CRM SOFTWARE BUILDING PROJECTS

CRM Software - Building Useful Systems That People Actually Use

Goal: Understand how to design and build CRM systems that deliver real value and achieve high user adoption—learning from the failures of 50-70% of CRM implementations.

The CRM Adoption Crisis You’re Solving

Before diving into projects, understand the problem you’re learning to solve:

Statistic	Source
50-70% of CRM implementations fail	Industry research
#1 cause of failure is poor user adoption	Not technology
17% fail due to lack of integration	Siloed systems
7% fail due to complexity	Too many features

The core insight: CRMs fail not because they lack features, but because they add friction to users’ workflows instead of removing it.

What Makes CRMs Actually Useful

Based on your Siebel/Salesforce experience, you know these systems can be powerful but overwhelming. Here’s what separates successful CRMs:

The Friction Equation

User Adoption = Value Delivered / Effort Required

High Adoption CRMs:

Auto-log emails, calls, meetings (zero data entry)
Surface relevant info at the right time
Integrate seamlessly with existing tools
Role-specific views (not one-size-fits-all)

Low Adoption CRMs:

Require manual data entry for everything
Complex navigation to find information
Duplicate work between CRM and other tools
“Big bang” implementations with all features at once

Project Comparison Table

Project	Difficulty	Time	What You’ll Learn	Business Potential
Contact Management Core	Beginner	1 week	Foundation data model	Service & Support
Sales Pipeline Kanban	Beginner-Intermediate	1 week	Deal stages, conversions	Micro-SaaS
Activity Auto-Logger	Intermediate	2 weeks	Integration, zero-friction UX	Open Core
Email Integration Engine	Intermediate	2 weeks	Real-time sync, threading	Service & Support
Workflow Automation Builder	Advanced	2-3 weeks	Rules engines, triggers	Open Core
Custom Object Platform	Advanced	3-4 weeks	Schema-less design, flexibility	Industry Disruptor
360° Customer View	Advanced	2-3 weeks	Data aggregation, timeline	Open Core
Final: Full CRM Platform	Expert	2-3 months	Complete SaaS architecture	Industry Disruptor

Project 1: Contact Management Core

File: CRM_SOFTWARE_BUILDING_PROJECTS.md
Main Programming Language: Python (FastAPI)
Alternative Programming Languages: Go, TypeScript (Node.js), Ruby on Rails
Coolness Level: Level 2: Practical but Forgettable
Business Potential: 3. The “Service & Support” Model
Difficulty: Level 1: Beginner
Knowledge Area: Data Modeling / CRUD / Relationships
Software or Tool: Contact Database
Main Book: “Designing Data-Intensive Applications” by Martin Kleppmann

What you’ll build: The foundational data layer of any CRM—Contacts, Companies (Accounts), and their relationships. This seems simple but teaches the core patterns every CRM uses.

Why it teaches CRM fundamentals: Every CRM is built on this foundation. You’ll understand:

Why contacts belong to companies (and sometimes multiple companies)
How to handle duplicates (a massive CRM problem)
The difference between “person” and “role at company”
Why data quality determines CRM success

Core challenges you’ll face:

Data model design (contacts, companies, relationships) → maps to entity relationships
Duplicate detection (fuzzy matching on names, emails) → maps to data quality
Merge functionality (combining duplicate records) → maps to data integrity
Search and filtering (find contacts quickly) → maps to indexing strategies
Import/export (CSV, vCard support) → maps to data migration

Key Concepts:

Entity-Relationship Modeling: “Designing Data-Intensive Applications” Chapter 2 - Martin Kleppmann
Full-Text Search: “Designing Data-Intensive Applications” Chapter 3 - Martin Kleppmann
Data Quality Patterns: “Data Quality” by Thomas Redman
REST API Design: “Design and Build Great Web APIs” - Mike Amundsen

Difficulty: Beginner Time estimate: 1 week Prerequisites: Basic web development, SQL basics

Real world outcome:

# Your Contact Management API in action
$ curl -X POST http://localhost:8000/contacts \
  -d '{"first_name": "Alice", "last_name": "Smith", "email": "alice@acme.com"}'
{"id": 1, "first_name": "Alice", "last_name": "Smith", "email": "alice@acme.com"}

$ curl -X POST http://localhost:8000/companies \
  -d '{"name": "Acme Corp", "domain": "acme.com", "industry": "Technology"}'
{"id": 1, "name": "Acme Corp", "domain": "acme.com", "contacts": []}

# Auto-associate by email domain
$ curl http://localhost:8000/contacts/1
{"id": 1, "name": "Alice Smith", "email": "alice@acme.com",
 "company": {"id": 1, "name": "Acme Corp"}, "role": "Unknown"}

