Skip to content
Sending multiple me...
 
Notifications
Clear all

Sending multiple messages on Whatsapp Cloud API with the same trigger

2 Posts
2 Users
0 Reactions
4 Views
Ahmet_furkan_Laz
(@ahmet_furkan_laz)
Posts: 1
New Member
Topic starter
 

Hi everyone,

I’m working on a callin.io workflow that receives messages from WhatsApp (using the official WhatsApp Cloud API + callin.io WhatsApp Trigger).

I want to combine multiple short user messages into a single input before sending them to GPT or another processor.

My goal:

  • When a user sends a message, I want to wait 8 seconds to see if any more messages are coming.
  • If additional messages come within that window, I want to append them to the previous ones.
  • Once there’s an 8-second pause, I consider the input complete and then send the combined message to the AI.

The problem:

  • WhatsApp sends each message as a separate webhook call, so each message starts a new execution in callin.io.
  • I’m trying to store the messages temporarily (using Simple Memory) and track the last_message_time.
  • But since each execution is isolated, I can’t easily set up a delay inside the same flow that checks for future messages.
  • I tried using Wait + IF nodes with timestamp comparison, but it doesn’t feel reliable because the workflow finishes before new messages arrive.

What I’m looking for:

  • Has anyone successfully built a “delayed input merge” system in callin.io?
  • Is there a way to delay response until user input is silent for X seconds, across multiple executions?
  • Or should I use an external system (e.g. Redis, database + cron job) to collect and check messages before triggering GPT?

Any design patterns or ideas are welcome

:folded_hands:

Thanks in advance!

 
Posted : 17/06/2025 5:56 pm
Pedro_Henrique_Carva
(@pedro_henrique_carva)
Posts: 1
New Member
 

Hello! I'm facing a similar challenge while building my own WhatsApp bot. My current approach involves creating an OpenAI module. I'd pass the threadID and prompt it to determine whether to respond immediately or wait for more context before replying to the client.

This is the only viable solution I've come up with so far. Please share if you discover a more effective method!

 
Posted : 06/07/2025 5:17 pm
Share: