How to connect an order system or carrier feed to a WhatsApp or website chatbot for where-is-my-order answers, when to send proactive updates, when delays should reach a person, and what it costs.
WISMO (where is my order) tickets dominate logistics support. A WhatsApp or web chatbot reads tracking data from Shopify or Shiprocket, or from your ERP or carrier tracking through Zapier or a webhook, so customers self-serve status, delivery windows, and pickup points. Start with customer support chatbot on WhatsApp chatbot.
"Where is my order" (WISMO) automation means a chatbot on WhatsApp, your website or email looks up a shipment in your order management system or carrier feed and answers with the latest scan, the delivery window and what to do next, without a person opening a ticket. It is for ecommerce brands, 3PLs, couriers and marketplaces whose support queue is dominated by status questions that the tracking page already answers.
The decision is which system the agent reads from, whether to send proactive updates or only answer when asked, which channel carries the updates, and when a delay or loss should reach a person. This guide covers each with a cost example built on Botcadence pricing. For the industry view, see AI chatbots in logistics.
The flow has four steps, and the second is where most projects stall.
Step three needs a mapping table that you own, because carriers use different codes for the same event.
| Carrier or OMS code | What the customer hears | Next action offered |
|---|---|---|
| Label created, no scan | "Your order is packed and waiting for carrier pickup." | Expected ship date |
| In transit | "Your parcel left [city] at [time] and is on its way." | Expected delivery window |
| Out for delivery | "Your parcel is on the van today." | Delivery instructions |
| Delivery attempted | "The driver could not deliver at [time]." | Reschedule or pickup point |
| Exception or delay | "There is a delay at [location]; the new estimate is [date]." | Apology, new date, escalation if the SLA is breached |
| Delivered | "Delivered at [time], signed by [name] if captured." | Report a problem |
| No scan for N days | "We are checking with the carrier." | Open a trace, escalate to a person |
Build the table before you train the agent. If a code has no row, the agent should say it is checking and escalate rather than guess.
| Channel | Strength | Constraint | Botcadence setup |
|---|---|---|---|
| Proactive alerts reach the customer's phone; replies come back in the same thread | Messages outside Meta's 24-hour customer service window need approved templates and stored opt-in | After Meta verification, usually 24 to 48 hours | |
| Website chat | Customers already on the tracking page; no opt-in needed | Reactive only | About 5 minutes after documents are uploaded |
| Email agent | Formal records, proof-of-delivery disputes, attachments | Slower; customers expect hours, not seconds | Same knowledge base |
| Voice | Customers who call about a missed delivery | Minute-based pricing | About 5 business days for a managed agent |
Many teams run WhatsApp for proactive updates and the website for reactive lookups, with email for disputes, all from one knowledge base and one status mapping.
Reactive lookup removes the human from the ticket. Proactive updates remove the ticket. A proactive update at three moments covers most of the questions customers would otherwise ask:
On WhatsApp, each of those outside an open 24-hour window is a template message, which Meta must approve and which carries a per-message conversation charge that varies by country and category. Botcadence sends these through templates and broadcasts; the WhatsApp Business API pricing guide covers how the charges work, and the WhatsApp Business API glossary entry defines the terms. Customers must have opted in at checkout or in the thread, and the message must be transactional, not promotional, to stay inside Meta's rules.
Botcadence web chat and WhatsApp start on the Free plan, with Starter at $39, Growth at $99 and Business at $199 a month, and Enterprise custom above that; the plan depends on reply volume, so check the pricing page. Meta's conversation charges for WhatsApp are billed separately.
A worked example, with every input labelled as an assumption. Say a brand ships 12,000 orders a month and, before automation, 6 percent of orders generate a status contact: 720 contacts. Say each takes a person 4 minutes including the carrier lookup: 48 hours of agent time a month. With a reactive agent reading from the OMS, say 80 percent of those 720 resolve without a person: 576 automated, 144 escalated, and the human time falls to about 10 hours. Add proactive updates at three touches per order, 36,000 template messages a month on WhatsApp, and the reactive volume usually falls further, but the template charges become the dominant cost line, so price them from Meta's current rate card for your country before deciding on three touches versus one.
The platform fee is the smaller number; the mapping table and the connector are the work. A sensible sequence is to launch reactive lookup first, measure how many contacts remain, then add the single proactive touch (shipped, with the tracking link) and measure again before adding out-for-delivery and exception messages.
| Situation | Agent action | Escalate when |
|---|---|---|
| Delay with a new carrier estimate | Acknowledge with the latest scan and the new date | Estimate is beyond your promised SLA |
| No scan for several days | Say a trace is being opened; set expectations | Immediately, with order details, to the carrier liaison |
| Damaged on arrival | Collect photos and order ID | Always; refunds and reships stay human |
| Customer says it was not delivered despite a delivered scan | Collect address and delivery note | Always; possible fraud or carrier error |
| Address change before dispatch | Collect new address, confirm | If the order already shipped |
Botcadence human handoff passes the transcript and collected fields with the escalation, and workflow automation can open the ticket in the helpdesk.
Choose it when status questions are a large share of tickets, when your OMS or carrier has an API or webhook, and when customers already message you on WhatsApp or chat on the tracking page.
Hold off when tracking data is unreliable (an agent that confidently repeats a wrong scan is worse than a person who checks), when you cannot store opt-in for proactive messages, or when you have not yet built the status mapping. Fix the data first; the customer support chatbot and FAQ automation pages cover the simpler policy questions you can launch in the meantime.
For the storefront side of the same problem, see the Shopify chatbot guide. The Botcadence chat agents page covers the website and WhatsApp agents, the WhatsApp Business Suite handles the proactive side, and together they give a logistics team one place to put status questions and the handoff rules around them.
Questions readers ask
Shopify feeds it through the native integration, Shiprocket through the in-product connector, and a custom OMS or carrier API through Zapier or a webhook. Map raw status codes to plain sentences before customers see them.
WhatsApp fits proactive delivery alerts; email fits formal POD disputes. Many teams do both from one knowledge base.
Auto-acknowledge delay with latest scan, then escalate to human if SLA breached or customer confirms loss.
Proactive outbound status messages at label, out-for-delivery, and delay events before customers ask.
Book a free demo and we will build a working AI agent for the channel this article covers, tailored to your business goals. No commitment required.
Ask AI about this page
