WISMO: How AI Agents Answer 'Where Is My Order?'

WISMO stands for "where is my order?" It is the contact a customer makes after buying, when they want to know where the parcel is. It arrives as a phone call, a WhatsApp message, an email, a chat or a form, and it is one of the most repeated contacts in ecommerce support. This guide covers why WISMO piles up, what to fix first, and how an AI agent answers it from your store record. If you want the wider picture across voice, WhatsApp and email, start with AI agents for ecommerce.

What is WISMO?

WISMO is shorthand for "where is my order?". Support teams use it as a ticket category: every contact whose only question is the status of an order that has already been paid for.

Related terms you will see:

• WISMO calls are the phone version of the same question.
• WISMO tickets are the email and helpdesk version.

The question is simple. The answer sits in a record your team already has: the order, the fulfilment, the tracking number, the carrier's last scan. Nobody needs judgment to read it back. They need a lookup and a reply.

That is exactly why it is a good job for software, and exactly why it eats a human team's day.

Why WISMO contacts pile up

Four things feed the queue, and most stores have all four.

1. The customer cannot find the status themselves. The confirmation email has a tracking link, but it is buried or expired, or the carrier page is confusing. Asking you is faster than searching.

2. Something went wrong and nobody told them. A delayed shipment, a split order or a stock-out creates a gap between what the customer expects and what happens. They fill the gap with a contact.

3. The question arrives on every channel. The same customer may call, then message on WhatsApp, then email. Each channel is staffed separately, so the same lookup is done three times.

4. It spikes at the worst moment. Sales, holidays and carrier delays push order volume and WISMO volume up together, which is when your team has the least spare time.

Each WISMO contact is quick for a person. That is the trap. It is quick, it never ends, and it crowds out the contacts that need a human: complaints, damaged goods, a customer deciding whether to buy again.

Fix it before the contact: the proactive layer

The cheapest WISMO contact is the one that never happens. Do these first, before any AI agent.

• A tracking page you control. One branded page where the customer enters an order number and sees status in plain language. It is better than a raw carrier link because it can explain what "in transit" means for this order.
• Notifications at the points that matter. Order confirmed, shipped, out for delivery, delayed. The delayed message is the most valuable and the one most stores skip.
• Honest delivery estimates at checkout. A date you cannot keep creates a WISMO contact on the day it passes.
• Clear exceptions. If an order is split or backordered, say so the moment it happens.

Proactive fixes remove the avoidable contacts. They do not remove all of them. Some customers will always ask, some orders will always go wrong, and some people simply prefer to talk to someone.

Answer the contact that still arrives: the reactive layer

After the proactive work, the contacts that remain are real questions from people who want an answer now, in a channel they chose. This is where an AI agent earns its place.

An agent can sit on the phone line, on WhatsApp or on email and answer the same question the same way on each, from the same store record. It does not replace the tracking page. It covers the customers who never opened it.

If you run B2B order tracking against an ERP, the same pattern applies with different systems behind it. That case is covered in how a voice agent handles order tracking and ETA calls, which is written for B2B buyers and suppliers rather than online stores.

How an AI agent answers a WISMO contact, step by step

This is the flow VoxDonna scopes for order status. It is the same on voice and WhatsApp, and the store connection is the part that makes it work. VoxDonna has connected agents to a live Shopify store for order status and for product and stock lookups, using scoped API access agreed during scoping.

Step 1: Understand the question. The customer says or types something like "where is my order?" or "my parcel hasn't come". The agent recognises an order-status request in the customer's own words, in their language, with no menu to press through.

Step 2: Identify the order. The agent asks for what it needs: an order number, or the email or phone number on the order. It confirms before it reads anything out. If a customer gives an order number only, a sensible store also asks for a second detail before sharing delivery information, so a stranger cannot read someone else's order.

Step 3: Look up the order in the store. Through scoped, read-only API access, the agent finds the order and reads its status and fulfilment details. It answers from the record, not from a script and not from a guess.

Step 4: Read the delivery stage. Where the store record carries the carrier and tracking number, the agent gives them. Where it also carries tracking status, the agent reads that back. Pulling live scans straight from a carrier is a separate integration, and it is scoped on its own rather than assumed.

Step 5: Answer in the customer's language. The reply is given in the language the customer used. A small team that is strong in one language but thinner in others gets cover where the desk is weakest.

Step 6: Offer the next step. Send the tracking link by message, or tell the customer when the record next updates. The agent does not invent a date. If the record has no date, it says so.

