Customer support

Ecommerce Customer Support: Request Types and Templates

A practical guide to ecommerce customer support: common request types, what to automate, reply templates per order stage and how to handle order tickets.

In this article
  1. The request types you will actually get
  2. What to automate and what to keep human
  3. Reply templates by order stage
  4. Handling order tickets
  5. Peak days: sales, holidays and campaign launches
  6. Frequently asked questions

Open the inbox of almost any online store and the same few messages fill it: where is my order, can I return this, my payment went through but the order says unpaid. Ecommerce customer support is less about clever answers than about answering a small set of questions quickly, with the right order details in front of you.

This guide goes through the request types you will actually see, which ones to automate, reply templates for each stage of an order, how to handle order tickets and how to get through peak days. For the tools themselves, see Hodhod for online stores.

The request types you will actually get

Sort a week of your conversations by topic and most of them will land in one of these groups.

Type What the customer really wants What the agent needs on hand
Order status To know where the parcel is Order number, shipping date, tracking link
Delivery problem A late, lost or wrongly delivered parcel fixed Carrier, last tracking event, address
Return or refund To send something back and get money or a replacement Return window, item condition rules, refund method
Payment To know whether they were charged and whether the order exists Order number, payment reference, time of payment
Sizing and product questions Confidence before buying Size charts, materials, stock
Order changes To change an address or cancel before shipping Current order stage, cut-off time

Add a label for each group. After a month, the label counts tell you what to fix on the site itself. If a third of your chats are "where is my order", a clearer tracking page will do more than a faster reply.

What to automate and what to keep human

Automate the questions whose answer is always the same, and keep people for the ones where a mistake costs money.

Good candidates for automation - Shipping costs and delivery times by city or region - Return policy and how to start a return - Accepted payment methods - Opening hours and when the next reply is expected - Collecting the order number and email before a person joins

Keep with a person - Payment charged but no order - Lost or damaged parcels - Refund exceptions - Angry customers; see how to handle angry customers

A decision-flow chatbot can offer these choices as buttons, ask for the order number and hand over to an agent or open a ticket. Canned replies cover the rest, so agents do not retype the same paragraph forty times a day. More on that in canned response templates.

Reply templates by order stage

Replace the parts in brackets. Keep them short; customers read the first two lines.

Before purchase: delivery question

Hi [name], orders placed before [cut-off time] ship the same working day. To [city], delivery usually takes [x] working days. Shipping costs [amount] and is free above [threshold]. Want me to check a specific product?

After payment: order not confirmed

Thanks for letting us know, [name]. Could you send the order number or the email you used at checkout, and the time of the payment? I will check it now and come back to you within [time].

Shipped: where is my order

Your order [number] left our warehouse on [date] with [carrier]. Tracking: [link]. The last update shows [status]. If it has not moved by [date], write back here and I will contact the carrier.

Delayed or lost

I am sorry, [name]. Your parcel [number] has been stuck at [status] since [date], and that is longer than it should take. I have opened a case with [carrier] and will update you by [date]. If it cannot be found, we will [resend / refund], your choice.

Return request

Sure. You can return items within [x] days of delivery if they are unused and in their packaging. Send me the order number and which item, and I will give you the return address and steps. The refund goes back to [method] within [y] days of us receiving the parcel.

Sizing

It depends on the cut. For [product], the size chart is at [link]. Tell me your usual size in [brand] or your height and weight and I will suggest one. If it does not fit, the exchange is covered by our return policy.

Say plainly what you will do and by when. Do not promise a date you do not control, such as a carrier's.

Handling order tickets

A conversation about a late parcel or a refund rarely ends in one sitting. When you are waiting for the carrier or the warehouse, turn it into a ticket so it does not disappear down the inbox. A workable routine:

  1. Put the order number in the subject. "Order 48213: parcel stuck in transit" is findable. "Problem" is not.
  2. Set a category such as delivery, return or payment, so each team sees its own.
  3. Set a priority. Payment charged without an order is urgent; a sizing exchange is not.
  4. Give it a due date. Use the carrier's or warehouse's promised date, not an optimistic one.
  5. Use the status honestly. "Waiting on customer" when you asked for photos; "In progress" when you are waiting for someone inside.
  6. Merge duplicates. Customers who write twice should end up in one ticket.
  7. Write what you promised in an internal note, so whoever picks it up tomorrow knows.

The customer gets replies by email as well, so they do not have to keep the chat open. Automation rules can set the priority from a label such as "payment-issue" and attach an SLA, so the urgent ones are never missed.

Peak days: sales, holidays and campaign launches

Volume does not rise evenly. Take a sale day with 600 orders shipped. If one customer in five asks where their order is, that is 120 conversations about status alone, before a single return or sizing question. If an agent handles about 12 chats an hour, those 120 take ten agent-hours. Your own ratios will differ, so work them out from last year's numbers.

What helps most is removing the repeat questions before they reach an agent: tracking links in shipping emails, a chatbot flow for status questions, and a banner in the widget for known delays. The full month-by-month preparation is in the peak season support checklist, and out-of-office messages cover the nights and holidays.

Afterwards, read the label counts and the CSAT survey results and keep what worked.

To see how this setup looks on your own store, request a free demo.

Frequently asked questions

Which requests should an online store automate first?

Start with order status and delivery times, then returns policy and payment methods. These are the most repeated and the least risky. Leave payment failures and lost parcels to a person.

How fast should a store reply to customers?

It depends on the channel and the stage. Chat before purchase is time-sensitive because the customer is deciding right then; a return question can wait a few hours. Set targets per request type; see what is an SLA in customer support.

Do I need a separate tool for order tickets?

Not necessarily. If chats and tickets live in the same inbox, the history moves with the conversation when you turn it into a ticket. Keep the order number in the subject either way, so anyone can find the ticket from the order.

How do I handle customers who write on several channels?

Keep website chat and Telegram in one inbox so each customer has one history, and ask for the order number at the start of every conversation.