Salesforce platform is opinionated PaaS for CRM and business apps. Heroku/AWS are general-purpose platforms.
Salesforce platform features:
- Pre-built data model (Account, Contact, Opportunity, Case, etc.).
- Multi-tenancy with strict governor limits.
- Declarative tools (Flow, validation rules, etc.).
- Pre-built UI (Lightning Experience, page layouts, dashboards).
- AppExchange ecosystem of pre-built solutions.
- CRM-specific features (forecasting, case routing, marketing).
- Built-in security model (sharing, FLS, profiles, OAuth).
- Salesforce-managed infrastructure — you don't manage servers.
Trade-offs vs generic PaaS:
Salesforce wins for:
- CRM-shaped problems (sales, service, marketing).
- Business apps with rich UI (forms, lists, reports).
- Multi-tenancy with shared schema.
- Quick start (most things "just work").
Generic PaaS wins for:
- Bespoke applications outside CRM.
- Custom data models that don't fit Salesforce.
- Heavy compute (image processing, ML inference).
- Complete control (own database, own runtime, own libraries).
- Lower per-user cost at very high scale.
Hybrid pattern (common):
- Salesforce for the CRM-shaped front-end + business workflows.
- Heroku/AWS for compute-heavy back-end (PDF generation, ML, custom integrations).
- Connected via APIs / Mulesoft.
This is increasingly the pattern: Salesforce as the system of engagement; cloud compute for capabilities Salesforce doesn't natively offer.
Architect insight: don't try to force everything onto Salesforce. The platform's value is for what it does well; for everything else, integrate with cloud-native tools.
Salesforce + AWS / Salesforce + Heroku architectures are common in enterprise. Architects design the boundary deliberately.