Claude Code for Business

Duration: 20-25 minutes

Difficulty: Beginner

Prerequisites: Module 0 (Setup complete)

Overview

Learn to interact with Claude Code using natural language. You'll read files, create documents, find information, and explore the FitHub Wellness business scenario.

What you'll learn:

• Launch Claude Code and navigate your workspace

• Read files using natural language

• Create simple files and documents

• Find specific information across files

• Explore the FitHub Wellness scenario

The FitHub Wellness Scenario

Throughout this course, you'll work with a realistic business scenario:

Your Role: New Member Experience Manager at FitHub Wellness, a regional fitness chain with 35 locations.

The Problem:

- FitHub launched "FitRewards" loyalty program 6 months ago

- Invested $40,000 in development

- 11,800 members signed up

- Only 1,950 are active monthly users (16.5% engagement)

- Target engagement: 35-40%

- Leadership is considering shutting it down

Your Mission: Turn this program around in 90 days using Claude Code to analyze the situation and develop a strategy.

Key Commands & Prompts

Launching Claude Code

claude

Reading Files

Ask naturally:

Read the previous manager notes

Show me the SCENARIO file

What's in the brand voice document?

Interactive Lesson

After watching the video, run the hands-on practice:

start-module-1

This interactive lesson will guide you through:

1. Verifying your workspace location

2. Understanding folder structure

3. Reading the FitHub scenario

4. Creating your first file

5. Finding specific metrics

6. Exploring member feedback

7. Free exploration time

What You'll Create

By the end of this module, you'll have:

my-first-notes.md - Your first Claude Code deliverable

Key Takeaways

1. Natural language works - No technical jargon needed

2. Claude can see your files - It reads and creates real documents

3. Ask like you would ask a colleague - Clear, simple questions work best

4. Explore freely - You can't break anything

Self-Assessment

After this module, you should be able to:

[ ] Launch Claude Code and navigate workspace

[ ] Ask Claude to read any file naturally

[ ] Create files with Claude's help

[ ] Find specific information across files

[ ] Understand the FitHub scenario

What's Next

Module 2: File Exploration - Learn to read multiple files simultaneously, search across your project, and create organized summaries.

Duration: 25-30 minutes

Difficulty: Beginner-Intermediate

Prerequisites: Module 1 complete

Overview

Move beyond single files. Learn to read multiple files at once, search across your entire project, and create organized summaries from scattered information.

What you'll learn:

• Read multiple files simultaneously

• Search for information across all files

• Create organized summaries and inventories

• Compare information from different sources

• Identify patterns in scattered data

The Power of Multi-File Analysis

In Module 1, you read one file at a time. Now imagine you have:

- 4 months of customer feedback

- 8 old campaign files

- 5 competitor research documents

- Scattered notes everywhere

Opening each file manually? Hours of work.

With Claude Code? Seconds.

Key Commands & Prompts

Reading Multiple Files

Examples:

Read all customer feedback files and summarize the main themes

Read all competitor research files and create a comparison chart

Read all campaign files and identify which mentioned "points"

Targeted Search

Example:

Find all mentions of "double points" across all files

Creating Organized Summaries

Example:

Create a file called campaign-inventory.md that lists all old campaigns with brief descriptions

Comparative Analysis

Example:

Compare January feedback to April feedback - did sentiment improve or worsen?

Interactive Lesson

Run the hands-on practice:

start-module-2

You'll practice:

1. Multi-file reading across feedback files

2. Searching for specific information

3. Creating a campaign inventory

4. Building a member complaints summary

5. Comparative competitor analysis

Key Takeaways

1. Multi-file reading saves massive time - Analyze months of data in seconds

2. Pattern recognition is powerful - Claude finds themes across sources

3. Organized summaries are easy - Turn chaos into clarity

4. Comparative analysis is strategic - Understand positioning

Real-World Applications

Think about your work:

- Sales: "Read all Q1 call notes and identify top objections"

- Product: "Compare all user feedback and find feature requests"

- Consulting: "Analyze all client documents and create executive summary"

- Marketing: "Review all campaign performance and identify what worked"

Self-Assessment

After this module, you should be able to:

[ ] Read and analyze multiple files in one request

[ ] Search for specific information across entire projects

[ ] Create organized summaries and inventories

[ ] Compare information from different sources

[ ] Identify patterns across scattered data

What's Next

Module 3: Deep Analysis & Synthesis - Move from gathering information to generating strategic insights. Learn root cause analysis, synthesis, and executive-level recommendations.

