Intercom automation to send post resolution survey in the messenger (2025 guide)

Kenneth Pangan

Stanley Nicholas
Last edited October 29, 2025
Expert Verified

Getting customer feedback right after a support chat ends is a golden opportunity. The conversation is still fresh in their mind, and it's the perfect moment to ask, "Hey, how did we do?"
But let's be real, the way you ask matters. A clunky survey can mean the difference between getting a useful reply and being completely ignored.
Too many teams are stuck wrestling with awkward survey tools or sending out generic requests that don’t actually help them get better. The point isn't just to get a thumbs-up or a thumbs-down; it's to figure out the why behind that rating.
This guide will show you a few ways to set up an Intercom automation to send post resolution survey in the messenger. We'll cover Intercom's own tools, check out some third-party apps, and then dive into a smarter, AI-driven method.
What is an Intercom automation to send post resolution survey in the messenger?
Simple enough, it's an automatic feedback request that pings a customer right after a support agent closes their conversation. Think of it as a quick "how'd we do?" at the end of the chat.
The whole point is to measure Customer Satisfaction (CSAT) for that one interaction. It helps teams see how agents are doing, spot places to improve, and get a feel for how customers are feeling in the moment. Usually, this is set up inside Intercom with an automation rule that kicks in when a conversation is marked as "closed."
Method 1: Using Intercom's native conversation ratings
Intercom has its own built-in feature for this, called "Conversation Ratings." It's the quickest way to get started and comes with most of their plans. It's a solid first step, but honestly, it has some big limitations, especially if your team is growing and needs more than just a simple emoji rating.
How to set up native CSAT surveys in Intercom
Setting this up is pretty simple. You definitely don't need to be a developer to get it working, which is nice.
According to Intercom's own guide, it only takes a few clicks:
-
Go to "Fin AI Agent > Simple automations" in your settings.
-
Find the "Ask for conversation rating (CSAT)" option and switch it on.
-
From there, you can set a few basic rules, like not sending it for outbound messages or deciding how long customers have to leave or change a rating.
If you want a little more control, you can also tweak these surveys using Intercom Workflows, but even that approach has its own set of constraints.
A view of Intercom's Fin AI Agent settings, where users can configure simple automations like CSAT surveys.
Limitations of Intercom's built-in survey tool
The setup might be easy, but you'll probably hit a wall with it sooner or later. Here’s where Intercom’s built-in tool starts to feel a bit cramped:
-
You're stuck with one question type: The native tool is all about CSAT, using that standard emoji rating scale. If you want to measure other things like Net Promoter Score (NPS) or Customer Effort Score (CES), you'll have to install a separate app.
-
The automation is pretty basic: You can set simple triggers, but the logic doesn't go very deep. You can't easily build complex rules based on what was said in the chat, a customer's history, or other important context. The feedback ends up feeling disconnected from the quality of the help they received.
-
You don't get much context: The reports will show you CSAT scores, which is a start. But they won't tell you anything about the conversations that led to those scores. You'll know a customer is unhappy, but you're left digging through tickets by hand to find out why, which just isn't practical as you grow.
Intercom's reporting dashboard displaying CSAT scores, highlighting the kind of data available natively.
This is where a smarter system really makes a difference. For example, a tool like eesel AI gives you a fully customizable workflow engine. You can build super-specific automation rules that are way more powerful than a simple "send survey on close" trigger.
Method 2: Integrating third-party survey tools
To work around the limits of Intercom's built-in feature, lots of companies connect specialized survey tools. These apps give you a lot more freedom with the kinds of surveys you can send and the reports you get back.
Popular third-party survey tools
If you browse the Intercom App Store, you'll find plenty of options. Tools like Zonka Feedback, Survicate, and Retently are popular picks. They all claim to give you more insight than Intercom's basic tool, but each one has its own way of doing things.
| Feature | Zonka Feedback | Survicate | Retently |
|---|---|---|---|
| Survey Types | NPS, CSAT, CES | NPS, CSAT, & more | NPS, CES, 5-Star |
| Integration Method | Messenger App, Email, Custom Bots | Messenger App, Email | Messenger App, Email |
| Key Advantage | Offers agent-based reporting. | AI-powered survey builder. | Strong follow-up automation. |
| Pricing Model | Starts from $49/month. | Custom pricing, contact sales. | Starts from $49/month. |
The hidden costs of third-party tools
While these tools do add more power, they also bring a new set of headaches you might not see coming.
-
Tool sprawl: You're adding another subscription to pay for, another dashboard to check, and another piece of software for your team to learn. Your feedback data lives in one place while your conversation data is still in Intercom. This creates annoying silos that make it tough to see the full customer story.
-
Survey fatigue: We've all been there. Customers are constantly bombarded with feedback requests. Sending a formal survey after every single interaction can kill your response rates and, frankly, just annoy people.
-
