70-20-10 Learning Activity Logger: See the Development Courses Miss

The problem this kills

You already know the hard truth of learning and development: most growth doesn't happen in a classroom. It happens on a stretch project, in a tough piece of feedback, in shadowing someone who's better at the thing. The 70-20-10 model puts numbers on it — roughly 70% of development from on-the-job experience, 20% from coaching and others, 10% from formal courses.

But what does your system of record actually track? Course completions. The 10%. So when someone asks "are we developing our people?" all you can show is a list of finished e-learnings — which makes it look like the only learning that counts is the kind that's easiest to log. Meanwhile the project that stretched someone for three months, the coaching their manager gave every week, the mentoring across teams — none of it shows up anywhere, so none of it gets credit, and nobody can see who's getting a balanced diet of development versus who has done nothing but click through compliance modules.

You don't need a six-figure LMS module to fix this. You can build a simple, trustworthy logger this afternoon — without being a developer.

What you'll build

A clean web app where:

• An employee logs a learning activity in seconds: what type it was (on-the-job, coaching/feedback, or formal course), a short description, how many hours, the date, and which development goal or skill it relates to.
• The tool automatically sorts it into 70 / 20 / 10 and ties it to a real goal — so logging is purposeful, not box-ticking.
• The person's manager reviews and approves each activity — confirming it actually happened and the category and hours are reasonable — before it counts toward their development record.
• Everyone (and their manager, and L&D) can see each person's development mix across the three categories, with a flag when someone is all formal courses and no on-the-job growth.
• You can export the whole validated learning log as a clean CSV for your LMS, your talent reviews, or your board deck.

What's inside the Implementation Plan

The plan is a single file you paste into an AI coding agent, which then builds the tool with you step by step. Crucially, it opens by interviewing you about your organization — your job titles and team structure, how you name goals and skills, your real category definitions, your typical and peak logging volumes, your approval rules, and the messy edge cases (a course someone started but didn't finish, a project that spanned two quarters, an activity logged for the wrong person). It reflects a short tailored spec back to you and waits for your thumbs-up. So you get a tool shaped around how your L&D function actually works — not a generic template you have to bend yourself to fit.

From there it walks through: setting up the database and login, building the logging form, the manager approval queue, the per-person 70-20-10 dashboard with the "all formal" flag, the email notifications, and the CSV export — each step ending with a ready-to-paste prompt.

The governance it includes (this is the point)

This is built to be trusted with people's development records, so the plan bakes in:

• Login so only your team can use it.
• Row-level security so people see their own activities and managers see their own team's — never the whole company's.
• A complete audit trail — who logged what, who approved it, and exactly when.
• A human-in-the-loop approval gate — nothing counts toward a development record until the right manager has reviewed and approved it. The AI helps categorize; a person validates.
• Duplicate guards so the same activity (same person + activity + date) can't be logged or approved twice.

Who it's for

L&D teams, people partners, and managers who know in their gut that most real development happens on the job — but whose systems only ever count formal courses, leaving them unable to show, encourage, or balance the other 90%.

You've got this. Paste the first prompt and let the agent interview you — by this evening you'll have a learning logger that finally sees the whole picture.