 2033, driven by the global aging population. (Grand View Research, 2023)
Pricing strategy: value-based premium. The reference is not a competing belt — it's the cost of not wearing one.
| Year | Target Market (M) | Penetration | Units Sold | Revenue (¥M) | Profit (¥M) |
|---|---|---|---|---|---|
| 1 | 20.00 | 0.04% | 8,000 | 14.4 | 8.0 |
| 3 | 22.52 | 0.10% | 22,520 | 40.5 | 22.5 |
| 5 | 25.35 | 0.19% | 46,900 | 84.4 | 46.9 |
Year-5 penetration is still under 0.3% of the 20M target — i.e. the forecast is conservative on purpose. Growth at CAGR 6.2%.
We pulled real reviews and interview voices from the existing fall-belt category. Six recurring complaints drove the iteration from V1 to V2.
"It beeps when I just bend over, it's annoying — and I can't hear it clearly anyway." Comments like this told us the comfort, alert and detection layers all needed a rebuild.
SureStepSafety covers the full timeline of a fall: before (smart early-warning), during (instant airbag), after (SOS auto-alarm via the app to family & emergency), and long-term (continuous health monitoring and trend reports for caregivers).
Two channels, two different audiences — the decider / purchaser online (adult children with budget authority) and the actual user offline (the senior who wears it).
A small group of early data and feedback that lets us project trend, sustainability and what to optimize next. Online vs. offline are tracked separately so we can decide whether to keep / scale offline stores.
A team of five. I owned the market & financial analysis on the Go / No-Go gate, and led the V1 → V2 iteration logic that turned user complaints into spec changes.
"Product manager" sounds like a collection of frameworks in class. Doing it taught me that the hardest part isn't the 4P — it's convergence. Picking one idea out of five means killing four others that all looked plausible.
Second thing: when budget is zero, AI is a leveler. We had no money for prototyping, no team for video shoots — but AI tooling pulled the visual density of the pitch up to "looks already real." I reused this exact playbook later in my internship.
The third thing — and the most important — who buys decides what the product looks like. The real buyer for this belt is the adult child, not the senior. So it has to look like an insurance product, not a medical device.