Project Maturity Tracking: Knowing When You’re Ready to Build

Project Maturity Tracking: Knowing When You’re Ready to Build

●10 ●23
calendar_today • schedule2 min read

Project Maturity: How Much of a Project Do You Actually Understand?
Socratic Maturity is part of an earlier research prototype. It is not maintained as a product. The percentages below are the library's scoring scale, not findings, and I have not tested at what score a project is ready to build.
Projects often start building before the problem is understood, and the rework comes later. The opposite also happens: a team keeps asking questions long after it knows enough. In both cases the person deciding has no number, only a feeling. I wanted to see whether a rough, explicit measure of "how much do we know" would help.
socratic-maturity is a small Python library that calculates that measure.
How the scale works
The library models a project as passing through four phases:

• Discovery: define the problem, scope and audience.
• Analysis: gather requirements and analyse data.
• Design: choose technology and architecture.
• Implementation: write code and test.

Each phase has scores in five quality categories: code quality, testing, documentation, architecture and performance, each from 0.0 to 1.0. The library:

• averages only the phases that are actually being worked on, so that starting a new phase does not lower the overall score;
• finds weak categories (below 0.6);
• estimates the current phase from the overall score;
• keeps a history of events, so you can see how the score changed.

It is a pure calculation. It has no side effects and no dependencies, so it gives the same answer for the same input.
from socrates_maturity import MaturityCalculator

phase_scores = {"discovery": 1.0, "analysis": 0.3}
overall = MaturityCalculator.calculate_overall_maturity(phase_scores)
current_phase = MaturityCalculator.estimate_current_phase(overall)
What the number means, and what it does not
The score is only as good as the scores put into it. Someone, or some model, has to decide that "documentation" is 0.4. The library does not do that judging. It turns judgements into one comparable number and shows which part is weakest. It does not say when to start building: that threshold depends on how costly a mistake is in your project, and I have no data for it.
What I changed in the current product
In the current Socrates app the same idea is used on a smaller scale. The score is not a checklist of categories. It is the share of the ledger entries for one question that the person has settled, in their own words (a term they have defined, a claim they have tested), over the entries under consideration. A question counts as thought through when that share and a few other conditions reach a line. I chose those thresholds as guesses and tuned them on a few simulated conversations. They will change when real beta data arrives.
The library is open source: github.com/Nireus79/Socratic-maturity, and the package is on PyPI.

Want more #ai & #productivity deep dives?

Join 5,951+ developers sharing real benchmarks, code breakdowns, and projects. Free forever.

Part 10 of 11 in Socrates-AI

More Posts

I’m a Senior Dev and I’ve Forgotten How to Think Without a Prompt

Karol Modelski - Mar 19

AAIF Hits 10 Projects. Its CTO on What to Adopt Now and the Security Gaps Left Open

Tom Smithverified - Oct 23

What Security Teams Want From AI Agents Before They Let Them Act Alone

Tom Smithverified - Nov 18

TypeScript Complexity Has Finally Reached the Point of Total Absurdity

Karol Modelski - Apr 23

Your AI Agent Doesn't Need Your Token: Edera's Ariadne Conill on Object Capabilities

Tom Smithverified - Nov 5
chevron_left
1.3k Points • 33 Badges
11Posts
2Comments
5Connections
Building Socrates: Production AI Multi-Agent Platform (37+ modules). Constitutional governance, RAG ... Show more

Commenters (This Week)

4 comments
2 comments
1 comment

Contribute meaningful comments to climb the leaderboard and earn badges!