How bad is your tech debt, really?
15 questions. About 5 minutes. A concrete score across 5 dimensions, plus an honest review of what it means for your AI plans.
- 15 questions
- About 5 minutes
- Score in 5 key dimensions
- An actionable start
- What it means
for your AI strategy
What you get at the end
- Your score (0–120) and your zone: Green, Yellow, Orange, or Red.
- Which of the 5 dimensions is dragging you down the most, and why.
- A clear, specific plan for what to do next. If you score Green, keep doing what you are doing.
- What your score means for your AI plans, without the hype.
“The result is specific enough to disagree with. That's the point.”
Where your debt is concentrated
- Team & Knowledge16/24
- Velocity14/24
- Stability12/24
- Organization12/24
- Technology10/24
Start with Team & Knowledge. It's your heaviest dimension, and where the cost compounds first.
Judge the questions before you spend the 5 minutes
You can tell a lot about a diagnostic from its questions. We write these for founders and leaders of growing software companies. Here are 3 of the 15 we ask.
Look at your roadmap from 12 months ago. How many planned features did you actually deliver?
- All of them
- 70-90%
- 40-69%
- Less than 40%
- We don't have a roadmap, to be honest
How often does fixing one bug cause another bug?
- Rarely
- Sometimes, about 1 in 10 times
- Often, we're afraid to change things
- Continuously, we have whole teams dedicated to fixes
You've hired developers in the past year. Did development speed up proportionally?
- Yes, it sped up noticeably
- Sped up a bit but not proportionally
- Stayed about the same
- Development actually slowed down
- We didn't hire anyone
If any of those made you wince, you already know which direction you are facing.
How the scoring works
15 questions, 5 key dimensions. Every answer carries points, which are weighted and normalized. Your final 0–120 score puts you in one of 4 zones.
- Velocity — How long things take, and whether your roadmap has any relation with reality.
- Stability — Is first-deploy super frustrating? How much time is spent on fire-fighting?
- Team & Knowledge — Who actually understands your core logic, and what happens when they go on vacation?
- Technology — Framework age, test coverage, and how much force is required to make things grow.
- Organization — Whether filing a bug gets fixed, or slips into a black hole.
The thresholds are based on empirical assessment of over 500 tech companies. This tool's purpose isn't to make you feel bad, but to provide an honest, objective baseline of your software environment.

Who's grading you
I'm David Melich. I've been building software for money since 2003, I teach web applications to master's students at UPJŠ in Košice, and I run GrownApps — the team companies call when the codebase that got them here starts fighting the roadmap ahead.
These 15 questions are the same ones we ask at the start of every audit. We made them free because most teams don't need an audit yet — they need to know which dimension is quietly getting worse.
- Rockpoint Legal Funding — from a portfolio under $2M to $150M+ over a 9+ year partnership
- Previo — a channel manager processing 100,000+ daily updates with zero lost reservations
- Tentacles IoT — a working demo in weeks after their previous supplier left the apps unfinished
The questions people ask before they start
You get your full result, then a short sequence of emails matched to your zone — what teams in your situation usually do next. Unsubscribe is one click. Nobody calls you unless you book a call yourself.
No. The zones are fixed before you answer — your score is the sum of your answers, nothing else. If you land in Green, the result says keep doing what you're doing, and the only link you'll want is a newsletter.
It maps your symptoms. It can't see causes, because it never reads your code. But self-reporting is more honest than it sounds. When you pick 'we have areas nobody wants to touch,' that's data no static analysis tool will ever give you.
That's the best time to run it. Most teams we work with aren't broken — they're healthy companies whose codebase is slowly falling behind their ambitions. Catching Yellow before it turns Orange is the whole point of measuring.
Please do. Comparing results is the interesting part: when the CEO and the CTO score the same codebase two zones apart, that gap is a finding in itself. Same link, separate runs.
No repo access, no metrics dashboard, no prep. If you can answer 'how long does a medium feature take from spec to production,' you're ready.
It can't read your code, so it can't tell you what to fix first or what the fix costs. That's what a technical audit does — ours start at €4,000 and end with a written report and a prioritized recovery plan. But the scanner first. If your result says you need more, you'll know.
Not ready to scan? The Hard Ship is our biweekly newsletter — notes on tech debt cleanup, platform renewal, and what AI actually changes about both. Written by me. Subscribe to The Hard Ship.
Find out where you actually stand
15 questions. About 5 minutes. No email to start.
Start the scannerTech debt compounds whether you measure it or not. Measuring it is the only part that's free.