Step 7: Hand over the exceptions with context. If the order is missing, late beyond what the agent may speak to, marked damaged or disputed, or the customer is upset, the agent passes the conversation to a person. The person gets the transcript, the order details and what was already said, so the customer does not repeat it.

An illustrative exchange, not a customer recording:

Customer: Where's my order? I ordered last week.
Agent: I can check that. What's the order number, or the email you used?
Customer: It's 4821.
Agent: Thanks. Can you confirm the email on the order so I can read the details out?
Customer: (gives email)
Agent: Order 4821 has shipped. It left on the date shown in your store record, and the tracking number is on the way to your phone now. Is there anything else?

The point of the example is the shape, not the wording: identify, verify, look up, answer, hand over when the record runs out.

What an AI agent should not do on WISMO

Good WISMO automation is defined as much by its limits as by its reach.

• No invented delivery dates. If the record does not hold a date, the agent says it does not have one and routes the question.
• No promises about compensation. Refunds, re-sends and goodwill belong to your policy and your team.
• No reading out an order without a check. Order details are personal. Verify before you share.
• No dead ends. Every path ends in an answer or a handover with context.

What to measure

You do not need a long dashboard. Five numbers tell you whether WISMO automation is working.

1. WISMO share of contacts. What fraction of all contacts are order-status questions? Pull this from a sample of your own call log, WhatsApp inbox and helpdesk before you start.
2. Resolved without a person. Of the WISMO contacts the agent took, how many ended with the customer answered and no handover?
3. Handover rate and reason. When the agent hands over, why? Missing order, delay, upset customer? The reasons show you where the store data or the process has a gap.
4. Time to answer. How long does a customer wait to get a status? Compare it with your pre-agent baseline, by channel.
5. Repeat contacts. Do customers contact you a second time about the same order? A high rate means the first answer was not clear or not complete.

Count before you automate, and keep the same count afterwards. A figure you did not measure on your own data is a guess.

For a worked example of an agent on a support line, Le Marquier, a French premium consumer brand rather than an online store, reports 98% of calls resolved by the agent, with order and delivery status read from the ERP. Those are VoxDonna's own published figures, and they are not Shopify results.

When not to use an AI agent for WISMO

An agent is not always the right tool. Skip it, or hold off, when:

• Your volume is small. If a person handles every WISMO contact in a few minutes a day, the proactive fixes will do more for less effort.
• Your order data is unreliable. If statuses in the store are often wrong or updated late, an agent will read out wrong answers faster. Fix the data first.
• Most of your WISMO contacts are really complaints. If the cause is late or lost parcels, the problem is fulfilment, not the answer. Automating the answer hides the symptom.
• Your customers need a human for every contact. High-value or sensitive orders sometimes call for a person by default.
• You cannot give the agent safe access. If the store setup cannot support scoped, read-only access, resolve that before you start.

A good agency will tell you when an agent does not fit. VoxDonna scopes one workflow end to end, such as order status, and says so plainly when the answer is no.

Where VoxDonna fits

VoxDonna builds custom AI agents rather than selling a self-serve app. Order status is one workflow among several on the AI agents for ecommerce page. The phone side is described on AI voice agents, and the messaging side on WhatsApp Donna agents. Custom builds are scoped on a call, so you agree the workflow, the store access and the handover rules before work starts.

Frequently asked questions

What does WISMO mean?

WISMO stands for "where is my order?". It is the support category for contacts whose only question is the status of an order the customer has already paid for. It reaches a store by phone, WhatsApp, email, chat or form.

What are WISMO calls?

WISMO calls are phone calls from customers asking where their order is. They are the voice version of WISMO tickets. A voice agent can answer them by finding the order in the store, reading back the status, and handing over to a person when the record does not settle the question.

How do you reduce WISMO calls?

Start with prevention: a branded tracking page, notifications at shipment, delay and delivery, and honest delivery estimates. Then cover the contacts that still arrive with a faster answer, either a trained team with a good lookup tool or an AI agent connected to your store. Measure your WISMO share before and after so you know what moved.

Can an AI agent look up orders in Shopify?

Yes. VoxDonna has connected agents to a live Shopify store for order status and for product and stock lookups, using scoped API access agreed during scoping. The agent reads from the store record and hands over when the record is missing or unclear. Other platforms are a scoping question, not a pre-built connector.

Does an AI agent replace my support team on WISMO?

No. It takes the repeatable status questions so your team can handle complaints, damaged goods and customers who need a person. The agent hands over with the full conversation, so the person starts with context rather than from zero.