How to qualify, categorize, route or mark conversation as spam with the Automatic Triage feature
The Triage feature is here to help you perform qualification-related actions anytime a message is received in your Crisp inbox.
You can define your own rules to silently and automatically add a segment to a conversation, mark it as spam, or use it to trigger various actions which we'll see in this article.
The Triage feature is available in all plans! Simply head over to Settings > Inbox Settings > Automatic Triage to get started.
Once you've opened up the Triage interface from Settings > Inbox Settings > Automatic Triage, you will find a couple of sections available:
Block Ruleset
Sort Ruleset
➞ Block rules allow you to mark certain messages/conversations as spam, which will redirect them to your Spam Inbox where you choose whether to delete them permanently or to authorize them back in your main inbox. These rules can also be used to directly and permanently "drop" messages (implying it won't go to your Spam inbox at all, the message will be deleted instantly).
These rules can be applied to any channels! You can detect spammy messages sent from Email, Chat, Instagram, Whatsapp... etc.
➞ Sort rules are instead used to automatically add a segment or custom data to your conversations. This allows you to effortlessly categorize your conversation right as they land in your inbox, whether you wish to simply be able to create a filter for them, assign them to specific agents, or even redirect them to sub-inboxes.
You can create rules for each one, which will then be evaluated in the order in which they are placed (when expanding a rule, you will find little arrows at the bottom to change their priority).
For each rule, you can configure:
If all or any condition must match (switch from or/and)
Channels for which this rule apply (if none are selected, your rule is applied in all of them)
Filters, where you defined the conditions of your rule (more on that below)
The action (add a segment or data) and their value
ℹ️ About Filters:
These can be applied to detect patterns either on the email address of the user, or the content of their message. You can select among various options to more accurately target certain terms (starts with, ends with, has certain patterns anywhere, etc).
A few examples:
To detect all emails from the acme.com group, we would use Email From Address with Ends With @acme.com
To detect a message containing offensive terms, we could catch relevant terms a Message Text and Has Patterns for values such as jerk and idiot
You do not need to add any wildcards to detect these terms inside of a sentence, the Triage rules will automatically do that for you.
Triage rules are very quick and easy to set up. Crisp evaluates them first, before your Operator routing rules, Sub-inbox routing rules, and before your Chatbot scenarios even fire.
➞ For Block Rules, this allows you to get rid of unsolicited messages, spam, junk before they land in your inboxes to keep it neat and clean and maintain the focus of your agents on the conversations which matter.
➞ For Sort Rules, this allows you to categorize your conversations and automate various actions straight away.
You can add a segment to all emails sent from specific email address and create an operator rules to assign those to specific agents:
How does Routing / Assign work?
Or route them to specific sub-inboxes for easier organisation of your conversations:
How can I create sub-inboxes in my Crisp Workspace?
Adding a segment in your conversation also allows you to filter and monitor their usage frequency in the Analytics:
How does Crisp Analytics work?
You can define your own rules to silently and automatically add a segment to a conversation, mark it as spam, or use it to trigger various actions which we'll see in this article.
The Triage feature is available in all plans! Simply head over to Settings > Inbox Settings > Automatic Triage to get started.
Getting started with the Automatic Triage
Once you've opened up the Triage interface from Settings > Inbox Settings > Automatic Triage, you will find a couple of sections available:
Block Ruleset
Sort Ruleset
➞ Block rules allow you to mark certain messages/conversations as spam, which will redirect them to your Spam Inbox where you choose whether to delete them permanently or to authorize them back in your main inbox. These rules can also be used to directly and permanently "drop" messages (implying it won't go to your Spam inbox at all, the message will be deleted instantly).
These rules can be applied to any channels! You can detect spammy messages sent from Email, Chat, Instagram, Whatsapp... etc.
➞ Sort rules are instead used to automatically add a segment or custom data to your conversations. This allows you to effortlessly categorize your conversation right as they land in your inbox, whether you wish to simply be able to create a filter for them, assign them to specific agents, or even redirect them to sub-inboxes.
You can create rules for each one, which will then be evaluated in the order in which they are placed (when expanding a rule, you will find little arrows at the bottom to change their priority).
For each rule, you can configure:
If all or any condition must match (switch from or/and)
Channels for which this rule apply (if none are selected, your rule is applied in all of them)
Filters, where you defined the conditions of your rule (more on that below)
The action (add a segment or data) and their value
ℹ️ About Filters:
These can be applied to detect patterns either on the email address of the user, or the content of their message. You can select among various options to more accurately target certain terms (starts with, ends with, has certain patterns anywhere, etc).
A few examples:
To detect all emails from the acme.com group, we would use Email From Address with Ends With @acme.com
To detect a message containing offensive terms, we could catch relevant terms a Message Text and Has Patterns for values such as jerk and idiot
You do not need to add any wildcards to detect these terms inside of a sentence, the Triage rules will automatically do that for you.
What can Triage Rules be used for
Triage rules are very quick and easy to set up. Crisp evaluates them first, before your Operator routing rules, Sub-inbox routing rules, and before your Chatbot scenarios even fire.
➞ For Block Rules, this allows you to get rid of unsolicited messages, spam, junk before they land in your inboxes to keep it neat and clean and maintain the focus of your agents on the conversations which matter.
➞ For Sort Rules, this allows you to categorize your conversations and automate various actions straight away.
You can add a segment to all emails sent from specific email address and create an operator rules to assign those to specific agents:
How does Routing / Assign work?
Or route them to specific sub-inboxes for easier organisation of your conversations:
How can I create sub-inboxes in my Crisp Workspace?
Adding a segment in your conversation also allows you to filter and monitor their usage frequency in the Analytics:
How does Crisp Analytics work?
Updated on: 08/04/2025
Thank you!