15 Customer Service Scenarios for Role Play - Ecommerce Scripts for WhatsApp, Chat and Instagram

Customer service role play scenarios are structured practice exercises where one person plays the customer and another plays the support agent, acting out realistic support conversations before they happen in real life.

Role play builds the skills that matter most in customer service: staying calm under pressure, finding the right words when a customer is upset, and knowing exactly what to do when a situation falls outside the standard process. Companies that invest in scenario-based training see measurably better outcomes — a Zendesk study found organisations using scenario-based training achieved a 16 percent increase in customer satisfaction scores and a 22 percent reduction in onboarding time for new agents.

The 15 scenarios in this guide are written specifically for ecommerce stores. Every script is formatted for the channels modern ecommerce agents actually use — WhatsApp, website chat, and Instagram DM — not for phone calls that represent a shrinking share of ecommerce support volume.

The 15 scenarios covered are: order not arrived, wrong item received, damaged product, return request within policy, return request outside policy, refund not received, discount code not working, angry customer, customer threatening a chargeback, customer asking for a manager, product quality complaint, delivery attempted when not home, order partially missing, pre-sale product question, and the customer who wants an exception.

How to run a role play session for ecommerce support

Role play for ecommerce support is slightly different from traditional call centre training because most conversations happen in writing, not on a call. This changes what agents practise.

Written support on WhatsApp and chat requires a different set of skills than a phone call. Agents must communicate empathy without tone of voice. They must be concise enough to be readable on a phone screen. They must know when a written conversation needs to escalate to a call rather than continuing in text.

Set up each session with two participants — one playing the customer and one playing the agent. The customer role player should stay in character for the full scenario and not immediately accept the first response if it does not fully resolve the situation. Real customers rarely do.

Use the actual tools your team uses. If agents handle WhatsApp on your shared inbox, run the role play on that interface. If they work on a website chat platform, use that. The interface shapes how agents think about response length, formatting, and urgency.

After each scenario, discuss three things: what the agent did well, what they could improve, and what the customer was feeling at each point in the conversation. The customer perspective discussion is the most valuable part of any role play debrief and the one most sessions skip.

Difficulty rating - which scenarios to practise first

Start new agents on beginner scenarios until they are consistently confident. Move to intermediate once the fundamentals — acknowledging, investigating, resolving — feel natural. Reserve advanced scenarios for when agents are comfortable handling pressure without retreating into scripts.

The 15 role play scenarios with scripts

Each scenario below includes the channel context, the customer's opening message, the agent's response, and the key skill being practised. Scenarios are written in the short, conversational style appropriate for WhatsApp and chat. Where a response would differ meaningfully between email and chat, both versions are provided.

Scenario 1 - Order not arrived (tracking available)

Difficulty: Beginner

Channel: WhatsApp or website chat

Skill being practised: Retrieving live order data and communicating it clearly without overexplaining

Customer message: "Hi, my order should have arrived by yesterday. Where is it?"

Agent response: "Hi, thanks for reaching out. I am sorry to hear it has not arrived yet. I have just checked your order and the latest update from the carrier shows your parcel is currently out for delivery today, estimated by end of day. If it does not arrive by tonight, please message us again and we will take this further immediately. Can I help with anything else?"

What the agent is practising: Finding order data quickly, communicating it confidently, and setting a clear next step without leaving the customer in uncertainty.

Common mistake to watch for: Vague answers like "it should arrive soon" without actual tracking data. Customers on WhatsApp want a specific answer, not reassurance.

Role play escalation: The trainer playing the customer replies: "The tracking has said out for delivery for two days. I do not believe it." The agent must now acknowledge that the customer's frustration is reasonable and escalate to a carrier investigation without making a promise they cannot keep.

Scenario 2 - Discount code not working

Difficulty: Beginner

Channel: Website chat

Skill being practised: Troubleshooting without frustrating the customer further

Customer message: "Your discount code is not working. I am trying to checkout with WELCOME10 and it keeps saying invalid."

Agent response: "I am sorry about that, let me sort this for you right away. WELCOME10 is valid for orders over £30 on full-price items and excludes sale products — can you confirm your cart total and whether any items are from the sale section? If everything qualifies and it is still not applying, I will generate a working code for you directly."

What the agent is practising: Checking eligibility conditions without making the customer feel accused of doing something wrong, while moving toward a resolution rather than just explaining why the code does not work.

Common mistake to watch for: Telling the customer "the code is working fine" without checking their specific situation. This dismisses their experience and increases frustration.

Role play escalation: The trainer confirms the order qualifies but the code still will not apply. The agent must now take ownership and generate a manual discount or apply it at the backend — not send the customer to another department.

