One of our favorite new features in Intercom is the ability to tag conversations with automated tag rules. While gathering feedback should be top of mind for every customer conversation, when you have hundreds or thousands of conversations a day, feedback gathering can be a difficult task to scale. With tag rules this becomes much easier.
Setting up tag rules
To get started simply jump over to your inbox, click automation, then choose tag rules. You need to think strategically about what kinds of feedback you’re hoping to capture with tag rules. Automation can be your best friend when it comes to efficiency, or your worst nightmare when it comes to data quality, so be careful to set up tag rules that won’t produce a bunch of false positives.
Examples of data to trigger tagging
Intercom lets you choose from attributes/events for message data, people data, or company data. While there are many options, you may want to start with one of these use cases to tag a conversation:
- Keywords or phrases in conversations
- Page URL when the conversation started
If it's a keyword/phrase try to think of common questions or feedback that could easily be identified as a specific type of feedback. For example, feedback about a Slack integration would be a great chance to set up tag rules.
A page URL could be useful too. Imagine you have a new beta feature or product that only exists on a certain URL string. You may like to see all feedback given on that page:
Once you gather all of this feedback you can report on conversation tag trends:
Taking this to the next level with Userfeed
While these are some good examples of how to get started with some basic tag rules, there are still some things you likely want to do when it comes to managing this feedback at scale:
- Report in a way that enables you to look at the data as well as the context. The chart above only tells you about the volume of tags instead of what was actually said in the convos or who all said those things.
- Tagging users, not just the conversation. This way you can create user segments in Intercom based on feedback. This is super important for messaging purposes.
- The ability for further categorizing of feedback
- A place to collaborate on the feedback
This is where Userfeed comes in.
As a quick reminder, Userfeed essentially allows you to create posts (ie. a specific feature request or issue) for which each post creates it’s own intercom tag (see below). When a conversation is tagged with a Userfeed generated Intercom tag, the conversation and the user in the conversation are tagged, and both are linked to that post in Userfeed.
Let’s look at the same examples. In the keyword example, imagine if you set up tag rules with a Userfeed tag.
Now all of these conversations and corresponding users are grouped within that Slack integration post in Userfeed:
Since the users were tagged as well, Intercom segments based on feedback are one click away where you can view data about those users, message them to send a survey, or close the loop on a feature update. Additionally, you’re able to collaborate both internally and externally about that specific feature request in one place with every important notification and conversation about this post tracked in that comments feed:
Lastly, it’s important to further categorize your feedback for high-level trends and the ability to filter through all the data for prioritization. Userfeed lets you push feedback into high-level categories called feeds. You can add another level of categorization with labels too.
In conclusion, Intercom tag rules are helpful in automating the gathering of more commonly suggested feedback. You still need to be careful you’re protecting data quality, and it’s important to continue to gather the more complex feedback in a bit more manual fashion. No matter how you gather feedback in Intercom, the most important thing is to treat each conversation as an opportunity to learn something, not a chance to improve your ticket closing rate. Ask questions. Then ask more questions. The more you talk to your customers, the easier it will be to gather quality feedback manually or with automation rules.
Interested in more content like this. Send us your email and we’ll keep you in the loop.