Duration: 30-35 minutes

Difficulty: Intermediate

Prerequisites: Module 2 complete

Overview

Move from information gathering to strategic insight. Learn to find root causes, connect disparate information, and generate recommendations that could be presented to executives.

What you'll learn:

- Conduct root cause analysis with evidence

- Synthesize information from multiple sources

- Create strategic recommendations

- Perform competitive positioning analysis

- Generate executive-ready deliverables

From Summary to Analysis

There's a critical difference:

"Engagement is low" | "WHY is engagement low?"

"Members are confused" | "What causes their confusion?"

"Campaigns didn't work" | "What made them fail?"

Analysis digs deeper. It asks "why" not just "what."

Key Commands & Prompts

Root Cause Analysis

Example:

Analyze all customer feedback, campaign performance, and previous manager notes. What are the root causes of low engagement?

Cross-Source Synthesis

Example:

How do member complaints relate to campaigns that were run? Are there clear patterns?

Competitive Positioning

Example:

Based on competitor research, what are FitHub's competitive advantages and disadvantages?

Strategic Recommendations

Example:

Create a strategic recommendation for turning around FitRewards. Include problem statement, root causes, strategic approach, and 3-5 specific recommendations. Save as strategic-recommendations.md

Executive Deliverables

Example:

Create a 5-minute presentation outline for FitHub executives covering problem, root causes, solution, and asks

Interactive Lesson

Run the hands-on practice:

start-module-3

You'll practice:

1. Root cause analysis vs symptom identification

2. Evidence-based analysis

3. Connecting campaigns to feedback patterns

4. Competitive positioning analysis

5. Creating strategic recommendations

6. Building executive presentations

What You'll Create

By the end of this module:

- what-worked-analysis.md - Campaign performance patterns

- strategic-recommendations.md - Actionable strategy

- exec-presentation-outline.md - Leadership presentation

The Analysis Mindset

Quality of your questions determines quality of analysis:

Weak Question | Strong Question

"What do the files say?" | "What patterns exist and what do they reveal about root causes?"

"Summarize feedback" | "Analyze feedback for root causes and connect to campaign performance"

"Tell me about competitors" | "What competitive gaps create opportunity for FitHub?"

Think like a strategist:

• Ask "why" not just "what"

• Request evidence, not opinions

• Connect multiple sources

• Generate recommendations, not just summaries

Key Takeaways

1. Analysis asks "why" - Go beyond surface observations

2. Root causes need evidence - Back up claims with specific files

3. Synthesis connects dots - Find hidden relationships

4. Recommendations are actionable - Specific and implementable

5. Executive-ready = concise + evidence-based - Clear and defensible

Self-Assessment

After this module, you should be able to:

[ ] Conduct root cause analysis with evidence

[ ] Synthesize information from multiple sources

[ ] Create strategic recommendations

[ ] Perform competitive analysis

[ ] Generate executive-level deliverables

What's Next

Module 4: Commands & Automation - Turn complex analysis workflows into one-word commands you can reuse.

Duration: 25-30 minutes

Difficulty: Intermediate

Prerequisites: Module 3 complete

Overview

Turn complex workflows into one-word shortcuts. Learn to create custom commands that automate recurring analysis tasks, saving time and ensuring consistency.

What you'll learn:

• Create custom command files

• Build parameterized commands that accept input

• Test and execute commands

• Chain multiple commands together

• Apply the "do it twice, automate it" mindset

The Automation Mindset

Think about recurring work:

• Weekly feedback summaries

• Monthly campaign reviews

• Quarterly competitive analysis

Without automation:

Read all customer feedback files, identify top 3 themes,

extract notable quotes, and create a dated summary

in the analysis folder.

With automation:

weekly-feedback-summary

Same result. 5 seconds instead of typing a paragraph.

Understanding Commands

Commands live in .claude/commands/ folder. Each is a file with instructions Claude follows when you run the command.

