What are Subtopics?
To obtain actionable insight sometimes you might need to drill down further into a Topic.
You can use the Subtopic feature to sub-divide any Topic into Subtopics by defining a set of keywords for a Subtopic similar to how you would create a Topic.
Just like with Topics, Subtopics are very flexible and can be modified, created, and deleted at any time. This means that if you come up with new ideas for sub-topics or keywords you can be tracking you can set those and see the matching mentions instantly – both for historic and incoming customer conversations.
Once you’ve got a sub-topic setup, the matching mentions will be automatically allocated to it. You don’t need to worry about manually categorizing mentions – Customer AI takes care of that.
How to create a Subtopic
- Open any Topic
- Under Subtopics, click Create
- Enter the Subtopic name
- Type in one or more keywords to define the Subtopic, confirming with Include each time
- Confirm your changes with Save Subtopic.
Tracking churn reasons
For example in a topic “Cancel Account”, you might want to create the following Subtopics to track potential churn reasons:
Each sub-topic can be defined by one or more keywords. For example, for “Pricing” you might want to set “cost”, “price”, “expensive”, “cheap”, “afford”, “budget” as trigger words.
This means that when people mention cancelling their account and any of the keywords that define “Cost”, all mentions will be aggregated in a separate tab alongside the mention count.
You can create a topic for tracking bugs defined by phrases like “bug”, “error”, “crash”. Then have a flick through the mentions to see if any patterns emerge. As you spot areas of your product or service that users are complaining about, you can create Subtopics to track each:
- Authentication (login, sign in, sign up, create account, access)
- Uploading (upload, import, add song)
- Syncing (sync, refresh, update song)
Monitoring common questions about a feature
Start off by creating a topic defined by a feature name. I will use an example feature “Playlists” from Spotify. As users talk about playlists they might mention specific questions to do with offline mode, sharing playlists, issues creating, etc. Each of these phrases is a good candidate for a sub-topic to help monitor various questions users ask about a feature.