§03//BLOG
§ log.008 — i scored 6 product ideas with a spreadsheet and still haven't found the right one
When EAC threatened WrapsRL's future, I built a weighted decision matrix across 6 niches, picked one, and shipped within 24 hours. Then I realized the spreadsheet can't tell you if an idea actually hits.
The EmotePack.ai pivot session from May 1 was the most productive 24 hours I've had this year.
Not because I shipped something amazing and launched into orbit. Because I stopped what I normally do — agonize over product ideas for weeks — and let a spreadsheet make the call so I could start building immediately.
The framework worked. The idea? Still figuring that out.
the trigger
WrapsRL is running ($215/mo MRR, growing slow). But EAC blocks mods in online play. That caps the ceiling whether I like it or not. I need something else that can grow without being limited by how many Rocket League players install BakkesMod.
The natural instinct is to panic-ship whatever comes to mind first. Done that. Never works.
So I opened a spreadsheet.
the contenders
Six niches that met my bar: not a feature, not a toy, small enough to dominate, big enough to matter.
1. EmotePack.ai — AI-generated custom emote packs for streamers
2. Adversec (adversec.io) — already built and live. API-first adversarial testing with Supabase auth and a Railway backend.
3. MCWraps — Minecraft resource pack skins via AI
4. Rocket League replay analysis — coaching tool
5. AI podcast clips — auto-highlight extraction
6. Discord bot marketplace — template-based bots
the matrix
Five weighted criteria. Scores out of 10. No gut feel — just math.
| Criterion | Weight | Why it matters |
|---|---|---|
| Demand signal | 30% | Is anyone actually searching/buying this? |
| Build complexity | 20% | Can I ship in under 2 weeks? |
| Margins | 20% | Can I charge $5-10/mo without being laughed at? |
| IP/legal risk | 15% | Will Nintendo/Epic/Mojang come for me? |
| Defensibility | 15% | How fast can someone clone it? |
the results
| Idea | Demand | Build | Margins | IP Risk | Defensibility | Total |
|---|---|---|---|---|---|---|
| EmotePack.ai | 8 | 7 | 9 | 9 | 6 | 7.85 |
| Adversec | 6 | 6 | 8 | 10 | 8 | 7.40 |
| MCWraps | 7 | 6 | 7 | 4 | 5 | 6.00 |
| RL replay analysis | 5 | 4 | 6 | 9 | 4 | 5.55 |
| AI podcast clips | 7 | 5 | 5 | 8 | 4 | 5.85 |
| Discord bot marketplace | 6 | 3 | 6 | 8 | 3 | 5.25 |
then i shipped
I deployed a landing page that same day. Vercel. Waitlist form. Drafted launch posts. Set pricing. Done.
The spreadsheet did its job: it killed the deliberation. I wasn't deciding what to build — I was building what was already decided.
the part the spreadsheet didn't catch
Here's the honest part.
I built EmotePack.ai. It works. But the website design didn't land the way I wanted. When I stepped back and looked at it, the whole thing felt basic. Like a generic AI wrapper with a waitlist form. Nothing about it felt differentiated or defensible in the way a product needs to be.
The matrix scored EmotePack.ai high on build complexity and margins. It didn't score how the idea felt once you built the landing page and stared at it. It didn't measure whether I'd be excited to wake up and work on it. It didn't measure whether a basic AI wrapper in a crowded space actually converts.
So I'm still looking.
why the framework is still useful
I'm not saying the matrix failed. It did exactly what I needed it to do — force a decision fast so I could stop theorizing and start building. The problem isn't the framework. The problem is that no spreadsheet can tell you if an idea has legs until you actually put it in front of people.
What the matrix did tell me:
- MCWraps would be a legal nightmare (IP risk 4/10)
- Adversec is already live and the most defensible option (8/10), just needs distribution push
- EmotePack.ai was the fastest path to a new shipped product (Adversec was already live)
It narrowed the search space from 6 to 2-3 viable candidates. That's valuable. It just didn't hand me a finished product.
what i'm doing now
Back to the board. The matrix narrowed things down. Execution revealed more questions. Now I need to either:
- Redesign EmotePack.ai and give it another shot with a better pitch
- Double down on Adversec — it's already live (adversec.io), just needs proper distribution
- Run a fresh round of brainstorming with tighter criteria
The D14 pivot plan I wrote during that session is still in effect. If the idea doesn't hit, move on. That part hasn't changed.
---
The spreadsheet got me from analysis paralysis to a deployed app in 24 hours. That's real. But deploying doesn't mean you're done — it means you've finally started the part of the process that actually tells you if you're onto something.
If you're stuck deciding between ideas, try the matrix. It'll force you to ship faster. Just don't confuse shipping fast with having found the answer.
$ echo "frameworks help you decide, execution tells you if you were right"
frameworks help you decide, execution tells you if you were right