Key components:

- What it does - Purpose of the command

- When to use - Scenarios where it's helpful

- Instructions - Step-by-step for Claude

- Output format - What gets created

Key Commands & Prompts

Creating a Basic Command

Example:

Create a command called weekly-feedback-summary that reads all customer feedback files, identifies top 3 themes, lists notable quotes, and creates a dated summary in the analysis folder.

Running a Command

Simply type the command name:

weekly-feedback-summary

Creating Parameterized Commands

Example:

Create a command called analyze-campaign that takes a campaign name, reads that campaign file, finds related member feedback, and creates an analysis report.

Using it:

analyze-campaign double-points-weekend

analyze-campaign tier-name-rebrand

Creating a Competitor Intelligence Command

Example:

Create a command called competitor-intel that reads all competitor research files, creates a comparison matrix, highlights gaps, and identifies opportunities.

Building a Quick Wins Finder

Example:

Create a command called find-quick-wins that analyzes feedback for easy-to-fix issues, prioritizes by frequency and effort, and lists top 5 quick wins.

Creating Documentation

Example:

Create command-reference.md that lists all my commands with descriptions and usage examples.

Interactive Lesson

Run the hands-on practice:

start-module-4

You'll create:

1. A weekly feedback summary command

2. A parameterized campaign analyzer

3. A competitor intelligence command

4. Command documentation

What You'll Create

By the end of this module:

- weekly-feedback-summary command

- analyze-campaign command (parameterized)

- competitor-intel command

- command-reference.md documentation

Command Chaining

Run multiple commands in sequence for powerful workflows:

1. Run feedback-summary

2. Run competitor-intel

3. Ask Claude to synthesize both into an executive brief

The Automation Rule

If you do something twice, automate it.

First time: 2 minutes to set up command

Every time after: 5 seconds to run it

Do that 50 times = hours saved.

Plus: Commands are consistent. Same quality output every time.

Real-World Command Ideas

Product Manager:

- sprint-summary

- feature-request-analysis

- user-feedback-review

Consultant:

- client-status-update

- stakeholder-synthesis

- project-risk-assessment

Marketing:

- campaign-performance-review

- content-calendar-analysis

- channel-effectiveness-report

Key Takeaways

1. Commands automate recurring work - Save time every week

2. One word replaces complex prompts - Fast and consistent

3. Parameterized commands are flexible - Same logic, different inputs

4. Build a library over time - Compound productivity gains

5. Commands ensure consistency - Same quality every time

Self-Assessment

After this module, you should be able to:

[ ] Create custom command files

[ ] Build parameterized commands

[ ] Test commands for reliability

[ ] Chain commands together

[ ] Document your command library

[ ] Apply automation mindset

What's Next

Module 5: Sub-Agents & Parallel Work - Work with multiple specialized AI agents simultaneously for diverse perspectives.

Duration: 30-35 minutes

Difficulty: Intermediate-Advanced

Prerequisites: Module 4 complete

Overview

Go from working with one AI to managing an entire team. Learn to orchestrate specialized AI agents that bring different perspectives, expertise, and priorities to your analysis.

What you'll learn:

• Work with specialized AI agents

• Get multiple perspectives on decisions

• Run parallel analysis across agents

• Create custom agents for your needs

• Orchestrate multi-agent complex projects

The Multi-Perspective Advantage

Consider a decision: Should FitHub simplify points or add earning opportunities?

Product Designer says: "Simplify. Users are confused by complexity."

Trainer Lead says: "Add opportunities. Give trainers engagement tools."

Executive says: "Simplify first, test additions in Phase 2. Lower risk."

Same question. Three different perspectives. All valid. All useful.

Usually this requires meetings with three different people. Now it takes 30 seconds.

Meet Your AI Team

Located in .claude/agents/:

Trainer Lead

- 10+ years gym floor experience

- Focuses on member experience and trainer enablement

- Thinks about practical implementation

Product Designer

- UX and app design expert

- Focuses on user experience and interface

- Thinks about simplicity and usability

Executive

- Business strategist

- Focuses on ROI, metrics, and risk

- Thinks about business impact and timeline

Key Commands & Prompts

Getting Multiple Perspectives

Example:

Get perspectives from the trainer lead, product designer, and exec on whether to simplify the points system or add more earning opportunities.

Agent-Specific Deep Dives

Examples:

Product Designer: Analyze all app-related complaints in customer feedback. What are the UX issues and how would you fix them?

