# Create a ramp plan for new support agents

A week-by-week onboarding plan that takes new agents from shadowing to a full queue with clear quality gates.

- Topic: Customer Support
- Framework: Few-Shot
- Level: Standard
- Tags: Customer Support, Plan, Fill-in Template
- Source: https://www.thinkwithprompt.com/prompts/create-a-ramp-plan-for-new-support-agents

## Prompt

```text
Create a ramp-up plan for new customer support agents.

**Product complexity:** [simple / moderate / technical]
**Channels they'll handle:** [email, chat, phone, social]
**Ticket types, from simplest to hardest:** [LIST - e.g. password resets, order status, billing, integrations]
**Tools they need to learn:** [help desk, CRM, billing system, internal tools]
**Time to full productivity we want:** [e.g. 6 weeks]
**Who can support them:** [trainer, buddy, team lead, QA]
**What a fully ramped agent handles:** [TICKETS PER DAY, QUALITY SCORE]

Build:
1. A week-by-week plan: product learning, tools, shadowing, reverse shadowing (they reply, a buddy reviews before sending), nesting (live tickets with close support), then the full queue.
2. Ticket-type unlocks: which categories they handle when, starting with low-risk ones. Keep refunds, security, and escalations until later.
3. Quality gates to pass before each stage (e.g. 5 reviewed tickets at a set QA score), not just time passed.
4. Weekly targets for volume and quality that rise gradually.
5. A daily check-in format for week 1 and a weekly format after that.
6. Signs an agent needs more time and how to support them without labeling it as failure.
7. A short glossary and "where to find answers" list template.

Keep the plan realistic for my team's capacity to support new hires.
```

## What to fill in

- simple / moderate / technical
- email, chat, phone, social
- help desk, CRM, billing system, internal tools
- e.g. 6 weeks
- trainer, buddy, team lead, QA
- TICKETS PER DAY, QUALITY SCORE

Free to copy and use. From ThinkWithPrompt (https://www.thinkwithprompt.com/), a library of 2,663 curated AI prompts.