Always playing catch-up: These tools are still reactive. They’re good at telling you about a problem after it's already happened. They don’t help you understand the root cause from the conversation itself or, more importantly, stop the issue from happening again.
Ideally, you want a solution that fits in so well it doesn't feel like yet another tool. eesel AI, for instance, plugs right into your helpdesk. It learns from all your past conversations and pulls your knowledge sources together. With this kind of setup, you might not even need a separate feedback tool, because the AI is smart enough to figure out the quality of an interaction on its own.
Beyond surveys: Using AI to analyze customer satisfaction
Alright, this is where we get into the next level of customer feedback. Instead of just asking if a customer is happy, what if you could tell just by reading the conversation? And what if you could use that info to make your whole support system better?
Why traditional post-resolution surveys fall short
Look, old-school surveys are better than nothing, but they have some major flaws that are tough to overlook.
-
Low response rates: Only a tiny fraction of your customers will ever fill out a survey. That means you're making big decisions based on feedback from a small, and often skewed, group of people (usually the really happy or the really mad ones).
-
Lack of context: A frowny face emoji doesn't tell you the whole story. Was the customer upset about the agent's tone? A frustrating company policy? A bug in your app? To find out, you have to go back and read the ticket manually, which is impossible to do for every single one.
How AI agents analyze customer sentiment automatically
This is where AI platforms like eesel AI do things differently. Instead of waiting for survey responses, eesel's AI Agent digs into thousands of your past support tickets. It learns your company's tone, gets to know your common problems, and figures out what a "good" resolution actually looks like for you.
From there, the AI can analyze the feeling, the language, and the result of a conversation to gauge customer satisfaction, all without sending a single survey. It can automatically spot moments of friction, confusion, or frustration. You basically get a 100% "response rate" because every single chat is analyzed for quality, not just the handful where customers take the time to click a button.
This workflow illustrates how an AI can draw from multiple knowledge sources to analyze conversations, going beyond a simple survey.
Improving support quality with AI insights
But the real magic is what happens next. A tool like eesel AI doesn't just hand you a satisfaction score and call it a day. Its reports show you trends and point out gaps in your knowledge base.
For example, if the AI sees that a bunch of unhappy customers are all asking about the same thing, it can draft a help center article to tackle that exact problem. This helps stop those tickets from ever being created. You shift from just putting out fires to actually fire-proofing your support.
Move beyond basic post-resolution surveys
So, we've walked through three ways to handle feedback in Intercom: the simple built-in ratings, the more powerful (but separate) third-party tools, and the smarter, all-in-one AI approach.
Sending a post resolution survey in the messenger is a decent start, but the real future of customer support is about understanding the conversation, not just slapping a score on it. Lasting improvements come from digging into what's actually being said and using that knowledge to make your whole support team smarter.
The goal isn't just to collect more data. It's to build a system that fixes problems well and learns from every single conversation, so you're constantly getting better.
Ready to move beyond post-resolution surveys?
With eesel AI, you can plug right into Intercom to analyze past tickets, automate resolutions, and get insights that go way beyond a simple CSAT score. You can get started in minutes, not months, and even test how it would have performed on your old tickets using our simulation mode.
Frequently asked questions
This automation automatically sends a feedback request to a customer immediately after their support conversation is closed. Its purpose is to quickly measure Customer Satisfaction (CSAT) for that specific interaction, helping teams understand agent performance and customer sentiment in real-time.
You can set this up by navigating to "Fin AI Agent > Simple automations" in your Intercom settings. Locate and enable the "Ask for conversation rating (CSAT)" option, then configure basic rules such as exclusion for outbound messages or rating window duration.
The native tool is limited to a single question type (CSAT emoji) and offers only basic automation rules, which can't adapt to complex conversation contexts. Crucially, it provides little insight into why a customer gave a particular rating, requiring manual investigation for deeper understanding.
Your team should consider third-party integrations when you need to collect more diverse feedback types like NPS or CES, require deeper survey customization, or need more advanced reporting capabilities than Intercom's native feature provides. Be mindful, however, of potential "tool sprawl" that can occur.
An AI-driven approach analyzes your existing support conversation data (like past tickets) to infer customer sentiment and satisfaction without sending explicit surveys. This method provides a "100% response rate" by evaluating every interaction, offering more comprehensive and contextual insights than a simple score.
Moving to an AI solution allows your team to identify the root causes of customer issues, spot emerging trends, and proactively suggest improvements to your knowledge base. This shifts your support from reactive problem-solving to a more proactive, preventative approach, constantly learning and improving.