Scenario 3 - Return request within policy

Difficulty: Beginner

Channel: WhatsApp

Skill being practised: Processing a return request clearly and efficiently

Customer message: "I want to return my order. I received it last week and I just do not like it."

Agent response: "Of course, I can help with that. Your order is within our return window so this is completely fine. Here is how to return it: [return instructions link]. Once we receive the item back, your refund will be processed within five working days to your original payment method. Do you need a returns label or do you have your own packaging sorted?"

What the agent is practising: Processing a legitimate request efficiently without making the customer justify their decision or feel judged for wanting to return.

Common mistake to watch for: Adding unnecessary friction — asking why they want to return, suggesting they keep it, or introducing conditions not in the policy. The customer has a right to return. The agent's job is to make that easy.

Scenario 4 - Refund not received

Difficulty: Intermediate

Channel: Email or website chat

Skill being practised: Managing expectation around financial timelines without being defensive

Customer message: "I returned my order two weeks ago and I have still not received my refund. This is not acceptable."

Agent response: "I completely understand your frustration and I am sorry this has taken longer than it should. I have checked your account and I can see your return was received on [date] and the refund was issued on [date]. Refunds typically take three to five working days to appear depending on your bank. If it has not shown by [specific date], please come back to us and we will escalate directly with our payments team. I can also provide a reference number for your bank if that would help."

What the agent is practising: Acknowledging frustration genuinely, sharing specific dates rather than vague timelines, and giving the customer a concrete next action without dismissing their concern.

Common mistake to watch for: "We already issued the refund" as a complete response. This may be technically accurate but does not address the customer's experience. Agents must bridge the gap between what happened operationally and what the customer is feeling.

Scenario 5 - Wrong item received

Difficulty: Intermediate

Channel: WhatsApp or Instagram DM

Skill being practised: Taking ownership of a clear error without delay

Customer message: "I ordered [item A] and received [item B]. This is wrong."

Agent response: "I am really sorry about that — that is our mistake and I want to fix it for you straight away. I have checked your order and you are absolutely right, you ordered [item A]. I am going to arrange the correct item to be sent to you today and organise collection of the wrong one at no inconvenience to you. Can you confirm your delivery address is still [address from order]? I want to get this moving immediately."

What the agent is practising: Owning the error immediately without hedging, moving to resolution in the same message, and avoiding the common trap of asking the customer to prove what they received.

Common mistake to watch for: "Can you send us a photo?" as the first response to a wrong item report. In most cases, this adds unnecessary friction for a clear operational error. Agents should lead with ownership and only request evidence if the situation genuinely requires verification.

Role play escalation: The item the customer should have received is now out of stock. The agent must offer an alternative or full refund without making the customer feel like a problem.

Scenario 6 - Damaged product

Difficulty: Advanced

Channel: Instagram DM or whatsapp chat

Skill being practised: High-empathy response to an emotionally significant failure

Customer message: "My order arrived completely smashed. I ordered this as a birthday gift and now it is ruined."

Agent response: "I am so sorry. Receiving a damaged item is disappointing enough, but when it was meant to be a gift, that makes it so much worse and I completely understand how frustrating this must be. This is not the standard we hold ourselves to. I am going to get a replacement sent to you on priority shipping — can you let me know when you need it by and I will do everything I can to make sure it arrives in time? Please do not worry about returning the damaged one."

What the agent is practising: Responding to the emotional context, not just the operational problem. The customer's frustration is about the gift, not just the product. The agent must acknowledge that specific loss before moving to the resolution.

Common mistake to watch for: Jumping straight to "send us a photo and we will look into it." An obviously damaged item that ruined a gift requires immediate empathy and ownership, not an investigation process. Save the investigation for fraud indicators, not for clearly genuine damage reports.

Scenario 7 - Return request outside policy

Difficulty: Intermediate

Channel: Email or website chat

Skill being practised: Communicating a policy boundary firmly and empathetically

Customer message: "I want to return this. I know it has been a while but I am not happy with it."

Agent response: "Thank you for getting in touch. I have checked your order and I can see it was placed on [date], which means it falls outside our standard [X]-day return window. I want to handle this fairly for you, so I am passing this to a member of our team who will be in touch within 24 hours to discuss the options available to you. I want to make sure you feel looked after even in a situation like this."

What the agent is practising: Not making the final decision themselves, not refusing outright, and not creating false hope. The escalation maintains goodwill while giving a senior team member the authority to make an exception decision.

Common mistake to watch for: A flat "our policy does not cover this" response with no next step. This ends the interaction without any path forward and creates a customer who has no option but to dispute the charge or leave a negative review.

Scenario 8 - Angry customer

Difficulty: Intermediate

Channel: Website chat or WhatsApp

Skill being practised: De-escalation without matching the customer's emotional temperature

Customer message: "This is absolutely ridiculous. I have been waiting three weeks and nobody has helped me. Your company is a joke."

