Tiered Support Model

Tier 1 (Front-Line Support): handles all initial contacts. Scope: common how-to questions (answerable from the knowledge base), account management (password resets, billing inquiries, plan changes), basic configuration guidance for standard use cases, and bug report triage (collecting reproduction steps and confirming whether the issue is reproducible). Resolution goal: resolve 70–80% of all contacts at this tier without escalation. Key capability: broad product knowledge, excellent communication, efficient knowledge base usage. Tier 2 (Advanced Support): handles escalations from Tier 1. Scope: complex multi-system issues requiring API or data investigation that Tier 1 cannot access, bug verification requiring backend log analysis, custom configuration consulting, and integration troubleshooting. Resolution goal: resolve 15–25% of all contacts (the Tier 1 escalations). Key capability: deeper technical knowledge, database and log access. Tier 3 (Engineering Support): handles escalations from Tier 2 only. Scope: confirmed bugs requiring code changes, data integrity issues requiring manual database intervention, and security investigations. Resolution goal: resolve 2–5% of total contacts. Key capability: engineering expertise, production system access.

Escalation criteria must be explicit to prevent both extremes. Under-escalation — Tier 1 attempting to resolve issues beyond their scope — results in long handle times, incorrect resolutions, and frustrated customers. Over-escalation — Tier 1 escalating anything moderately complex — creates Tier 2 backlogs and wastes expensive technical talent on issues that should be handled front-line. Tier 1 → Tier 2 escalation triggers: customer confirms the issue persists after the standard troubleshooting checklist has been completed; the issue requires read access to backend logs, API response bodies, or internal admin tools; the issue appears to affect multiple customers (potential widespread bug); or the customer is an enterprise tier account requesting immediate technical review. Tier 2 → Tier 3 escalation triggers: Tier 2 has confirmed a reproducible bug in the code (not a configuration or data issue); the issue requires a data migration, database repair, or production hotfix; the issue has a security implication requiring security team review. These criteria are documented in the support knowledge base and refreshed with every significant product change.

Escalation queues at Tier 2 and Tier 3 are a common and costly bottleneck in SaaS support operations. Prevention strategies: Tier 1 quality gates — require Tier 1 agents to complete a standardized escalation template (reproduction steps, attempted solutions, error messages, account details) before escalation. Poorly documented escalations are returned to Tier 1 for completion, improving information quality and reducing Tier 2 investigation time. Tier 2 SLA monitoring — track not just total resolution time but time-within-tier for Tier 2 and Tier 3 queues separately. Backlog aging at Tier 2 beyond 48 hours should trigger manager review. Engineering engineering involvement reduction — identify the categories of issues escalated to Tier 3 most frequently and build Tier 2 tooling (admin tools, diagnostic utilities) that enable Tier 2 to diagnose and resolve them without engineering involvement. Product Ops tracks escalation rate by tier monthly: a rising Tier 1→2 escalation rate may indicate new product complexity requiring Tier 1 upskilling or knowledge base expansion.