How to set up Zendesk auto assign tickets using sentiment and priority rules

Kenneth Pangan

Amogh Sarda
Last edited October 29, 2025
Expert Verified

Let's be honest, staring at a giant, unassigned ticket queue is a headache. Your team has to manually read through each one, figure out who's upset, who's just asking a simple question, and who needs a specialist, all while urgent issues are piling up. Every minute spent on triage is a minute a frustrated customer is left waiting.
Getting the right ticket to the right person, fast, is what separates good support from great support.
This guide will walk you through how to get this done automatically. We’ll show you how to set up Zendesk to auto assign tickets using sentiment and priority rules, so your most critical tickets jump to the front of the line. We'll start with the built-in Zendesk method and then look at a more straightforward way to get the same result with less fuss.
What you'll need to get started
Before we jump in, let’s make sure you have what you need. The requirements are a bit different depending on which path you take.
For the native Zendesk method:
-
The Zendesk Advanced AI add-on. This is a paid extra and is necessary for the sentiment analysis part.
-
Admin access to your Zendesk account, since you'll be tinkering with triggers and settings.
For the eesel AI method:
-
Any Zendesk plan will do.
-
An eesel AI account. You can get one set up in just a few minutes.
-
Admin access to your Zendesk account for the one-click integration.
Method 1: Setting up directly in Zendesk
Zendesk can handle this workflow using a feature called Intelligent Triage, which is part of that Advanced AI add-on I mentioned. The process involves turning the feature on and then building some specific triggers to tell Zendesk what to do.
Step 1: Turn on intelligent triage for sentiment
First things first, you need to give Zendesk permission to actually read the sentiment of incoming tickets.
-
Head over to the Admin Center in Zendesk.
-
Look for AI in the sidebar, then go to Intelligent triage and click on Sentiment.
-
Hit Manage Settings and make sure sentiment detection is switched on for the right channels, like Email and Web form.
Once you do this, Zendesk will start adding a "Sentiment" field to new tickets, labeling them as "Negative", "Neutral", or "Positive".
Step 2: Understand how Zendesk's AI predictions work
With Intelligent Triage enabled, Zendesk adds a few new, un-editable fields to your tickets. The one we care about here is the Sentiment field. Zendesk's AI looks at the customer's first message and assigns a value. This value is what you'll use to build your automation rules. It’s good to remember that the accuracy here is based entirely on Zendesk's own model.
Step 3: Create a trigger for negative sentiment tickets
Now for the fun part. Let's build a trigger that automatically snags any tickets with negative sentiment and does something with them.
-
In the Admin Center, go to Objects and rules > Business rules > Triggers.
-
Click Create trigger.
-
Under Meet ALL of the following conditions, you’ll want to add:
- "Ticket" > "Status" | "Is" | "New"
- "Ticket" > "Sentiment" | "Is" | "Negative"
- You could even add a condition for "Very negative" if you want to escalate the really angry ones even faster.
-
Under Actions, tell Zendesk what to do with these tickets:
- "Ticket" > "Set priority" | "High"
- "Ticket" > "Group" | "(Your Tier 2 or Escalations Team)"
- "Ticket" > "Add tags" | "negative_sentiment_escalation"
Step 4: Fine-tune your triggers
You can let the sentiment trigger handle setting the priority, or you can get a bit more detailed. For instance, maybe you want any ticket from a VIP customer to be marked as top priority, no matter what their sentiment is.
To do that, you'd just create a separate trigger that fires before the sentiment one. This trigger would check for a VIP customer tag or organization and set the priority accordingly. Then, your sentiment trigger can work alongside it.
Step 5: Test and watch your new workflow
Here's a word of caution with Zendesk triggers: there's no "practice" mode. The moment you activate them, they’re live. The only way to test is to watch your incoming tickets and see what happens. Keep a close eye on your queue to make sure everything is being routed correctly. If something goes wrong, be ready to disable the trigger quickly so you don't accidentally send a bunch of tickets to the wrong place.
Method 2: A simpler approach using eesel AI
While Zendesk’s own tools can get the job done, they can feel a bit rigid and require that expensive add-on. If you're looking for something more flexible and intuitive, a tool like eesel AI that connects directly to your helpdesk might be a better fit.
This approach gives you more precise control, lets you test your rules safely before they go live, and uses AI that learns from all your company knowledge, not just what's in your helpdesk.
Step 1: Connect eesel AI to Zendesk
Getting set up is refreshingly simple. You can create an eesel AI account and connect your Zendesk account with just one click. No developers needed, no complicated setup screens.
Right away, eesel AI starts learning from your past tickets, macros, and help center articles. You can also connect it to other places where your team keeps information, like Confluence or Google Docs. This gives its AI a much richer understanding of your company's specifics, which leads to more accurate analysis of what customers are saying.
Step 2: Build your automation workflow
