DEV Community

kanta13jp1
kanta13jp1

Posted on

The Indie Dev 90-Day Plan: KPI Design to Milestone Management

#ai #automation #indiedev #buildinpublic

The Indie Dev 90-Day Plan: KPI Design to Milestone Management

Stop building "somewhat." Three months, designed, measured, improved. Here's the framework I actually use.

Why 90 Days? 

1 week:   only urgent tasks get done
1 month:  slightly strategic, but too short to see trends
90 days:  trends emerge, tactics can be measured
1 year:   too long to maintain motivation
Enter fullscreen mode Exit fullscreen mode

90 days is the sweet spot: strategic enough to move the needle, short enough to stay motivated.

Phase 0: Goal Setting (Day 1-3) 

OKR format:
  Objective:
    "Provide [value] to [user type]"
    Example: Give indie developers a daily AI-powered development workflow

  Key Results:
    KR1: MAU 500 (current: 0)
    KR2: NPS +30 (current: unknown)
    KR3: MRR ¥50,000 (current: ¥0)
Enter fullscreen mode Exit fullscreen mode

KRs must be measurable. "Better UX" is not a KR. "4.2-min avg session" is.

Phase 1: Validation (Day 1-30) 

Month 1 goal:
  - Complete 5 user interviews
  - Release MVP (one core use case only)
  - Set a trial price and observe response

Success criteria:
  - 3 of 5 interviewees say "I'd use this every week"
  - DAU/MAU ratio > 20% (retention signal)
Enter fullscreen mode Exit fullscreen mode

Month 1 prioritizes learning over building.

Weekly check (every Friday):
  ✅ What did I learn this week?
  ✅ Was the hypothesis right?
  ✅ What's the top priority for next week?
Enter fullscreen mode Exit fullscreen mode

Phase 2: Growth (Day 31-60) 

Month 2 goal:
  - Publish 4 SEO blog posts
  - Add 20 /vs-{competitor} pages
  - Automate weekly X posts via GHA
  - Land first paying customer

Success criteria:
  - Organic traffic > 300 sessions/month
  - Churn rate < 10%/month
  - MRR > ¥10,000
Enter fullscreen mode Exit fullscreen mode

Month 2 balances building with distributing.

Phase 3: Stabilization (Day 61-90) 

Month 3 goal:
  - Zero critical bugs
  - GHA Schedule automation reduces ops time 50%
  - Design the next 90-day plan

Success criteria:
  - MAU up 20% month-over-month
  - Support time < 2 hours/week
  - Goals achieved within 20 hours/week of development
Enter fullscreen mode Exit fullscreen mode

Month 3 is about making the pace sustainable.

KPI Dashboard Design 

CREATE TABLE weekly_kpis (
  id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
  week_start DATE NOT NULL,
  mau INT,
  new_users INT,
  mrr_jpy INT,
  nps INT,
  churn_rate DECIMAL(4,2),
  recorded_at TIMESTAMPTZ DEFAULT NOW()
);
Enter fullscreen mode Exit fullscreen mode 
# Auto-collect KPIs every Sunday via GHA
on:
  schedule:
    - cron: '0 8 * * SUN'
Enter fullscreen mode Exit fullscreen mode

Monthly Retrospective Template 

## [Month X] Retrospective

### Done
- [KR1] MAU: target 500 → actual 320 (64%)
- [KR2] NPS: target +30 → actual +22 (73%)

### Not done
- [KR3] MRR: target ¥50,000 → actual ¥18,000 (36%)

### What I learned
- Free-to-paid conversion is higher with a shorter trial

### Adjustments for next month
- Shorten trial from 30 days to 14 days
Enter fullscreen mode Exit fullscreen mode

Summary 

90-day framework rules:
  1. OKR with measurable targets
  2. Month 1: validate / Month 2: grow / Month 3: stabilize
  3. Weekly check-ins for course correction
  4. Automate KPI collection (GHA + Supabase)
  5. Monthly retro to carry learnings forward
Enter fullscreen mode Exit fullscreen mode

Shifting from "building somewhat" to "designing, measuring, adjusting" was the single biggest lever for making indie development sustainable long-term.

Top comments (0)