What’s this template for? This is a step-by-step guide for creating a Custom Bot that triages conversations in a platform with different user types. This example is based on an education platform, with organizations, teachers, and students. A similar method would also suit ‘marketplace’ apps with vendors and customers.

“EdShare” our example platform, doesn’t offer the same level of support for all types of users. In some cases, no chat support at all. Rather than limiting which users can get in touch, the Custom Bot gives all users a welcome, and guides them towards the best resources for getting support, whether that is from another user of the platform, or self serve content.

As an added bonus, the Custom bot automatically tags the conversations, so you can report on why people are looking for support in the first place.

Identify your users

The first step here is to identify who is writing in, so your welcome message should offer all the available possibilities. In this example, there are two types of teachers (independent, or organisation) and students:

Tip: If your users can belong to more than one of these groups, you should encourage them to identify with whichever is relevant to the issue at hand. 

Since you don’t offer chat support for students, the composer is disabled. This means everyone must select from one of the available options:

Clarify why they’re getting in touch

The paths for teachers here are quite similar to begin with, but there’s an important distinction in the final steps.

Independent teacher:

  1. Offers common reasons for getting touch like “reporting a bug”, “feature request”, “help with my account” or “something else”. 

  2. Tags the conversation based on the previous response.

  3. Asks them to share details on why they need help.

  4. Assigns the conversation to your support team. — They’ve got a head start on the conversation, as they have specific information up front, and know this is a user they should be supporting.

Teacher at an organisation:

  1. Offers common reasons for getting touch like “reporting a bug”, “feature request”, “help with my account” or “something else”. 

  2. Tags the conversation based on the previous response.

  3. Asks them to share details on why they need help.

  4. Assigns them directly to their account owner.

  5. Marks the conversation as priority.


The student path is a little more nuanced than the teacher paths, because even though you don’t offer chat support for students, there will be situations you want to be informed about. You’ll also be aware of common confusion that students face, and should offer self service content to assist.

  1. I have a question about my course  — Questions about course content are likely the most common choice, and are handled by teachers, so you want to get this out of the way up front.

  2. I need help using the platform — This path leads to the common causes of confusion. You can keep things general with a link to your Help Center (or the Article Search app), or offer more paths for specific questions.
    Tip: The Custom Bot is doing the work here, so more detail doesn’t mean more time spent, just a better experience for your users. 🤩

  3. My grades aren’t showing — This is an example of a bug that you’re aware of, and would like to gather more reports, so you can notify those affected.

End each path with a next step

None of these conversations are going to be assigned to the support team, but that doesn’t mean they won’t be supportive. This approach means that all your users have a better experience on the platform.

I have a question about my course:

Use this as an opportunity to set expectations with the student, let them know what you do and don’t offer chat support for. Then, ensure you point them towards where they can get more support. In this case, from the teacher:

Tip: If you can’t link to their next step directly, include a GIF or screenshot with clear instructions on where to go.

I need help using the platform:

  1. Offers common areas where students might need help like “Viewing my grades”, “Downloading course content”, “Viewing courses on mobile” or “something else”. 

  2. Tags the conversation accordingly — If you see one area coming up often in your conversation tags report, you can pass it on to your product team, or provide more detailed documentation.

  3. Shares an Article with the Article Inserter App that answers their question.
    Tip: It’s a good idea to include a link to your Help Center’s homepage too, along with a reminder that they should check there first if they ever run into trouble.

  4. Closes the conversation.

My grades aren’t showing:

  1. Lets them know you’re aware of the bug, and working on it. Also provides a link to your status page.
    Tip: You could also share this with the status page app directly.

  2. Tags the conversation with a specific tag for this bug, like: “bug report - no grades”.

  3. Closes conversation.

When you fix the issue, you can easily follow up in these conversations to inform anyone who ran into the issue. The bot handles the tricky part, and you get to share the good news 😉

See this Custom Bot in action

💡 Tip

Need more help? Get support from our Community Forum
Find answers and get help from Intercom Support and Community Experts

Did this answer your question?