Agent response: "I hear you and I am genuinely sorry. Three weeks without a resolution is completely unacceptable and your frustration makes total sense. I am looking at your case right now and I am going to stay with you until this is sorted. Can you give me your order number so I can see exactly what has happened and fix it?"

What the agent is practising: Absorbing the emotional energy without becoming defensive, not apologising for things they cannot control, and moving toward resolution rather than explanation.

Common mistake to watch for: Explaining why the delay happened before acknowledging the customer's experience. Customers do not want reasons first. They want to know the person they are speaking to understands why this matters.

Role play escalation: The customer responds with more anger. The agent must stay calm for a second exchange without repeating the same empathy statement — which starts to feel scripted. The trainer should push agents to find genuine, varied language for sustained de-escalation.

Scenario 9 - Customer threatening a chargeback

Difficulty: Advanced

Channel: Email or website chat

Skill being practised: Remaining composed under financial pressure while protecting the business

Customer message: "If you do not refund me today, I am going to contact my bank and dispute the charge."

Agent response: "I understand you feel strongly about this and I want to resolve it properly for you. You are absolutely entitled to contact your bank, and I would not want to stop you doing that if you feel it is the right step. What I would like to do first is look at your case and see if I can resolve this directly for you today, which is usually faster than a dispute process. Can you share your order number so I can see exactly what has happened?"

What the agent is practising: Not treating the chargeback threat as an attack, not panicking into an immediate refund they cannot verify, and redirecting toward resolution while acknowledging the customer's right to take that step.

Common mistake to watch for: "If you dispute the charge, we will flag your account." This escalates the situation significantly and puts the business on poor legal ground. Agents should never threaten consequences for a customer exercising their consumer rights.

Scenario 10 - Customer asking for a manager

Difficulty: Advanced

Channel: Website chat or phone

Skill being practised: Handling the request without feeling bypassed or defensive

Customer message: "I want to speak to your manager. You are not helping me."

Agent response: "I hear you, and I want to make sure you get the help you need. I am going to do everything I can to resolve this for you right now, and if I cannot get it to a place you are happy with, I will absolutely get a senior member of our team involved. Can you help me understand what outcome you are looking for? I want to make sure whoever handles this has the full picture."

What the agent is practising: Not immediately routing to a manager, which rewards the escalation request and trains customers to skip front-line agents. Instead, making a genuine attempt to resolve while making clear that escalation is available.

Common mistake to watch for: "I am the manager" as a response to avoid escalation. This is dishonest and makes the situation significantly worse if the customer discovers otherwise.

Scenario 11 - Product quality complaint

Difficulty: Intermediate

Channel: Instagram DM or email

Skill being practised: Responding to subjective quality issues without admitting a defect prematurely

Customer message: "The quality of this product is really poor. It does not look anything like the photos on your website."

Agent response: "I am sorry to hear that — this is definitely not the experience we want you to have. Quality matters to us and I take feedback like this seriously. Could you describe what specifically feels different or share a photo if you are happy to? I want to understand exactly what you are seeing so I can figure out the best way to make this right for you."

What the agent is practising: Gathering information before committing to a resolution, validating the feeling without validating a defect that has not been confirmed, and moving toward a genuine solution rather than a defensive response.

Scenario 12 - Delivery attempted when not home

Difficulty: Intermediate

Channel: WhatsApp

Skill being practised: Navigating third-party carrier limitations clearly

Customer message: "The courier came when I was not in and now I cannot get a redelivery booked. Can you help?"

Agent response: "I am sorry about the missed delivery, that is frustrating. I can see your parcel is currently held at [carrier name]. The quickest way to get this rebooked is directly through their app or website — here is the link: [carrier link]. They should be able to arrange a specific delivery time that works for you. If you have trouble getting that sorted, come back to me and I will contact them on your behalf. Which carrier is showing on your tracking?"

What the agent is practising: Giving the most efficient resolution path (direct carrier contact) while making clear they will step in if needed. Agents should not take ownership of logistics processes they cannot control, but they also should not abandon the customer to navigate alone.

Scenario 13 - Order partially missing

Difficulty: Intermediate

Channel: Website chat or email

Skill being practised: Investigating before resolving — not promising before checking

Customer message: "My order arrived but one item is missing from the box."

Agent response: "I am sorry to hear that — let me look into this straight away. Sometimes items are dispatched separately if they come from different locations, so I want to check whether that is the case before we take the next step. Can you confirm your order number? I will check your dispatch notes and come back to you within a few minutes with a clear answer."

What the agent is practising: Not immediately promising to send the missing item before confirming whether it was shipped separately. This prevents the situation where a replacement is sent but the original arrives the next day.

