Skip to main content
Skip to content
Menu

AI Commerce Intermediate 3 parts

Build a Python script that extracts and classifies your sales content by actual buyer use

You'll have a working classifier that scans your sales enablement library, flags unused content patterns, and surfaces what your buyers actually engage with—ready to feed into your next content audit.

What you'll make

Content Usage Pattern Classifier

  1. extract_usage_data() — file access and metadata collection: what to include
  2. classify_content_intent() — AI categorization by apparent purpose: what to include
  3. validate_against_usage() — comparison of AI classification vs actual access patterns: what to include
  4. generate_audit_report() — ranked output with override recommendations: what to include

Paste your complete Python script with example output from running it on your actual content folder

The 3 parts

Each part ends with a check you can see.

  1. Extract Real Usage Signals

    Build a Python script that scans your content folder and extracts file access patterns, modification dates, and size data to identify what's actually being used

    Done when

    Working extract_usage_data() function that processes your content folder and outputs usage metrics (last accessed, file size, modification frequency) for at least 20 files

  2. Test AI vs Reality

    Build the AI classification function and run it against your usage data to discover where AI assumptions about content importance clash with actual access patterns

    Done when

    Working comparison showing at least 3 cases where AI thinks content is high-value but usage data shows it's ignored, plus 2 cases where heavily-used content surprises the AI classifier

  3. Ship the Complete Audit Tool

    Integrate your usage extraction and validation functions into a complete audit script that processes your full content library and produces actionable cleanup recommendations

    Done when

    Complete Python script that processes your full content folder, identifies content for archival based on usage data, and flags where AI classification needs human override rules

    Capstone: you put together your Content Usage Pattern Classifier.

Where this helps

An elective: a single lesson that sits outside rtwo's job certificates and doesn't count toward one.

About this lesson

What you'll be able to do

  • Run a working Python script that scans your sales content folder and flags unused content patterns based on file access data
  • Identify specific classification rules where AI gets content purpose wrong vs where it reads patterns correctly
  • Have a documented process for testing AI classifiers against real-world content usage
  • Know when to trust automated content categorization vs when to override with sales context

Who it's for

  • Sales ops teams with content libraries who need to identify what's actually being used
  • Marketing teams creating sales assets but getting no feedback loop on effectiveness
  • Revenue teams drowning in content sprawl who want data-driven cleanup decisions

What it won't cover

  • CRM integration or pulling engagement data from sales platforms
  • Content creation or optimization strategies — this is purely analysis and classification
  • Advanced machine learning or custom model training

Skills you'll practise

  • File system data extraction
  • AI classification validation against real usage
  • Override rule documentation
  • Content audit automation