Introducing AI Insights

Before starting this lesson, make sure the following is in place:

• Fin is live and handling conversations: You should have Fin deployed and actively resolving customer queries across your channels.

In a traditional support operation, we were always working from an incomplete picture, relying on manual tagging, CSAT surveys that maybe ten percent of our customers actually filled out, and reports that were telling us what's happening weeks ago, not what's happening right now, which meant we were always making decisions based on a small sample of data. But with a FIN first setup, this small sample turns into complete visibility. This is because FIN classifies every conversation it touches, even the ones it hands off to your team. This means every single interaction becomes a data point, which in turn powers a set of tools designed to help us understand what's happening and keep improving. Let's take a quick look at these tools. So first off, there's standard reporting with all the table stakes features you'd expect, including the ability to drill into the conversations behind any metric. But here's where things start to get interesting. Every conversation gets an AI generated quality rating, not just the ones where a customer filled out a survey. We can see the rating here on this conversation and then zoom out and track it across our whole support operation. Now this is called the CX score, and this is the thing that turns thousands of conversations into one clear signal. Moving on from that, we've got recommendations, which flag things like content we should add or update. So at all times, the system is basically showing us the next best steps to take. Then there's topics explorer and trends. Now topics explorer groups every conversation into themes. So when we want to understand what customers are talking about, we can use this to investigate. Trends is the opposite. It's a weekly report on what's changing in our support operation. What's up, what's down, what's new. And it's delivered proactively without us having to go looking for it. And lastly, there is monitors. Honestly, this is a feature that our customers go crazy for. It's always on QA that allows us to review both Fin and our team's performance, enabling us to create monitors, scorecards, and alerts, which help us spot and react to issues in real time. And that's the difference with an AI first setup. We're not piecing things together from a sample anymore. We actually get to see the whole picture. And in the next few lessons, we're gonna dig into these tools in a bit more detail.