In this video, we’ll show you how we’ve automated the entire customer interaction and order submission process for a printing company, using AI and WhatsApp.
Here’s how it works:
The scenario begins when a message is received via a WhatsApp webhook — for example, a customer asking for a print job.
That message is connected to HubSpot, where the system retrieves it.
It first checks if a customer record already exists. If it does, it updates the record. If not, a new one is created.
Next, the system sends a message asking the customer to confirm whether they have more documents to upload.
It checks the datastore to see if there’s already an active thread for that client.
The AI assistant processes the message and updates the thread if one exists.
This is the part where the AI actually responds to the client.
It gathers all the printing information needed. Once everything’s collected, the AI summarizes the data and generates a Google Docs file with all the order details.
That document is then sent to an external printing API or document service.
Meanwhile, on the lower branch of the workflow, if a human agent steps in, the system saves a “stop” status in the datastore to prevent the AI from continuing the conversation.
Here’s how the AI prompt works:
It receives the client’s actual message, the type of document requested (since different documents trigger different actions), plus any additional details such as pickup or courier delivery options.
To wrap up: this workflow automates the entire process — from receiving a customer message to preparing and sending print orders.
It uses AI to understand requests, collect key details, generate the necessary documents, and hand things off efficiently.
If a human takes over, the AI knows when to step back.
Thanks for watching — and if you want to see more workflows like this, visit www.makeitfuture.com.