Trainer Lead: Based on member feedback, what do trainers need to better promote FitRewards on the gym floor?

Exec: Analyze the ROI of the FitRewards program and recommend budget allocation.

Parallel Agent Work

Example:

Have the trainer lead, product designer, and exec each analyze the Fitness Personality Quiz concept. Should we pursue it? Each agent should create a recommendation.

Creating Custom Agents

Example:

Create a new agent called Marketing Specialist with expertise in campaign effectiveness, brand messaging, and member acquisition.

Then use it:

Marketing Specialist: Review all old campaigns and recommend which strategies to repeat and which to avoid.

Agent Orchestration

Example:

Create a comprehensive FitRewards improvement plan. Have the trainer lead identify gym floor opportunities, the product designer create UX improvements, the exec build budget and timeline, then synthesize all three into one plan.

Interactive Lesson

Run the hands-on practice:

start-module-5

You'll practice:

1. Getting multi-perspective recommendations

2. Agent-specific deep dive analysis

3. Running parallel analysis

4. Creating a custom agent

5. Full multi-agent project orchestration

What You'll Create

By the end of this module:

- Agent-specific analysis documents

- Custom agent file

- Multi-agent synthesis documents

- comprehensive-improvement-plan.md

Building Agent Teams

For different projects, assemble different teams:

Product Launch:

- Product Manager agent

- Marketing Specialist agent

- Operations Manager agent

Strategy Consulting:

- Financial Analyst agent

- Change Management agent

- Industry Expert agent

Marketing Campaign:

- Creative Director agent

- Data Analyst agent

- Brand Strategist agent

Key Takeaways

1. Different agents = different perspectives - Avoid blind spots

2. Agents have domain expertise - Get specialized insights

3. Parallel work multiplies output - Analyze from multiple angles simultaneously

4. Custom agents for your needs - Build the team you need

5. Orchestration = project management - Coordinate complex work

Self-Assessment

After this module, you should be able to:

[ ] Get multiple perspectives from agents

[ ] Use agents for specialized analysis

[ ] Run parallel analysis across agents

[ ] Create custom agents

[ ] Orchestrate multi-agent projects

[ ] Facilitate agent debates

[ ] Give agents project context

What's Next

Module 6: Project Memory & Long-Term Context - Build persistent memory systems so Claude remembers your work across sessions.

Duration: 30-35 minutes

Difficulty: Intermediate-Advanced

Prerequisites: Module 5 complete

Overview

Solve the "fresh start" problem. Learn to build persistent project memory so Claude remembers your work across sessions, creating true continuity for long-term projects.

What you'll learn:

• Build project memory files that persist

• Create decision logs and knowledge bases

• Maintain living project documentation

• Establish session summary habits

• Apply memory systems to real projects

The Fresh Start Problem

Without project memory:

Day 1: Analyze FitHub, find root causes, make decisions.

Day 2: Start Claude again and ask "What should I focus on?"

Claude: "What project are you working on?"

You have to re-explain everything. Every. Single. Time.

On Day 45 of a 90-day project? That's a lot of context to rebuild.

Solution: Document your work in files Claude can read.

The Memory File System

Core Memory Files

project-brief.md | Living overview - goal, status, metrics, next steps

decision-log.md | What you decided and why (prevents re-debating)

knowledge-base.md | Organized learnings by category

progress-tracker.md | Timeline, milestones, blockers

session-summary-[date].md | What happened each work session

Supporting Files

agent-context.md | Background info for sub-agents

README.md | How the project is organized

Key Commands & Prompts

Creating Project Brief

Create a project-brief.md that includes: project goal, current status, key metrics, major insights discovered, work completed, and immediate next steps.

Starting a New Session

Read my project brief and decision log. What should I focus on today?

Logging Decisions

Add to the decision log: Decided to simplify points system first before adding earning opportunities. Rationale: Product Designer analysis showed complexity is the primary UX barrier. Date this decision.

Building Knowledge Base

Update the knowledge base with today's insight: Members aged 25-34 respond better to gamification features than point accumulation.

Creating Session Summaries

Create a session summary for today's work including: what I analyzed, key findings, decisions made, files created, and next session priorities.

Agent Context

Create an agent-context.md that gives all agents FitHub background, current challenges, decisions made, and work-in-progress.

