Often in support tickets, we always know the big trends that cause the most contacts. However, it's difficult to action broad AI tags like Late Delivery, or Payment Failed. That's why it's important to break down these broad issues as much as possible using Discover, common phrases, and specific conversations to make big issues more actionable.

Identify rising or big issues

The first step is to choose which AI tag you want to break down further. For example here, we can see that Damaged Packaging is a top reason for contact and has started to rise in occurrence recently. A quick glance at Summary for the last week or month of conversations will give you an idea of which AI tags to dive into.

Use Discover to break down into the root cause

By going to Discover, you will see your list of AI tags with the highest mentioned at the top. By clicking on an AI tag, you will see a more detailed chart (see next step), the common phrases mentioned alongside the AI tag, and the associated conversations.

Explore common phrases for potential trends

To understand why Damaged Packaging has been mentioned, we can scroll through the Common Phrases that come up below the detailed bar chart for Damaged Packaging. As we can see below, 'Leaked' has come up 23 times in the same sentence where Damaged Packaging was also identified, and the amount of mentions have increased by 5% since the last week.

Clicking on these Common Phrases will show the specific conversations where Damaged Packaging and 'Leaked' were mentioned in the same sentence.

Highlight specific conversations

Now we can explore a more specific sub-set of conversations that are more likely to be relevant to your root cause analysis, and therefore help in decreasing the time spent on manual analysis. While going through the conversations, it can help to 'Star' any relevant ones to reference later.

Making your case

When reporting on these specific issues, we recommend presenting the overall trend of the AI tag such as Damaged Packaging so that the impact of the whole issue is felt; this will help answer why a problem needs to be fixed.

Then reporting on the specifics of the issue with real examples will help answer what can be fixed. When using real-life examples, teams will feel more able to act because they understand straight from the customer what went wrong.

Did this answer your question?