Instead of juggling multiple Zendesk triggers and hoping they don't conflict, you can map out your entire workflow in eesel AI's visual editor.
-
Create a new AI Triage rule.
-
Set your condition: "IF" the ticket content's "sentiment" is "negative"...
-
Define the actions: "THEN" Set priority in Zendesk to "Urgent", Assign to group "Escalations", and Add tag "routed_by_eesel_ai".
You can easily layer in more logic, like checking for certain keywords or identifying a specific customer type, all in the same rule.
Step 3: Simulate your workflow on past tickets
This is where you can really feel confident about your changes. Before your new rule touches a single live ticket, you can run it in simulation mode against your historical ticket data.
eesel AI gives you a full report showing:
-
Exactly which of your past tickets the rule would have caught.
-
A preview of what it would have done (like changing the priority or assigning it to a new group).
-
A clear forecast of how the rule will affect your team's workflow.
This lets you tweak your logic until it's just right, all without any risk of disrupting your live support queue. It’s a safety net that you just don't get with native Zendesk triggers.
Step 4: Go live and monitor your workflow
Once you're happy with how the rule performs in the simulation, you can activate it. You can even roll it out gradually, maybe starting with just one email address or a specific form, to ease into it.
The analytics dashboard gives you a straightforward look at how your automation is performing, which can help you find ways to improve and even spot gaps in your help articles that might be causing customer frustration.
Extra tips for flawless ticket assignment
Whether you stick with Zendesk's tools or use a solution like eesel AI, here are a few best practices to keep in mind:
-
Look at intent, not just sentiment. A customer could sound perfectly happy but be asking about a complicated billing issue. The best routing also figures out why a customer is writing, not just how they feel.
-
Have clear priority levels. Make sure your team agrees on what makes a ticket "Urgent" versus "High" priority. Is it the customer's subscription level? The part of your product they're asking about? Having this defined makes automation much easier.
-
Start small. Don't try to automate your entire routing system overnight. Pick one or two obvious pain points, like routing angry feedback or urgent bug reports, and build from there.
-
Keep your knowledge base up to date. Any AI is only as smart as the information it learns from. A tool like eesel AI can help by automatically spotting common questions that don't have a help article yet.
Which method should you choose?
So, you have two solid options to have Zendesk auto assign tickets using sentiment and priority rules. The native Zendesk method works and is a decent choice if your team is already all-in on the platform and has the Advanced AI add-on.
But for most teams looking for a more powerful, flexible, and frankly, less stressful way to automate, a dedicated AI platform is the way to go. Tools like eesel AI not only make the setup process a lot easier but also tend to be more accurate because they learn from all your scattered company knowledge. And features like risk-free simulation let you build automations with confidence, knowing you won’t break anything or make your customers' day worse.
Ready to get your ticket queue under control?
Automate Zendesk ticket assignment with eesel AI
Stop letting manual triage and clunky triggers slow your team down. With eesel AI's AI Triage, you can build, test, and deploy smart routing rules in minutes, not hours.
Sign up for a free trial and see how easy it can be.
Frequently asked questions
Yes, absolutely. While the native Zendesk method requires a Professional plan and the Advanced AI add-on, alternative solutions like eesel AI enable you to set up Zendesk auto assign tickets using sentiment and priority rules on any Zendesk plan, without needing the extra add-on.
Zendesk's native AI uses its own proprietary model to detect sentiment, which forms the basis for its routing. External tools often leverage AI that learns from a broader range of your company's knowledge, which can lead to more nuanced and accurate analysis for routing.
The native Zendesk method does not offer a simulation mode, meaning any testing occurs live on your active tickets. Tools like eesel AI provide a simulation mode, allowing you to test your rules against historical data to ensure accuracy and prevent unintended consequences before activation.
You can combine sentiment with various other conditions, such as customer VIP status, specific keywords in the ticket, organization details, or even ticket forms. Dedicated AI platforms generally offer more flexibility to layer complex logic within a single rule, beyond what native triggers might easily allow.
For native Zendesk triggers, it's crucial to monitor your queue closely after activation and be ready to disable triggers quickly if they misbehave. With external tools, utilize their simulation features and consider a gradual rollout to minimize risk and ensure rules perform as expected.
The native Zendesk method involves configuring multiple triggers, which can become complex for intricate rules and requires familiarity with Zendesk's AI features. Solutions like eesel AI simplify the process with visual editors and straightforward integrations, often making complex rule setup quicker and more intuitive.