# Duplicate detection on import
$ curl -X POST http://localhost:8000/import \
  -F "file=@contacts.csv"
{
  "imported": 847,
  "duplicates_found": 23,
  "duplicates": [
    {"existing": {"name": "Alice Smith"}, "new": {"name": "Alice B. Smith"},
     "confidence": 0.94, "action": "review"}
  ]
}

# Search across all fields
$ curl "http://localhost:8000/search?q=alice+acme"
{"results": [{"type": "contact", "name": "Alice Smith", "company": "Acme Corp"}]}

Learning milestones:

CRUD operations work → You understand basic data operations
Relationships correctly modeled → You understand CRM entity relationships
Duplicates detected on import → You understand data quality challenges
Search returns relevant results → You understand indexing

Project 2: Sales Pipeline Kanban

File: CRM_SOFTWARE_BUILDING_PROJECTS.md
Main Programming Language: TypeScript (React + Node.js)
Alternative Programming Languages: Python (FastAPI + HTMX), Go + Templ, Elixir + LiveView
Coolness Level: Level 3: Genuinely Clever
Business Potential: 2. The “Micro-SaaS / Pro Tool”
Difficulty: Level 2: Intermediate
Knowledge Area: State Management / Real-time Updates
Software or Tool: Pipeline Board (like Pipedrive)
Main Book: “Fundamentals of Software Architecture” by Mark Richards & Neal Ford

What you’ll build: A visual sales pipeline where deals move through stages (Lead → Qualified → Proposal → Negotiation → Closed). This is the “money view” of any CRM.

Why it teaches CRM value: The pipeline view is where salespeople live. You’ll learn:

How to model deal stages and transitions
Why weighted pipeline (probability × value) matters
How to track conversion rates between stages
The importance of “time in stage” metrics

Core challenges you’ll face:

Drag-and-drop with persistence (real-time state sync) → maps to optimistic updates
Stage transition rules (can’t skip stages, required fields) → maps to workflow validation
Pipeline metrics (weighted value, conversion rates) → maps to real-time aggregation
Multi-user collaboration (two people viewing same deal) → maps to concurrency
Deal history (who changed what, when) → maps to audit logging

Key Concepts:

Optimistic UI Updates: “Learning React” - Alex Banks
Event Sourcing for Audit: “Designing Data-Intensive Applications” Chapter 11 - Kleppmann
Real-time Sync: “Building Microservices” Chapter 4 - Sam Newman
State Machines for Workflows: “Domain Modeling Made Functional” - Scott Wlaschin

Difficulty: Beginner-Intermediate Time estimate: 1 week Prerequisites: Project 1, basic frontend knowledge

Real world outcome:

┌─────────────────────────────────────────────────────────────────────────────┐
│  🎯 Sales Pipeline - Q1 2025                      Weighted: $847,500        │
├─────────────┬─────────────┬─────────────┬─────────────┬─────────────────────┤
│   LEAD      │  QUALIFIED  │  PROPOSAL   │ NEGOTIATION │      CLOSED         │
│   $50,000   │   $125,000  │   $200,000  │   $175,000  │      $450,000       │
│   (10%)     │    (25%)    │    (50%)    │    (75%)    │       (100%)        │
├─────────────┼─────────────┼─────────────┼─────────────┼─────────────────────┤
│ ┌─────────┐ │ ┌─────────┐ │ ┌─────────┐ │ ┌─────────┐ │ ┌─────────────────┐ │
│ │Acme Corp│ │ │TechStart│ │ │BigCo Inc│ │ │MegaCorp │ │ │ ✅ DataFlow     │ │
│ │ $25,000 │ │ │ $75,000 │ │ │$150,000 │ │ │$175,000 │ │ │    $200,000     │ │
│ │ 5 days  │ │ │ 12 days │ │ │ 8 days  │ │ │ 3 days  │ │ │ Won: Jan 15     │ │
│ └─────────┘ │ └─────────┘ │ └─────────┘ │ └─────────┘ │ └─────────────────┘ │
│ ┌─────────┐ │ ┌─────────┐ │ ┌─────────┐ │             │ ┌─────────────────┐ │
│ │NewLead  │ │ │CloudCo  │ │ │StartupX │ │             │ │ ✅ WebTech      │ │
│ │ $25,000 │ │ │ $50,000 │ │ │ $50,000 │ │             │ │    $250,000     │ │
│ │ 1 day   │ │ │ 20 days │ │ │ 15 days │ │             │ │ Won: Jan 8      │ │
│ └─────────┘ │ └─────────┘ │ └─────────┘ │             │ └─────────────────┘ │
└─────────────┴─────────────┴─────────────┴─────────────┴─────────────────────┘

Pipeline Analytics:
  • Average days to close: 34 days
  • Lead → Qualified: 45% conversion
  • Qualified → Closed: 28% conversion
  • Deals at risk (stale >30 days): 2 ⚠️

Learning milestones:

Drag-and-drop moves deals → You understand frontend state management
Weighted pipeline calculates correctly → You understand sales metrics
Multiple users see real-time updates → You understand WebSocket sync
Deal history shows all changes → You understand audit trails

