# The Infrastructure Mismatch

*Why You're a Staff Engineer at Work and a Ghost at Home*

![Hero](https://cdn.marblism.com/TxyEpFYzOGB.webp "The Infrastructure Mismatch: Why You're a Staff Engineer at Work and a Ghost at Home")

You crushed Q4. Shipped three major features. Led a migration that touched every system in your platform. Your manager called you "the glue" that holds the team together.

Then you come home.

That side project? The one that's been sitting in your head for eighteen months? Still there. Gathering dust. You spent maybe three weekends writing code, read some blog posts, felt momentum building, then nothing. It's back to being a vague cloud on your mental horizon, competing with everything else that screams louder.

Here's what you tell yourself: *I'm lazy. I lack discipline. Other people have side hustles, why can't I?*

Wrong diagnosis.

You're not lazy. **You're operating without the infrastructure that makes you effective at work.**

## The Biology of Paralysis

Your brain isn't wired to execute vague, sprawling projects with no clear path forward. At your day job, you operate inside a machine: standups, roadmaps, sprint planning, weekly check-ins, a manager who asks about blockers, a definition of "done" that everyone agrees on.

For your side project? You have a wish and a to-do list.

Your brain treats uncertainty like a threat. A project with no deadlines, no scope boundaries, and no accountability triggers the same ancient threat response as staring into a dark cave. Your brain doesn't know what's inside. It could be a predator. So it conserves energy and avoids the risk.

This isn't weakness. This is biology.

You can't beat biology with willpower. You beat it with structure.

![Project chaos vs structured infrastructure](https://cdn.marblism.com/kqiKz-YBLB2.webp "Transforming tangled uncertainty into organized systems")

## The Infrastructure Audit: Where Are You Actually?

Answer these eight questions. No judgment. Just data.

1. Do you have a specific, non-negotiable ship date for your project?
2. Can you describe your V1 outcome in a single, concrete sentence?
3. Have you broken the project down into 3-5 major, provable milestones?
4. Do you have a weekly check-in scheduled with yourself or an accountability partner?
5. Have you told at least one person your ship date?
6. Is your next single action identified and scheduled in your calendar?
7. Do you have a written "NOT NOW" list for features that are out of scope for V1?
8. Do you have a dedicated, recurring time block for project work in your calendar?

**Give yourself 1 point for every "Yes."**

- **7-8 Points**: You're already building like a pro. You probably don't need this article.
- **4-6 Points**: You have some structure, but it's inconsistent. You're here because you need to systematize what you're already trying to do.
- **0-3 Points**: Your project is operating in a vacuum. That vague feeling that you "should" be working on it but aren't? This is why.

Most people score between 0-3. Not because they're incompetent — they're staff engineers, for crying out loud — but because they've never built infrastructure for personal projects. Why would they? No one teaches this.

## The RICE Reality Check: Most Ideas Aren't Worth Your Time

Let's be direct: most ideas are bad. Not bad as concepts, but bad as *things for you to build right now*.

You need a framework that cuts through enthusiasm and shows you reality. Enter **RICE**: Reach, Impact, Confidence, Effort.

Score each dimension on a scale of 1-10:

- **Reach**: How many people would use this in the first 3 months? (1 = you, 10 = thousands)
- **Impact**: How much would it improve each person's life/work? (1 = nice to have, 10 = life-changing)
- **Confidence**: How sure are you people want this? (1 = guessing, 10 = validated with dozens of conversations)
- **Effort**: How many weeks to ship? (1 = one week, 10 = three+ months — this one's inverted)

**Formula**: `(Reach × Impact × Confidence) / Effort`

![RICE framework decision matrix](https://cdn.marblism.com/SI0Tv4Zhra2.webp "Scoring and validating project ideas by reach, impact, and effort")

### Example 1: Personal Finance App

**The Idea**: A web app where users log expenses and see a dashboard by category.

- **Reach**: 50 people (your network + some organic reach) = 5
- **Impact**: Moderately helpful, but Mint and YNAB exist = 6
- **Confidence**: Asked 3 friends, got lukewarm "yeah, maybe" responses = 4
- **Effort**: Auth, database, UI, dashboard, deployment. 5-6 weeks = 8

**RICE Score**: (5 × 6 × 4) / 8 = **15**

### Example 2: PDF Template Kit for Freelancers

**The Idea**: Five customizable PDF templates (invoice, proposal, contract, timesheet, feedback form) that freelancers buy once and use forever.

- **Reach**: Posted in freelancer communities, got consistent interest. 200+ people first 3 months = 7
- **Impact**: Saves 30-60 minutes per project. For 10 projects/year, that's 5-10 hours saved = 7
- **Confidence**: Surveyed 15 freelancers. 12 said they'd buy. Found a 5,000-person Facebook group actively seeking this = 9
- **Effort**: Design in Canva (one week), set up Gumroad (one hour). Total: 1.5 weeks = 2

**RICE Score**: (7 × 7 × 9) / 2 = **220.5**

The finance app *feels* more impressive. It's a "real app" with a backend. It's what you imagine when you think "startup."

