Cloud Service Model · 3 of 6

SaaS — Software as a Service

Subscribe to finished software running in someone else's cloud. The user-facing layer of cloud computing — Gmail, Salesforce, Slack, Notion, Figma. Probably how you spent today.

SubscriptionMulti-tenantBrowser-firstZero installAPI-friendly
← Back to Cloud
Quick Facts

At a Glance

Basic Concepts

  • Pay subscription — per user, per seat, per usage, or freemium.
  • Multi-tenant — your data lives in a shared system, isolated logically.
  • Vendor owns everything — infrastructure, code, updates, security.
  • Access via browser (and increasingly API + mobile + desktop apps).
  • Updates are continuous — no "version 14 upgrade project."
Categories

The SaaS Universe

CategoryExamples
Productivity / OfficeMicrosoft 365, Google Workspace, Notion, Coda
CommunicationSlack, Teams, Zoom, Discord
CRM / SalesSalesforce, HubSpot, Pipedrive
HR / PayrollWorkday, BambooHR, Gusto, Rippling
Finance / AccountingQuickBooks Online, Xero, NetSuite
ERPNetSuite, SAP S/4HANA Cloud, Oracle Fusion
Design / CreativeFigma, Canva, Adobe Creative Cloud
Dev toolsGitHub, GitLab, Linear, Jira, Sentry, Datadog
E-commerce platformsShopify, BigCommerce
MarketingMailchimp, HubSpot, Marketo, Braze
Customer supportZendesk, Intercom, Front, Help Scout
AI toolsChatGPT, Claude, Cursor, Notion AI
Mechanics

How SaaS Works

Multi-Tenancy Models
  • Pooled (shared DB): all customers in one schema, isolated by tenant_id. Cheapest, smallest customer-by-customer overhead.
  • Silo (DB per tenant): stronger isolation, easier compliance, more ops.
  • Hybrid: shared compute, dedicated databases for big customers.
  • Single-tenant cloud: dedicated stack per customer — for regulated / huge accounts.
Pricing Models
  • Per seat / user — Slack, Notion, GitHub.
  • Tiered features — Free / Pro / Team / Enterprise.
  • Usage-based — Stripe, Twilio, OpenAI (per token, per call).
  • Freemium — free tier funnels to paid.
  • Hybrid — base subscription + overages.
Integration & APIs

Modern SaaS isn't a silo — it's a node in a graph:

  • REST / GraphQL APIs for data access.
  • Webhooks for real-time events.
  • OAuth / SSO for delegated auth.
  • iPaaS tools (Zapier, Make, Workato, MuleSoft, Boomi) wire SaaS together.
  • Reverse ETL (Hightouch, Census) syncs data from warehouse → SaaS.
SaaS Compliance & Trust

Buyers ask about:

  • SOC 2 Type II — table stakes for B2B.
  • ISO 27001 / 27018 — international info-sec.
  • HIPAA / PCI / FedRAMP — domain-specific.
  • GDPR / CCPA / DPDP — data privacy.
  • Subprocessor list, data residency, encryption keys (BYOK).
SaaS Sprawl & Governance

The downside: a typical mid-market company runs 100+ SaaS apps.

  • SSO / SCIM centralize identity & provisioning.
  • SaaS management platforms (Zylo, Productiv, Torii) discover & track usage.
  • Spend management (Vendr, Tropic) negotiates renewals.
Trade-offs

Strengths & Weaknesses

Strengths
  • No installation, no infra, no upgrades.
  • Predictable per-seat pricing.
  • Vendor handles security & uptime.
  • Always the latest version.
Weaknesses
  • Data lives off-prem; vendor lock-in.
  • Per-seat costs add up fast at scale.
  • Limited customization vs self-hosted.
  • Outages = your outage; no fallback.
Building SaaS

If you're the one building a SaaS, the technical picks usually look like:

Continue

Other Service Models