Project 3: Activity Auto-Logger (Zero Friction CRM)

File: CRM_SOFTWARE_BUILDING_PROJECTS.md
Main Programming Language: Python
Alternative Programming Languages: Go, TypeScript, Rust
Coolness Level: Level 4: Hardcore Tech Flex
Business Potential: 4. The “Open Core” Infrastructure
Difficulty: Level 2: Intermediate
Knowledge Area: Integration / Background Processing
Software or Tool: Activity Capture System
Main Book: “The Linux Programming Interface” by Michael Kerrisk

What you’ll build: An activity logger that automatically captures emails, calendar events, and calls WITHOUT manual data entry. This solves the #1 CRM adoption killer.

Why it teaches CRM adoption: Manual data entry is why salespeople hate CRMs. You’ll learn:

How to integrate with Gmail/Outlook APIs
How to match activities to contacts/deals automatically
How to run background sync jobs reliably
The difference between “push” and “pull” integrations

Core challenges you’ll face:

OAuth2 integration (Gmail, Outlook, Calendar APIs) → maps to authentication flows
Email parsing (extracting sender, recipients, threads) → maps to data extraction
Contact matching (which contact is this email from?) → maps to entity resolution
Background job processing (sync without blocking UI) → maps to async architecture
Rate limiting and retries (API quotas, failures) → maps to resilient integrations

Key Concepts:

OAuth2 Flows: Google OAuth2 Documentation
Background Jobs: “Designing Data-Intensive Applications” Chapter 10 - Kleppmann
Entity Resolution: “Data Quality” - Thomas Redman
Idempotent Operations: “Release It!” - Michael Nygard

Resources for key challenges:

Difficulty: Intermediate Time estimate: 2 weeks Prerequisites: Projects 1-2, OAuth basics

Real world outcome:

# Connect Gmail account
$ ./crm-sync auth gmail
Opening browser for authentication...
✓ Connected to alice@company.com
✓ Found 3,247 emails in last 90 days
✓ Matched 892 emails to 156 known contacts

# Automatic activity timeline for a contact
$ ./crm-sync timeline "Bob Smith"
Contact: Bob Smith (bob@acme.com)
Company: Acme Corp

Activity Timeline:
─────────────────────────────────────────────────────────
📧 Jan 20, 2025 10:30 AM - Email Received
   Subject: "Re: Proposal Review"
   "Thanks for sending over the updated proposal..."
   [Auto-logged from Gmail]

📅 Jan 18, 2025 2:00 PM - Meeting
   "Product Demo - Acme Corp"
   Duration: 45 minutes
   Attendees: Bob Smith, Carol White
   [Auto-logged from Google Calendar]

📧 Jan 15, 2025 3:45 PM - Email Sent
   Subject: "Proposal for Acme Corp"
   "Please find attached our proposal for..."
   [Auto-logged from Gmail]

📞 Jan 14, 2025 11:00 AM - Call
   Duration: 12 minutes
   [Auto-logged from phone integration]
─────────────────────────────────────────────────────────
Total: 23 activities (last 30 days)
Last contact: 2 days ago

# Real-time sync status
$ ./crm-sync status
Sync Status:
  Gmail (alice@company.com): ✓ Synced 2 min ago
  Calendar: ✓ Synced 5 min ago
  Next sync: in 13 minutes

  Last 24 hours:
    • 47 emails logged
    • 8 calendar events logged
    • 3 new contacts discovered

Learning milestones:

OAuth flow works end-to-end → You understand authentication
Emails match to correct contacts → You understand entity resolution
Background sync runs reliably → You understand job scheduling
Rate limits handled gracefully → You understand resilient integrations

Project 4: Email Integration Engine

File: CRM_SOFTWARE_BUILDING_PROJECTS.md
Main Programming Language: Go
Alternative Programming Languages: Rust, Python, Elixir
Coolness Level: Level 4: Hardcore Tech Flex
Business Potential: 3. The “Service & Support” Model
Difficulty: Level 3: Advanced
Knowledge Area: Email Protocols / Threading
Software or Tool: Email Sync Engine
Main Book: “TCP/IP Illustrated, Volume 1” by W. Richard Stevens

What you’ll build: A bidirectional email sync engine that shows email threads directly in the CRM AND allows sending emails from the CRM that appear in the user’s sent folder.

