Phase 1: Planning (An Afternoon, Not a Month)

In one line: Solo planning is short and direct — a paragraph pitch, five features, a sketch, and a go/no-go. If it takes longer than an afternoon, you're overthinking.

In plain English

Big companies spend weeks in product requirements docs because they have to align dozens of people. You have to align you. Your "PRD" is a paragraph in Notes.app. Your "design review" is a rectangle drawn on paper. Your "stakeholder sign-off" is asking yourself whether you'd actually use this in three months.

Step 1: Write the Pitch

In one paragraph, answer:

• What does it do?
• Who is it for?
• Why would they use it instead of alternatives?

If you can't answer in one paragraph, you don't understand it well enough yet.

Example:

"ShelfTrack is a web app for solo readers who want to track which books they're reading without a social network. Goodreads is bloated with social features and ads; ShelfTrack is just a clean shelf with progress tracking, ratings, and optional reading streaks. For me first, then anyone like me."

Step 2: List the v1 Features

Five or fewer. If you have more, you're not at v1 yet.

Example for ShelfTrack:

1. Sign up / sign in.
2. Add a book to my shelf (search by ISBN or title).
3. Mark a book as currently reading, finished, or want-to-read.
4. Rate finished books (1–5 stars).
5. See my shelf as a list.

That's it. Reading streaks, recommendations, social features, exports — all v2+.

Step 3: Sketch the UI

Paper, Figma, or just a Word doc with rectangles. Don't spend more than two hours.

Step 4: Decide if It's Worth Building

Ask:

• Would I use this every week?
• Is there a clear next user beyond me?
• Can I build v1 in 2–4 weekends?
• Will I still care in 3 months?

If "yes" to most: build it. If "no" to most: skip it. Side projects you don't finish accumulate in your graveyard and demoralize you.

Worked example: a real go/no-go

Idea: "An app that helps me track which of my friends I've seen each month."

• Would I use it weekly? Yes, I forget who I haven't seen recently.
• Clear next user? Yes — anyone with a wide social circle.
• 2–4 weekends? Yes — it's a CRUD app over a "friends" table with a date column.
• Still care in 3 months? Honest answer: probably no. I'd use it for a month then forget.

Verdict: Don't build. Set a calendar reminder instead.

That five-minute thinking exercise just saved a month of work on a project that wouldn't have stuck.

Highlight: this is not skipping the work

"Plan less" is not "don't plan." It's "plan in proportion to the project." A weekend tool deserves an afternoon of thought, not three. A 6-month SaaS deserves more — but still in days, not weeks. The point is to match planning depth to project depth so you don't spend more time planning than building.

Common mistakes

Where people commonly trip up

• Inflating the pitch into a pitch deck. You don't need a problem statement, market sizing, and a competitor matrix — those are artifacts for convincing other people. The fix is one paragraph in plain English. If it doesn't fit, you don't understand the idea yet.
• Smuggling v2 features into the v1 list. "Just a small social layer" or "just basic notifications" or "just an export button" — each one quietly doubles the build. The fix is brutal: if a feature isn't required for the first user to get value once, it's v2. Write it down somewhere else and move on.
• Lying to yourself on "will I still care in 3 months." Most solo developers say yes here automatically because the idea feels exciting today. The fix is to remember an idea from six months ago that excited you the same way and is now untouched. Same neurons, same trap.
• Sketching in Figma for two days. Pixel-perfect mockups before code is enterprise muscle memory. The fix is rectangles on paper, photographed, dropped in a Notes file. The sketch exists to clarify your thinking, not to brief a designer.

Page checkpoint

Checkpoint Quiz

Did solo planning stick?

Required

What's next

→ Continue to Phase 3: Stack Selection where the 2026 default stack is basically pre-decided.