Using it:

Product Designer: Read agent-context.md first, then recommend next UX priorities.

Interactive Lesson

Run the hands-on practice:

start-module-6

You'll create:

1. Decision log

2. Project brief

3. Knowledge base

4. Agent context document

5. Session summary template

6. Progress tracker

7. Final turnaround plan

Final Challenge: Turnaround Plan

Create fithub-turnaround-plan.md synthesizing all 6 modules:

1. Executive Summary

2. Situation Analysis

3. Root Cause Analysis

4. Strategic Approach

5. Specific Recommendations (5-7)

6. Implementation Timeline (90 days)

7. Success Metrics

8. Resource Requirements

9. Risk Mitigation

10. Appendices

What You'll Create

By the end of this module:

- decision-log.md - All decisions tracked

- project-brief.md - Living overview

- knowledge-base.md - Organized learnings

- agent-context.md - Context for agents

- session-summary-template.md - Reusable template

- progress-tracker.md - 90-day timeline

- final-turnaround-plan.md - Complete strategy

The Memory Habit

Session Start (2 minutes)

1. Ask Claude to read project-brief.md and latest session-summary.md

2. Get oriented on status and priorities

3. Start work with full context

During Session

4. Update decision-log.md when making decisions

5. Add insights to knowledge-base.md as discovered

Session End (3 minutes)

6. Fill out session-summary.md

7. Update progress-tracker.md

8. Update project-brief.md if status changed

5 minutes of memory work saves hours of reconstruction.

Key Takeaways

1. Memory prevents starting over - Instant context every session

2. Project brief = single source of truth - Always current status

3. Decision logs prevent re-debating - Settled questions stay settled

4. Knowledge bases compound - Institutional memory grows

5. Session summaries create history - Never lose progress

6. Context makes agents smarter - Better recommendations

Course Completion!

You've completed the core course. You can now:

• Analyze complex business situations

• Generate strategic recommendations

• Automate recurring workflows

• Manage AI specialist teams

• Build institutional knowledge

All without writing a single line of code.

Self-Assessment

After this module, you should be able to:

[ ] Create comprehensive project briefs

[ ] Maintain decision logs

[ ] Build organized knowledge bases

[ ] Write effective session summaries

[ ] Create agent context documents

[ ] Track project progress

[ ] Maintain memory across sessions

What's Next

Module 7: Creating Custom Skills (Bonus) - Become a power user by learning to create, install, and customize your own Claude Code skills.

Or apply what you've learned to a real project!

Duration: 20-25 minutes

Difficulty: Advanced

Prerequisites: Module 6 complete (Bonus Module)

Overview

Graduate from user to power user. Learn that the skills you've been using are just files—and you can create your own. This module transforms how you think about Claude Code.

What you'll learn:

• How skills work under the hood

• Install skills from the community

• Modify existing skills for your needs

• Create completely new skills from scratch

• Build a personal skills library

The Secret Revealed

Remember those 8 skills you've been using? email-draft, meeting-notes, analyze-folder

They're just files. Markdown files with instructions.

Open .claude/commands/ and look inside any skill file. You'll see:

- What it does

- When to use it

- Instructions for Claude

When you type email-draft, Claude reads this file and follows the instructions.

Which means: You can write your own.

Skill File Structure

Every skill follows this template:

markdown

Skill: [Name]

What This Does

[Brief description of the skill's purpose]

When to Use

[Scenarios where this skill helps]

How It Works

[What output it creates]

Instructions

[Detailed instructions for Claude to follow]

[Output format/template]

[Tone and style guidelines]

Usage Example

[command-name]

[Example input the user might provide]

Key Commands & Prompts

Exploring Existing Skills

Look in .claude/commands/ and open:

- meeting-notes

- analyze-folder

- competitive-analysis

Study their structure.

Installing Community Skills

1. Browse [awesome-claude-skills](https://github.com/ComposioHQ/awesome-claude-skills/tree/master)

2. Find a useful skill

3. Click "Raw" to get clean markdown

4. Create new file in .claude/commands/ with skill name

5. Paste and save

Installing Anthropic Skills

1. Open Terminal in VS Code/Cursor

2. Type “/” and find Plugins

3. Go to Markeplace

4. Add new marketplace

5. Paste anthropic git repository URL https://github.com/anthropics/skills.git

6. Click enter

7. Confirm you have the skills saved by typing /skills in the terminal

Installing full Repo of Skills

1. Browse to the skills you want

2. Copy the Git repository or the installation command of the skill you want

3. Ask Claude Code to install them for you :)

Modifying Existing Skills

Copy and customize:

Example:

Copy email-draft to email-draft-fithub and modify to match FitHub's brand voice, including their typical greeting style, tone preferences, and standard sign-offs.