Why it teaches CRM integration: Email is where sales conversations actually happen. You’ll learn:

IMAP/SMTP protocols at a low level
Email threading (References, In-Reply-To headers)
Bidirectional sync challenges
Why email integration makes or breaks CRM adoption

Core challenges you’ll face:

IMAP sync (efficient incremental updates, IDLE push) → maps to protocol implementation
Email threading (reconstructing conversation from headers) → maps to message correlation
Bidirectional sync (CRM changes reflect in Gmail) → maps to conflict resolution
HTML email rendering (displaying rich content safely) → maps to security sanitization
Attachment handling (storage, preview, size limits) → maps to file management

Key Concepts:

IMAP Protocol: RFC 3501 - Internet Message Access Protocol
Email Threading: “Programming Internet Email” - David Wood
Conflict Resolution: “Designing Data-Intensive Applications” Chapter 5 - Kleppmann
Content Sanitization: OWASP Guidelines

Difficulty: Advanced Time estimate: 2 weeks Prerequisites: Project 3, network programming basics

Real world outcome:

┌────────────────────────────────────────────────────────────────────────────┐
│  📧 Email Thread with Bob Smith (Acme Corp)                    Deal: $150k │
├────────────────────────────────────────────────────────────────────────────┤
│                                                                            │
│  ┌──────────────────────────────────────────────────────────────────────┐  │
│  │ From: Bob Smith <bob@acme.com>                                       │  │
│  │ To: You                                                              │  │
│  │ Date: Jan 20, 2025 10:30 AM                                          │  │
│  │                                                                      │  │
│  │ Thanks for sending over the updated proposal. I've reviewed it with  │  │
│  │ our team and we have a few questions:                                │  │
│  │                                                                      │  │
│  │ 1. Can you clarify the support SLA?                                  │  │
│  │ 2. Is the pricing negotiable for a 3-year commitment?                │  │
│  │                                                                      │  │
│  │ Looking forward to your response.                                    │  │
│  │                                                                      │  │
│  │ - Bob                                                                │  │
│  └──────────────────────────────────────────────────────────────────────┘  │
│                                                                            │
│  ┌──────────────────────────────────────────────────────────────────────┐  │
│  │ [Your previous message - Jan 18, 2025]                              ▼│  │
│  └──────────────────────────────────────────────────────────────────────┘  │
│                                                                            │
│  ┌──────────────────────────────────────────────────────────────────────┐  │
│  │ Reply:                                                               │  │
│  │ ┌──────────────────────────────────────────────────────────────────┐ │  │
│  │ │ Hi Bob,                                                          │ │  │
│  │ │                                                                  │ │  │
│  │ │ Great questions! Let me address them:                            │ │  │
│  │ │ _                                                                │ │  │
│  │ └──────────────────────────────────────────────────────────────────┘ │  │
│  │                                                                      │  │
│  │ [📎 Attach] [📝 Template] [⏰ Send Later]        [Send from Gmail ▼]  │  │
│  └──────────────────────────────────────────────────────────────────────┘  │
│                                                                            │
│  💡 AI Suggestion: Based on similar deals, offer 15% discount for 3-year  │
│     commitment. This has 78% close rate.                                   │
└────────────────────────────────────────────────────────────────────────────┘

# Sending email from CRM appears in Gmail sent folder
✓ Email sent via Gmail API
✓ Logged to deal activity
✓ Thread updated in real-time

Learning milestones:

Emails sync incrementally → You understand IMAP IDLE
Threading reconstructs conversations → You understand email headers
Sent emails appear in Gmail → You understand bidirectional sync
Rich HTML renders safely → You understand content sanitization

Project 5: Workflow Automation Builder

File: CRM_SOFTWARE_BUILDING_PROJECTS.md
Main Programming Language: Python
Alternative Programming Languages: TypeScript, Go, Elixir
Coolness Level: Level 4: Hardcore Tech Flex
Business Potential: 4. The “Open Core” Infrastructure
Difficulty: Level 3: Advanced
Knowledge Area: Rules Engines / Event-Driven Architecture
Software or Tool: Workflow Engine (like Salesforce Flow)
Main Book: “Enterprise Integration Patterns” by Gregor Hohpe

What you’ll build: A visual workflow builder where users can create automations: “When a deal moves to Proposal stage AND value > $50k, assign to senior rep AND send notification.”

Why it teaches CRM power: Automation is where CRM ROI comes from. You’ll learn:

How to design a trigger/condition/action system
Event-driven architecture for real-time response
How to make automation accessible to non-programmers
The dangers of automation loops

Core challenges you’ll face:

Event capture (detecting changes across all entities) → maps to change data capture
Condition evaluation (complex boolean logic on entity fields) → maps to expression engines
Action execution (email, field update, create record) → maps to command pattern
Loop prevention (automation A triggers automation B triggers A) → maps to cycle detection
Visual builder UI (drag-and-drop workflow design) → maps to graph visualization

Key Concepts:

Event Sourcing: “Designing Data-Intensive Applications” Chapter 11 - Kleppmann
Rules Engines: “Enterprise Integration Patterns” - Gregor Hohpe
Change Data Capture: “Designing Data-Intensive Applications” Chapter 11 - Kleppmann
Graph-Based UIs: Study n8n, Zapier, or Salesforce Flow

Resources for key challenges:

n8n (open source) - Study their workflow architecture
Temporal.io - Workflow orchestration patterns

