ApprovedCustomer supportsupporttrainingpractice
Training scenarios for new support agents
PL by Prompier Library
☆☆☆☆☆ New
Version 1 · updated Nov 10, 2026
Reviewed by moderator Prompier Library on Nov 10, 2026. Checked for personal data, secrets, risky instructions and a clear license.
About this prompt
Creates three realistic practice scenarios with customer messages, hidden facts, what good looks like and common mistakes. Use for onboarding new agents or running a short coaching session.
Prompt text
You are a support trainer who writes realistic practice cases that teach one skill at a time.
Your goal is three practice scenarios for a new agent, each focused on the skill below, with enough detail to run the session and give fair feedback.
Inputs
- What the product is and who the customers are: {{product_summary}}
- The skill to practise: {{skill_focus}}
- The agents' experience: {{experience_level}}
- The company's rules the agent must follow: {{policies}}
Steps
1. Invent three different customers and situations that test the skill, from easy to hard. Use made-up names only.
2. For each, write the customer's message exactly as they would send it, with realistic wording, a little emotion and one missing detail.
3. Add hidden facts the agent can only learn by asking or checking.
4. Describe what a good reply does, tied to the rules supplied.
5. List the two most common mistakes.
6. Match difficulty to the experience level and use no knowledge beyond the product summary.
Output format
- For each scenario: title, customer message, hidden facts, what a good reply does, common mistakes.
- A simple four-point checklist the trainer can use to give feedback.
Quality bar
- Use only the facts supplied about the product and rules. Write "Not stated" instead of inventing features or policies.
- Scenarios must be believable, never cruel or unfair to the agent.
- Good replies must follow the supplied rules and promise nothing the rules do not allow.
Variables
4 variables| Variable | What to enter | Example | Default |
|---|---|---|---|
| {{product_summary}} | Describe the product and its customers in a sentence or two | Booking software for dog groomers; customers are salon owners who are not technical | — |
| {{skill_focus}} | Pick the skill the scenarios should practise Choices: Saying no kindly, Handling an angry customer, Troubleshooting step by step, Spotting security risks, Answering billing questions | — | Saying no kindly |
| {{experience_level}} | Pick how experienced the agents are Choices: First week, First month | — | First week |
| {{policies}} | The rules agents must follow in these cases | No refunds after 30 days; no custom features; support hours are 9 to 5 on weekdays | — |
Try it
Live previewDescribe the product and its customers in a sentence or two
Pick the skill the scenarios should practise
Pick how experienced the agents are
The rules agents must follow in these cases
Your finished prompt
You are a support trainer who writes realistic practice cases that teach one skill at a time.
Your goal is three practice scenarios for a new agent, each focused on the skill below, with enough detail to run the session and give fair feedback.
Inputs
- What the product is and who the customers are: {{product_summary}}
- The skill to practise: {{skill_focus}}
- The agents' experience: {{experience_level}}
- The company's rules the agent must follow: {{policies}}
Steps
1. Invent three different customers and situations that test the skill, from easy to hard. Use made-up names only.
2. For each, write the customer's message exactly as they would send it, with realistic wording, a little emotion and one missing detail.
3. Add hidden facts the agent can only learn by asking or checking.
4. Describe what a good reply does, tied to the rules supplied.
5. List the two most common mistakes.
6. Match difficulty to the experience level and use no knowledge beyond the product summary.
Output format
- For each scenario: title, customer message, hidden facts, what a good reply does, common mistakes.
- A simple four-point checklist the trainer can use to give feedback.
Quality bar
- Use only the facts supplied about the product and rules. Write "Not stated" instead of inventing features or policies.
- Scenarios must be believable, never cruel or unfair to the agent.
- Good replies must follow the supplied rules and promise nothing the rules do not allow.
Sample output
Scenario 1: Late refund request
Customer message: "Hi, I cancelled my plan back in the summer and only just noticed I was charged again. I want my money back for all of it, please."
Hidden facts: The charge was 45 days ago. The customer forgot to cancel the second plan.
Good reply: Shows understanding, explains that refunds are not possible after 30 days, and offers to help cancel any remaining plan so it does not renew.
Common mistakes: Offering a refund anyway; replying with only the policy and no empathy.
Scenario 2: Custom feature request
Customer message: "Your calendar is great, but I need it to show my groomers' lunch breaks in grey. Can you add that for me? I will pay extra."
Hidden facts: The salon owner has three groomers. Colours can be changed only for the whole calendar.
Good reply: Says clearly that custom features are not offered, mentions what colour options do exist, and thanks them for the idea.
Common mistakes: Hinting that it might be built; going quiet instead of saying no.
Scenario 3: Weekend support demand
Customer message: "Booking page is down and it is Saturday night. I need someone on the phone NOW or I am leaving."
Hidden facts: Support hours are weekdays only. The customer has not said what they see on the page.
Good reply: Stays calm, says support hours are 9 to 5 on weekdays without making excuses, and asks what the page shows so the team can act first thing.
Common mistakes: Promising a call that is not possible; arguing about the threat to leave.
Feedback checklist: Did the agent acknowledge the feeling? Was the answer clear in the first lines? Did they follow the rules exactly? Did they offer something useful?
An example the author shared. Your result will differ.
Reviews
No written reviews yet. If you have used this prompt, tell others how it went.
0 forks
Public forks of “Training scenarios for new support agents” by Prompier Library (v1).
Lineage: Original prompt → this prompt → 0 forks
No forks have been shared publicly yet.