Services-as-Software

The new business model unlocked by Business-as-Code - AI-delivered services that operate autonomously

The same way Infrastructure-as-Code unlocked Software-as-a-Service in the Cloud era, Business-as-Code unlocks AI-delivered Services-as-Software in the AI era.

The Paradigm Shift

Then: Cloud Era (2006-2024)

Infrastructure-as-Code (Terraform, CloudFormation, Kubernetes) enabled developers to:

Define infrastructure declaratively
Version control infrastructure
Deploy infrastructure automatically
Scale infrastructure elastically

This unlocked Software-as-a-Service businesses:

Salesforce: CRM delivered remotely
Stripe: Payments infrastructure as a service
Shopify: E-commerce platform as a service
Twilio: Communications as a service

Business Model: Deliver software applications remotely. Customers still need humans to operate the software.

Now: AI Era (2024+)

Business-as-Code ($.Subject.predicate.Object patterns, semantic graphs, AI agents) enables developers to:

Define businesses declaratively
Version control business logic
Deploy businesses automatically
Scale operations autonomously

This unlocks AI-delivered Services-as-Software:

Not software to operate → Services that execute autonomously
Not features to configure → Outcomes delivered automatically
Not licenses to buy → Results to consume
Not humans required → AI agents operate 24/7

Business Model: Deliver complete services autonomously. No human operators required.

What Changed?

From Software to Services

Traditional SaaS (2006-2024):

You buy: Software application
You operate: Configure, train staff, monitor, maintain
You get: A tool to use

Services-as-Software (2024+):

You buy: Service subscription
You operate: Nothing - service operates itself
You get: Outcomes delivered autonomously

Why Now?

Three convergences make Services-as-Software possible today:

AI Models Reached Human-Level Capability
- GPT-5, Claude Sonnet 4.5, Llama 4, Gemini 2.5 Pro
- Reasoning, planning, tool use, memory
- Multi-modal understanding (text, images, code)
Semantic Standards Matured
- Schema.org: 817 types, 1,518 properties
- GS1: Supply chain and product standards
- O*NET: 1,459 occupations, 17,588 tasks
Infrastructure Became Elastic
- Cloudflare Workers: Global, instant scale
- Edge computing: 300+ locations
- Serverless: Pay-per-execution pricing

Result: You can now build services that AI agents operate at global scale with zero human intervention.

graph TB SaaS[Services-as-Software Platform] SaaS --> Build[Build Services] SaaS --> Deploy[Deploy Globally] SaaS --> Monetize[Monetize & Bill] SaaS --> Operate[Operate Autonomously] Build --> B1[AI Integration] Build --> B2[Event Handlers] Build --> B3[Workflows] Deploy --> D1[300+ Edge Locations] Deploy --> D2[Auto-Scale] Deploy --> D3[Zero Downtime] Monetize --> M1[Subscription] Monetize --> M2[Usage-Based] Monetize --> M3[Outcome-Based] Operate --> O1[AI Agents 24/7] Operate --> O2[Self-Healing] Operate --> O3[Continuous Learning]

Create services that:

Combine AI, workflows, and business logic
Monetize through subscriptions or usage-based pricing
Scale automatically with demand
Integrate seamlessly with existing systems
Deliver measurable business outcomes autonomously

Key Capabilities

AI-Native Services: Built on foundation models (GPT-5, Claude Sonnet 4.5, Llama 4, Grok 4, Gemini 2.5 Pro) that provide intelligence, understanding, and generation capabilities.

Event-Driven Architecture: Services react to events ($.Event.created, $.Order.placed) and trigger actions across your business.

Semantic Integration: Using $.Subject.predicate.Object patterns, services integrate naturally with your existing data model and business processes.

Automatic Scaling: Built on Cloudflare Workers, services scale from zero to millions of requests without infrastructure management.

Built-in Monetization: Subscription, usage-based, and hybrid pricing models with automatic billing and payment processing.

What is a Service?

A Service is a packaged solution that delivers specific business outcomes:

Customer support automation
Content generation
Data analysis
Process automation
Integration services
Specialized AI agents

Creating Services

Define services using code:

import { service } from 'sdk.do'

export default service({
  name: 'SEO Content Generator',
  description: 'AI-powered SEO content creation',

  // Define pricing
  pricing: {
    model: 'subscription',
    plans: [
      { name: 'Starter', price: 29, features: ['10 articles/mo'] },
      { name: 'Pro', price: 99, features: ['50 articles/mo', 'priority'] },
      { name: 'Enterprise', price: 299, features: ['unlimited', 'custom'] },
    ],
  },

  // Define capabilities
  capabilities: {
    async generateArticle({ topic, keywords, length }) {
      // Use AI to generate SEO-optimized content
      const article = await $.ai.generateText({
        prompt: `Write an SEO article about ${topic}`,
        context: { keywords, length },
      })

      // Analyze and optimize
      const optimized = await $.ai.optimize(article, {
        for: 'seo',
        keywords,
      })

      return optimized
    },
  },
})

Monetization

Subscription Model

pricing: {
  model: 'subscription',
  plans: [
    { name: 'Basic', price: 19, interval: 'month' },
    { name: 'Pro', price: 199, interval: 'year' }
  ]
}

Usage-Based Model

pricing: {
  model: 'usage',
  tiers: [
    { up_to: 1000, price_per_unit: 0.10 },
    { up_to: 10000, price_per_unit: 0.08 },
    { above: 10000, price_per_unit: 0.05 }
  ]
}

Hybrid Model

pricing: {
  model: 'hybrid',
  subscription: { base: 49, includes: 1000 },
  overage: { price_per_unit: 0.05 }
}

Service Marketplace

List your services in the .do marketplace:

Automatic discovery
User reviews and ratings
Built-in billing
Usage analytics
Customer management

Integration

Services integrate seamlessly:

// Install a service
const seo = await $.services.install('seo-content-generator')

// Use the service
const article = await seo.generateArticle({
  topic: 'AI in Healthcare',
  keywords: ['AI', 'healthcare', 'diagnosis'],
  length: 1500,
})

Service Templates

Start with pre-built templates:

Customer Support Agent
Content Marketing Suite
Data Analysis Service
Email Marketing Automation
Social Media Manager
Sales Assistant

Managing Services

Dashboard

Monitor service performance:

Active users
Usage metrics
Revenue
Error rates
Customer feedback

Scaling

Automatic scaling based on:

Request volume
Response time
Resource usage
Cost optimization

Versioning

Manage service versions:

Semantic versioning
Gradual rollouts
A/B testing
Rollback support

Service Composition

Combine multiple services:

const workflow = $.workflow('content-pipeline', async (topic) => {
  // Use SEO service
  const article = await services.seo.generateArticle({ topic })

  // Use image service
  const images = await services.images.generate({
    descriptions: extractImageNeeds(article),
  })

  // Use publishing service
  await services.cms.publish({
    content: article,
    images,
    schedule: 'next_available',
  })
})

Analytics

Built-in analytics track:

Usage patterns
User engagement
Revenue metrics
Performance data
Error rates
Customer satisfaction

Revenue Model: Per-seat licenses or subscriptions
Cost Structure: Software development + customer support + infrastructure
Margin Constraint: Human support team doesn't scale linearly

Example: Zendesk charges $19-99/agent/month. Customer needs humans to use the software.