Role play escalation: The trainer confirms the item was all in one shipment. The agent must now commit to sending a replacement without further investigation delays.

Scenario 14 - Pre-sale product question

Difficulty: Beginner

Channel: Instagram DM or website chat

Skill being practised: Converting a product question into a purchase conversation

Customer message: "Does this come in a size 14? I cannot see it on the website."

Agent response: "Great question — yes, size 14 is available in most colourways but it does vary. Which colour were you looking at? I can check stock for you right now and give you the exact availability. If the one you want is low in stock, I can also let you know what the restock timeline looks like."

What the agent is practising: Giving a specific answer rather than directing the customer back to the website, adding value by checking stock proactively, and creating a purchase moment from a simple availability query.

Common mistake to watch for: "Please check the website for availability." This sends the customer somewhere they already checked. Pre-sale queries on chat and Instagram DM deserve specific, instant answers.

Scenario 15 - Customer who wants an exception

Difficulty: Advanced

Channel: Email or website chat

Skill being practised: Holding a policy boundary with empathy and without apology

Customer message: "I know your return policy is 30 days but my circumstances were difficult and I am asking you to make an exception. I have spent a lot of money with your brand over the years."

Agent response: "I appreciate you reaching out and I can hear that this has not been a straightforward situation. Your loyalty to us genuinely matters and I want to handle this thoughtfully. I am not able to make that decision myself, but I am passing your case to a senior member of our team right now who has the authority to review it properly. They will be in touch within [timeframe] and they will have your full order history in front of them. I want to make sure this gets the right level of attention."

What the agent is practising: Not refusing outright, not making a promise they cannot keep, and not making the customer feel penalised for asking. The escalation to someone with decision-making authority is the correct move — not a flat policy answer and not an unauthorised exception.

Role play escalation: The customer pushes further: "I just want a yes or no." The agent must maintain the same boundary with warmth rather than caving under social pressure or becoming cold.

From role play to AI automation

The connection most training guides miss: every scenario your agents handle confidently through role play is a candidate for AI automation.

A beginner scenario is one with a clear, data-driven answer. Order tracking queries. Discount code eligibility checks. Return requests within policy. These scenarios have defined inputs and defined correct outputs. Once your agent can handle them consistently, an AI chatbot connected to your store data can handle them automatically at scale — in seconds, on any channel, at any hour.

An intermediate scenario often has a data layer and an empathy layer. The data layer — checking return eligibility, verifying dispatch dates, confirming order contents — AI handles directly. The empathy layer — sustaining a calm, genuine tone when a customer is frustrated — AI handles through well-configured response templates and tone settings.

Advanced scenarios require human judgment. A customer asking for an exception. A chargeback threat that may or may not be legitimate. A product quality complaint that needs investigation. These are the scenarios where AI should acknowledge and escalate, not resolve.

The framework is simple: use role play to identify which scenarios your team handles consistently well. Those are the first scenarios to automate with AI. Reserve your team's energy for the scenarios that genuinely need human judgment.

For stores wanting to see how this works in practice, the AI chatbot for customer service guide covers which query types AeroChat automates automatically and how the escalation to human agents is configured.

The Shopify complaint handling guide covers specifically how the AI handles the scenarios in this list that are appropriate for automation, with the actual response templates the chatbot sends.

Frequently asked questions

How do you run a customer service role play session?

Assign roles, provide the scenario brief to both participants, and run the conversation using your actual support tools. After the scenario, debrief on three things: what the agent did well, what they could improve, and what the customer was feeling at each point. Use the customer perspective discussion to build genuine empathy rather than just practising scripted responses.

Which customer service scenarios should new agents practise first?

New agents should start with beginner scenarios that have clear, data-driven resolutions and low emotional stakes. Order tracking queries, discount code troubleshooting, and return requests within policy are the right starting points. Move to intermediate scenarios — angry customers, damaged products, refund delays — once agents are consistently confident with the basics. Reserve advanced scenarios for experienced agents.

How are ecommerce customer service scenarios different from general scripts?

Ecommerce customer service in 2026 happens primarily on WhatsApp, Instagram DM, and website chat — not on phone calls. Scripts written for these channels need to be shorter, more conversational, and mobile-friendly. The most common ecommerce scenarios — WISMO queries, wrong item received, return policy questions, discount code issues — are specific to online retail and require product and order data access that general customer service training scripts do not account for.

Can AI replace customer service role play training?

No. Role play training builds the human judgment, empathy, and composure that agents need for the situations AI cannot handle — damaged product complaints with emotional context, exception requests from loyal customers, chargeback threats, and escalation management. What AI replaces is the repetitive execution of the scenarios agents master through role play — the routine queries with clear answers that do not require human judgment. The two approaches work together: training builds capability, AI handles repetition.