But the RICE score is 15.

The template kit is lower-tech, less glamorous, and the RICE score is **220.5**. Nearly 15 times higher.

This is the power of RICE. It cuts through ego and shows you what's actually worth your time.

**What's a good score?**

- Below 50: Effort too high relative to reach/impact. Your idea is too big or the market signal is too weak.
- 50-150: Viable, but you need strong validation. If you haven't talked to real people, you're guessing.
- Above 150: Strong signal. Worth your 30 days.

## The 10-Conversation Method: Google Won't Validate Your Idea

Here's what most people do when they want to validate: they Google it. They read 47 tabs about market size and competition. Three hours later, they still have no idea if their specific idea is something people actually want.

This is research theater, not validation.

Real validation comes from exactly one place: **conversations with real humans who might use your product.**

You don't need thousands. You need exactly 10. With the right people. In real-time (phone, video, in-person — async doesn't work).

![Customer validation through conversations](https://cdn.marblism.com/H9u5ss6iW_e.webp "Conducting market research calls to validate product ideas")

### Who to Talk To

**5 Warm Contacts:**

1. Someone in your professional network who fits your target user
2. Someone from a community you're in (Slack, Discord, forum)
3. Someone who's mentioned the problem you're solving
4. Someone who's bought a similar product
5. Someone who's asked you for related advice

**5 Cold Contacts:** Post in relevant communities (Reddit, Facebook groups, Indie Hackers). Message people directly. Offer a $10 gift card for 20 minutes. Most say no. That's fine. You only need 10 yes's.

### The 5 Core Questions

When you get someone on the phone, you have 20 minutes. Don't pitch. Ask:

1. **"Walk me through your current workflow for [problem]. What's the messiest, most frustrating part?"** (Listen for genuine frustration — that's signal.)
2. **"How much time do you spend dealing with this? How does it impact your work/life?"** (This tells you magnitude. "Annoying" is different from "costs me $10K/year.")
3. **"Have you tried existing solutions? What happened?"** (This reveals real competition and whether there's a legitimate gap.)
4. **"What would an ideal solution look like?"** (Let them describe it. Don't suggest features. Listen.)
5. **"If I built [very basic version], would you use it?"** (This is the validation question. "Yes, I'd definitely use that" is signal. "Yeah, maybe" is not.)

### How to Interpret

- **Strong signal**: 8+ out of 10 say "yes, I would use that." Bonus: they say "I need this" or "I'd pay for that."
- **Weak signal**: 4-7 say yes, or yes with caveats like "once you add X feature."
- **No signal**: 3 or fewer say yes, or they're just being polite.

Weak or no signal means the same thing: your idea needs to pivot, or you're talking to the wrong people.

## The Real Competence Trap

You're a staff engineer. You *can* build the full-featured app with auth and a backend and real-time sync and all the bells and whistles.

But should you?

The competence trap isn't about lacking skills. It's about using the wrong skills at the wrong time. You're bringing enterprise-level engineering to a validation problem. You're over-engineering because that's what you're good at.

Here's what I learned after 20 years in tech: your V1 is not your vision. Your V1 is your launch pad. It's the embarrassingly small thing you ship to gather evidence that your idea has legs.

When I finally shipped my first product, it wasn't the comprehensive video course with professional filming and community forums I'd envisioned. It was a **4-module PDF with fillable worksheets, sold on Gumroad for $47**.

It took 5 weeks to build. I made $2,400 in the first month selling to complete strangers.

That $2,400 did more to build my confidence than any number of compliments. It was evidence. It was real.

![Over-engineering vs MVP](https://cdn.marblism.com/OpbIYOGZm11.webp "Choosing a simple V1 launch over complex feature-heavy development")

## Stop "Wanting" to Ship. Start Building Infrastructure.

You don't need more motivation. You don't need to "find the time." You need to build the same infrastructure that makes you effective at work.

Here's where to start:

1. **Run your Infrastructure Audit.** Score yourself on those 8 questions. Know where you actually stand.
2. **Score your idea with RICE.** If it's below 50, kill it and move on. If it's above 150, you have signal worth pursuing.
3. **Have 10 conversations.** Not 3. Not 5. Ten. Real humans. Real-time.
4. **Define your V1 in one sentence.** If you can't, your scope is still too big.
5. **Pick a ship date 30 days from today.** Tell at least one person.
6. **Block recurring build time on your calendar.** Treat it like a meeting you can't miss.

The mismatch only exists because you haven't built the structure yet. So build it.

---

## About the Author

Molly Shelestak is a Build Partner for Side-Project Shippers. With 20+ years in tech — from Google to Heap to Contentsquare — she helps senior tech employees stop tinkering and actually ship their side projects in 6 weeks.

## Related

- [The V1 Manifesto](https://www.unstuckwithmolly.com/writing/v1-manifesto)
- [The Embarrassment Test](https://www.unstuckwithmolly.com/writing/embarrassment-test)
- [Work With Me](https://www.unstuckwithmolly.com/work-with-me)