Creating From Scratch

Example:

Create a skill called weekly-update that helps write status updates for my manager. It should ask for accomplishments, next week's priorities, and blockers, then format in a clean consistent structure.

Interactive Lesson

Run the hands-on practice:

start-module-7

You'll practice:

1. Examining skill file structure

2. Installing a community skill

3. Customizing a skill for FitHub

4. Creating a skill from scratch

5. Building a skills library

What You'll Create

By the end of this module:

- Understanding of skill mechanics

- At least 1 installed community skill

- At least 1 modified/customized skill

- At least 1 skill created from scratch

- Personal skills library foundation

Skills Worth Creating

Communication:

- email-draft-[audience] - Tailored for different recipients

- slack-message - Quick, informal updates

- meeting-agenda - Structured meeting prep

Analysis:

- [industry]-analysis - Industry-specific frameworks

- quarterly-review - Recurring review format

- risk-assessment - Standardized risk evaluation

Documentation:

- project-proposal - New project pitches

- post-mortem - Project retrospectives

- onboarding-doc - New team member guides

Role-Specific:

- PM: sprint-planning, feature-brief, release-notes

- Marketing: campaign-brief, content-calendar, brand-audit

- Sales: deal-summary, objection-handler, proposal-draft

Key Takeaways

1. Skills are just instructions - No magic, just structured text

2. Community has hundreds - Don't reinvent the wheel

3. Customization is powerful - Make skills match YOUR workflow

4. Creation is straightforward - If you can describe it, you can build it

5. Libraries compound value - Build once, use forever

The Power User Mindset

From now on, when you do something twice, ask:

"Should this be a skill?"

If the answer is yes, build it. Your future self will thank you.

Self-Assessment

After this module, you should be able to:

[ ] Explain how skills work

[ ] Install community skills

[ ] Modify existing skills

[ ] Create skills from scratch

[ ] Build a personal skills library

[ ] Calculate time savings from skills

Course Complete!

You've finished Claude Code for Business Professionals.

You can now:

• Explore and analyze complex file structures

• Generate strategic insights and recommendations

• Automate workflows with custom commands

• Orchestrate teams of specialized AI agents

• Build persistent project memory

• Create custom skills for your exact workflow

You're not just using Claude Code. You're making it work the way YOU work.

What's Next

This Week:

• Create 2-3 skills for your most repetitive tasks

• Apply course skills to a real work project

This Month:

• Build skills library with 10+ custom skills

• Install 5+ community skills

• Track time saved

Ongoing:

• Think in skills: "Do it twice, make it a skill"

• Share useful skills with colleagues

• Continuously refine based on actual use

Congratulations!

You've completed the entire course. You now have professional-grade AI productivity skills that most people don't even know exist.

Go build something remarkable.

Bonus Module: Claude Desktop

What You'll Learn

Throughout this course, you've been using Claude Code in the terminal inside VS Code or Cursor. But there's another way to access Claude Code through the Claude Desktop app.

In this bonus video, you'll discover:

• How to connect Claude Code to your project folders directly from the Claude Desktop app (the same app you use for regular Claude conversations)

• The key difference between Claude Desktop and Claude Code: Skills and MCPs are configured separately in each, so you can set up different tools for different purposes

• A practical workflow: Use Claude Desktop for planning and quick conversations, then move to Claude Code in your IDE when you're ready to build

• Live examples of useful MCPs like Firecrawl (for scraping websites) and how they work alongside your skills

Why This Matters

You don't always need the full power of Claude Code in your terminal. Sometimes you just want to chat, brainstorm, or run a quick task. Claude Desktop lets you do that while still accessing your project files and custom tools.

Think of it this way:

• Claude Desktop → Day-to-day conversations, planning, quick tasks

• Claude Code in VS Code/Cursor → Focused project work, building, creating

This module shows you how to get the best of both worlds.