Difficulty: Advanced Time estimate: 2-3 weeks Prerequisites: Projects 1-4, event-driven architecture basics

Real world outcome:

┌────────────────────────────────────────────────────────────────────────────┐
│  ⚙️ Workflow: High-Value Deal Assignment                      [Active ✓]  │
├────────────────────────────────────────────────────────────────────────────┤
│                                                                            │
│  ┌───────────────┐      ┌───────────────────────┐      ┌────────────────┐  │
│  │   TRIGGER     │      │      CONDITIONS       │      │    ACTIONS     │  │
│  │               │      │                       │      │                │  │
│  │ Deal Stage    │─────▶│ Value >= $50,000      │─────▶│ Assign to      │  │
│  │ Changed       │      │ AND                   │      │ "Senior Reps"  │  │
│  │               │      │ Industry = "Finance"  │      │ team           │  │
│  └───────────────┘      └───────────────────────┘      │                │  │
│                                                        │ Send Slack to  │  │
│                                                        │ #big-deals     │  │
│                                                        │                │  │
│                                                        │ Create Task:   │  │
│                                                        │ "Schedule      │  │
│                                                        │  exec meeting" │  │
│                                                        └────────────────┘  │
│                                                                            │
│  Execution History (last 7 days):                                          │
│  ─────────────────────────────────────────────────────────────────────     │
│  ✓ Jan 20 14:32 - Deal "Acme Corp" ($75k) → Assigned to Sarah              │
│  ✓ Jan 18 09:15 - Deal "FinTech Inc" ($120k) → Assigned to Mike            │
│  ⚠️ Jan 17 16:45 - Deal "StartupX" ($55k) → Skipped (Industry: Tech)       │
│  ✓ Jan 15 11:20 - Deal "BankCo" ($200k) → Assigned to Sarah                │
│                                                                            │
│  Stats: 3 executions, 1 skipped, 0 errors                                  │
└────────────────────────────────────────────────────────────────────────────┘

# CLI for workflow management
$ ./crm workflow list
Active Workflows:
  1. High-Value Deal Assignment (triggered 12 times this week)
  2. Stale Deal Alert (triggered 5 times this week)
  3. New Lead Welcome Email (triggered 34 times this week)

$ ./crm workflow run "High-Value Deal Assignment" --dry-run --deal=deal_123
Dry Run Results:
  Trigger: Deal "MegaCorp" stage changed to "Proposal"
  Conditions:
    ✓ Value ($85,000) >= $50,000
    ✓ Industry (Finance) = Finance
  Actions (would execute):
    → Assign deal to team "Senior Reps"
    → Send Slack message to #big-deals
    → Create task "Schedule exec meeting" for owner

Learning milestones:

Triggers fire on entity changes → You understand change detection
Complex conditions evaluate correctly → You understand expression engines
Actions execute reliably → You understand workflow execution
Loops detected and prevented → You understand graph cycle detection

Project 6: Custom Object Platform

File: CRM_SOFTWARE_BUILDING_PROJECTS.md
Main Programming Language: Python + PostgreSQL
Alternative Programming Languages: Go, Elixir, TypeScript
Coolness Level: Level 5: Pure Magic
Business Potential: 5. The “Industry Disruptor”
Difficulty: Level 4: Expert
Knowledge Area: Schema-less Design / Metadata-Driven Systems
Software or Tool: Platform Builder (like Salesforce Platform)
Main Book: “Designing Data-Intensive Applications” by Martin Kleppmann

What you’ll build: A platform that lets users create custom objects, fields, and relationships WITHOUT code changes—like Salesforce’s Custom Objects feature.

Why it teaches CRM extensibility: Every business is different. You’ll learn:

How to store arbitrary schemas in a relational database
The Entity-Attribute-Value (EAV) pattern and its tradeoffs
How Salesforce’s metadata architecture actually works
The balance between flexibility and query performance

Core challenges you’ll face:

Schema-less storage (arbitrary fields per object) → maps to EAV pattern
Dynamic queries (querying custom fields efficiently) → maps to query generation
Relationship types (lookup, master-detail, many-to-many) → maps to metadata modeling
Validation rules (per-field, cross-field) → maps to dynamic validation
UI generation (auto-generate forms from schema) → maps to metadata-driven UI

Key Concepts:

EAV Pattern: “SQL Antipatterns” Chapter 8 - Bill Karwin
Metadata-Driven Systems: Study Salesforce’s architecture docs
Dynamic Schema: “Designing Data-Intensive Applications” Chapter 4 - Kleppmann
Query Optimization for EAV: PostgreSQL JSONB approaches

Resources for key challenges:

Salesforce Metadata API - Study how SF does it
PostgreSQL JSONB - Modern alternative to EAV

Difficulty: Expert Time estimate: 3-4 weeks Prerequisites: All previous projects, advanced SQL

Real world outcome:

# Create a custom object via API
$ curl -X POST http://localhost:8000/api/objects \
  -d '{
    "name": "Project",
    "plural_name": "Projects",
    "fields": [
      {"name": "name", "type": "text", "required": true},
      {"name": "status", "type": "picklist", "options": ["Planning", "Active", "Complete"]},
      {"name": "budget", "type": "currency"},
      {"name": "start_date", "type": "date"},
      {"name": "account", "type": "lookup", "references": "Account"}
    ]
  }'
{"id": "Project__c", "status": "created", "fields_created": 5}

# Query custom object with custom fields
$ curl "http://localhost:8000/api/query" \
  -d '{"object": "Project__c", "fields": ["name", "status", "budget"],
       "where": "budget > 100000 AND status = Active"}'
{
  "records": [
    {"id": "proj_001", "name": "Website Redesign", "status": "Active", "budget": 150000},
    {"id": "proj_002", "name": "Mobile App", "status": "Active", "budget": 200000}
  ],
  "total": 2
}

# Schema introspection
$ ./crm describe Project__c
Object: Project__c (Project)
Created: 2025-01-15
Records: 47

Fields:
  name         Text(255)     Required
  status       Picklist      [Planning, Active, Complete]
  budget       Currency
  start_date   Date
  account      Lookup        → Account

Relationships:
  → Account (lookup via account field)
  ← Task (related list via project__c field)

# Auto-generated admin UI at http://localhost:8000/admin/Project__c

Learning milestones:

Custom objects created dynamically → You understand metadata storage
Custom fields queryable → You understand EAV or JSONB querying
Relationships work across custom objects → You understand dynamic joins
Forms auto-generate from schema → You understand metadata-driven UI

Project 7: 360° Customer View

File: CRM_SOFTWARE_BUILDING_PROJECTS.md
Main Programming Language: TypeScript (React)
Alternative Programming Languages: Python + HTMX, Elixir + LiveView, Go + Templ
Coolness Level: Level 4: Hardcore Tech Flex
Business Potential: 4. The “Open Core” Infrastructure
Difficulty: Level 3: Advanced
Knowledge Area: Data Aggregation / Timeline Design
Software or Tool: Customer 360 Dashboard
Main Book: “Designing Data-Intensive Applications” by Martin Kleppmann

What you’ll build: A unified view of everything about a customer—all contacts, deals, emails, meetings, support tickets, invoices—in a single timeline. This is the “holy grail” of CRM.

Why it teaches CRM value: Fragmented customer data is why CRMs exist. You’ll learn:

How to aggregate data from multiple sources
Timeline design patterns
How to make the “360 view” actually load fast
The challenge of keeping aggregated data fresh

Core challenges you’ll face:

Data aggregation (pulling from contacts, deals, activities, emails) → maps to query optimization
Timeline rendering (chronological, grouped, filtered) → maps to pagination strategies
Real-time updates (new email arrives, timeline updates) → maps to push notifications
Performance (accounts with 10,000+ activities) → maps to caching strategies
Cross-linking (clicking email shows thread, clicking deal shows pipeline) → maps to navigation UX

Key Concepts:

Materialized Views: “Designing Data-Intensive Applications” Chapter 3 - Kleppmann
Pagination Patterns: Cursor-based vs offset pagination
Caching Strategies: “Designing Data-Intensive Applications” Chapter 5 - Kleppmann
Timeline UX: Study HubSpot, Salesforce Lightning

Difficulty: Advanced Time estimate: 2-3 weeks Prerequisites: Projects 1-4, frontend performance optimization

Real world outcome:

┌────────────────────────────────────────────────────────────────────────────────┐
│  🏢 Acme Corporation                                                  ⭐ Key   │
├────────────────────────────────────────────────────────────────────────────────┤
│                                                                                │
│  ┌─────────────────┬─────────────────┬─────────────────┬─────────────────┐    │
│  │   OVERVIEW      │    CONTACTS     │     DEALS       │    ACTIVITY     │    │
│  └─────────────────┴─────────────────┴─────────────────┴─────────────────┘    │
│                                                                                │
│  Industry: Technology        │  Annual Revenue: $50M   │  Employees: 200      │
│  Website: acme.com           │  Customer Since: 2023   │  Health Score: 85/100│
│                                                                                │
│  ┌──────────────────────────────────┐  ┌────────────────────────────────────┐ │
│  │ 👥 Contacts (4)                  │  │ 💰 Deals                           │ │
│  │                                  │  │                                    │ │
│  │ ⭐ Bob Smith (Decision Maker)    │  │ ✅ Platform License    $150,000    │ │
│  │    VP Engineering                │  │    Closed Jan 2024                 │ │
│  │    Last contact: 2 days ago      │  │                                    │ │
│  │                                  │  │ 🔄 Expansion Deal      $75,000     │ │
│  │    Alice Chen (Champion)         │  │    Stage: Proposal                 │ │
│  │    Engineering Manager           │  │    Close: Feb 2025                 │ │
│  │    Last contact: 1 week ago      │  │                                    │ │
│  │                                  │  │ Total Value: $225,000              │ │
│  │ [+2 more contacts]               │  │                                    │ │
│  └──────────────────────────────────┘  └────────────────────────────────────┘ │
│                                                                                │
│  📅 Activity Timeline                                        [Filter ▼] [🔍] │
│  ──────────────────────────────────────────────────────────────────────────── │
│                                                                                │
│  TODAY                                                                         │
│  ├─ 📧 10:30 AM  Email from Bob Smith                                         │
│  │              "Re: Q1 Planning Discussion"                                   │
│  │              "Thanks for the proposal update. I've shared with..."         │
│  │                                                                             │
│  YESTERDAY                                                                     │
│  ├─ 📞 3:15 PM   Call with Alice Chen (12 min)                                │
│  │              Notes: Discussed timeline for expansion. She needs...         │
│  │                                                                             │
│  ├─ 📧 11:00 AM  Email sent to Bob Smith                                      │
│  │              "Q1 Planning Discussion - Updated Proposal"                    │
│  │                                                                             │
│  JAN 18                                                                        │
│  ├─ 📅 2:00 PM   Meeting: Product Demo                                        │
│  │              Attendees: Bob Smith, Alice Chen, Carol White                  │
│  │              [📹 Recording] [📝 Notes]                                      │
│  │                                                                             │
│  ├─ 🎫 9:30 AM   Support Ticket #4521 Opened                                  │
│  │              "API rate limiting issues"                                     │
│  │              Status: Resolved (2 hours)                                     │
│  │                                                                             │
│  [Load more...]                                                                │
│                                                                                │
│  📊 Insights                                                                   │
│  ──────────────────────────────────────────────────────────────────────────── │
│  • 47 touchpoints in last 90 days (above average for this segment)            │
│  • Response time: 2.3 hours average (good)                                     │
│  • Engagement trend: ↗️ Increasing                                            │
│  • Next best action: Schedule QBR (last one: 45 days ago)                     │
└────────────────────────────────────────────────────────────────────────────────┘

Learning milestones:

Timeline aggregates all data sources → You understand data aggregation
Infinite scroll loads smoothly → You understand pagination
New activities appear in real-time → You understand push updates
Large accounts load in <2 seconds → You understand caching

Final Project: Full CRM Platform

File: CRM_SOFTWARE_BUILDING_PROJECTS.md
Main Programming Language: Go (backend) + TypeScript (frontend)
Alternative Programming Languages: Rust + React, Elixir + LiveView, Python + HTMX
Coolness Level: Level 5: Pure Magic
Business Potential: 5. The “Industry Disruptor”
Difficulty: Level 5: Master
Knowledge Area: Full-Stack Platform Engineering
Software or Tool: Complete CRM Platform
Main Book: “Fundamentals of Software Architecture” + “Designing Data-Intensive Applications”

What you’ll build: A production-grade CRM platform combining everything: multi-tenant architecture, custom objects, workflow automation, integrations, API, and AI-powered features.

Why it teaches CRM at the highest level: You’re building what Salesforce built. You’ll learn:

Multi-tenant SaaS architecture
API-first design
AI integration (lead scoring, next-best-action)
The complete CRM technology stack

Core challenges you’ll face:

Multi-tenancy (data isolation, per-tenant customization) → maps to SaaS architecture
API design (REST + GraphQL, rate limiting, versioning) → maps to API platforms
AI integration (lead scoring, email suggestions, forecasting) → maps to ML in production
Performance at scale (1M+ records, 1000+ users) → maps to distributed systems
Security (RBAC, field-level security, audit logs) → maps to enterprise security
Billing/licensing (feature gating, usage tracking) → maps to SaaS monetization

Key Concepts:

Multi-Tenant Architecture: “Building Multi-Tenant SaaS Applications” - Tod Golding
API Design: “Design and Build Great Web APIs” - Mike Amundsen
ML in Production: “Designing Machine Learning Systems” - Chip Huyen
Distributed Systems: “Designing Data-Intensive Applications” - Kleppmann

Difficulty: Master Time estimate: 2-3 months Prerequisites: All previous projects

Real world outcome:

# Multi-tenant API
$ curl -H "X-Tenant: acme" https://api.yourcrm.com/v1/contacts
{"data": [...], "meta": {"tenant": "acme", "total": 1247}}

# AI-powered lead scoring
$ curl https://api.yourcrm.com/v1/leads/lead_123/score
{
  "lead_id": "lead_123",
  "score": 87,
  "factors": [
    {"factor": "engagement_velocity", "impact": "+15", "detail": "5 emails in 3 days"},
    {"factor": "company_fit", "impact": "+25", "detail": "Matches ICP: Tech, 100-500 emp"},
    {"factor": "title_match", "impact": "+20", "detail": "VP title = decision maker"},
    {"factor": "competitor_tech", "impact": "-8", "detail": "Uses competitor product"}
  ],
  "recommendation": "High priority - schedule demo within 48 hours"
}

# Natural language query (AI-powered)
$ curl -X POST https://api.yourcrm.com/v1/ask \
  -d '{"query": "Show me deals over $50k that have been stuck in proposal for more than 2 weeks"}'
{
  "interpretation": "Deals WHERE stage = 'Proposal' AND value >= 50000 AND stage_age > 14 days",
  "results": [
    {"name": "Acme Expansion", "value": 75000, "stage_age": "18 days", "owner": "Sarah"},
    {"name": "TechCorp License", "value": 120000, "stage_age": "22 days", "owner": "Mike"}
  ],
  "insight": "These 2 deals represent $195k at risk. Average close rate for deals stale >14 days is 23% vs 67% baseline."
}

# Webhook for integrations
$ curl https://api.yourcrm.com/v1/webhooks \
  -d '{"event": "deal.stage_changed", "url": "https://yourapp.com/webhook", "filters": {"stage": "Closed Won"}}'
{"webhook_id": "wh_123", "status": "active"}

# Admin dashboard
https://yourcrm.com/admin
┌─────────────────────────────────────────────────────────────────────────┐
│  YourCRM Admin                                                          │
├─────────────────────────────────────────────────────────────────────────┤
│  Tenants: 47 active                                                     │
│  Total Users: 1,247                                                     │
│  API Calls (24h): 2.3M                                                  │
│  Avg Response Time: 45ms                                                │
│                                                                         │
│  Health: ✓ All systems operational                                     │
└─────────────────────────────────────────────────────────────────────────┘

Learning milestones:

Multi-tenant data isolation works → You understand SaaS architecture
API handles 1000 req/sec → You understand performance engineering
AI suggestions are accurate → You understand ML integration
Custom objects work across tenants → You understand platform engineering

Recommended Learning Path

Given your Siebel experience and Salesforce learning, here’s my recommendation:

Phase 1: Foundation (2-3 weeks)

Project 1 (Contact Core) → Project 2 (Pipeline)

You know these concepts from Siebel—now build them yourself to understand the implementation.

Phase 2: The Adoption Secret (3-4 weeks)

Project 3 (Auto-Logger) → Project 4 (Email Integration)

This is what makes modern CRMs successful. Zero-friction data capture is the difference between 30% and 90% adoption.

Phase 3: Power Features (4-5 weeks)

Project 5 (Workflows) → Project 7 (360° View)

These are the features that justify CRM investment—automation and unified customer intelligence.

Phase 4: Platform (3-4 weeks)

Project 6 (Custom Objects)

This is what makes Salesforce a platform not just an app. Understanding this changes how you think about CRM architecture.

Phase 5: Production (2-3 months)

Final Project (Full CRM)

Combine everything into a production-grade SaaS.

Key Insight: Why CRMs Fail

After building these projects, you’ll understand the core truth:

CRM Success = (Data Quality × Automation) / User Effort

Siebel/Traditional CRMs: High effort, moderate automation, poor data quality → Low adoption

Modern CRMs (HubSpot, Pipedrive): Low effort (auto-capture), high automation, good data → High adoption

Your Goal: Build CRMs that capture data automatically, automate repetitive tasks, and deliver insights—requiring minimal user effort.

Essential Resources

Books

“Designing Data-Intensive Applications” by Martin Kleppmann
- Foundation for everything: data modeling, caching, scaling, event sourcing
“Fundamentals of Software Architecture” by Mark Richards & Neal Ford
- Architecture patterns for building platforms
“Enterprise Integration Patterns” by Gregor Hohpe
- Essential for workflow automation and system integration
“Domain-Driven Design” by Eric Evans
- Modeling the CRM domain properly

Online Resources

Salesforce Architecture Documentation - Study how SF does it
HubSpot Developer Docs - Modern CRM API design
n8n (open source) - Workflow automation patterns

CRM Software - Building Useful Systems That People Actually Use

The CRM Adoption Crisis You’re Solving

What Makes CRMs Actually Useful

The Friction Equation

Project Comparison Table

Project 1: Contact Management Core

Project 2: Sales Pipeline Kanban

Project 3: Activity Auto-Logger (Zero Friction CRM)

Project 4: Email Integration Engine

Project 5: Workflow Automation Builder

Project 6: Custom Object Platform

Project 7: 360° Customer View

Final Project: Full CRM Platform

Recommended Learning Path

Phase 1: Foundation (2-3 weeks)

Phase 2: The Adoption Secret (3-4 weeks)

Phase 3: Power Features (4-5 weeks)

Phase 4: Platform (3-4 weeks)

Phase 5: Production (2-3 months)

Key Insight: Why CRMs Fail

Essential Resources

Books

Online Resources

Sources