Building a Prioritization System
Advanced AI-Driven Workflows for Complex Support Needs
🔥 Tired of manual conversation assignments?
Discover how to build a smart, flexible escalation system that lets your AI assistant route chats to the right person or team—automatically.
In this tutorial, you’ll learn how to transform your support workflow into a seamless, scalable machine using intelligent routing, chat flow switches, and user properties.
🚀 What You’ll Learn:
- 🧠 How to build a universal escalation flow for all use cases
- 🔀 Smart routing based on plan type, intent, or owner ID
- 🧩 When and how to trigger the escalation chat flow from others
- 📈 Real-world escalation examples that improve customer experience
- ⚙️ How to assign chats to specific agents with precision
🎓 Real-World Examples Include:
- Routing Free, Pro, and Enterprise users to the right support teams
- Directing sales inquiries straight to your sales team
- Assigning VIP customers to their account managers
📌 With this system, you’ll save hours of repetitive work and boost customer satisfaction.
👉 Ready to elevate your AI assistant’s capabilities? Watch now!
👍 Like, comment, and subscribe to keep learning smarter support strategies.
- 00:00 Introduction
- 01:35 Why Escalation Systems Matter
- 02:23 Step-by-Step: Building Your Escalation Flow
- 03:34 Assign based on Subscription type
- 07:00 Assign based on Customer intent
- 07:56 Assign to account's owner
- 09:00 What’s Next & CTA
Transcript
Auto-generated from the video captions; it may contain errors.
[Introduction]
Imagine this. Your AI assistant is handling conversation like a champ. But then comes the moment where a customer issue needs to go to just the right person or team. Whether it's a sales question, a subscription upgrade, or a support escalation, you need a system that's not just fast, but smart. Nice. That's exactly what we're building today. A prioritization and escalation system powered by your AI assistant. No more repetitive setups. No more manual assignments. Let's make escalations simple, smart, and scalable. All right, so here's the scenario where Aura, your AI systems, is already doing a great job answering FAQs and handling complex intents. But when it comes to escalation, things get get messy fast.
Why? Because different conversations might need to go to different people or teams based on for example the user type, the subscription plan or even the intent of the conversation. Manually building escalation system for every single scenario takes time. What next? But today I'm going to show you how to create one flexible escalation system that works for all your use cases. We will do this using properties like plant type or owner ID switches in your chat flows and reusable chat flows that can be triggered anytime installation is needed. By the end of this lesson, you will have a scalable system that prioritize conversation and gets them to the right person or team fast.
Very good. Let's quickly talk about why this
[Why Escalation Systems Matter]
matter. Not all conversations are created equal. A free tier user asking a general question about pricing might not need the same level of attention of an enterprise customer with an urgent issue, isn't it? But if your AI doesn't know how to prioritize, those conversation could end up in the wrong place or worst unsolved. With a smart visitation system, your AI can first escalate conversation based on user type or subscription plan. Route intents like sales inquiries or upgrade requests directly to the right team. Turn assign escalation to a specific person using owner ID or owner name. Therefore, your team stays focused on what matter most and your customers get faster, better support. Here is how to set it up step
[Step-by-Step: Building Your Escalation Flow]
by step. Step one, let's create a dedicated escalation flow. First we need to go to our chat loss over here and we're going to select AI recognize intent. Now we're going to call it escalation a human user is asking for a human. So this is already built in your AI assistant why you need an additional chat flow with an intent because then you can launch this chat flow from any other chat flow at any other stage. So we're going to create this at first and then we're going to see how to use it then later. Okay. So whenever there is an escalation now I want to redirect the people to the right team.
Right? So if you got one single team it's fairly easy. You just assign it to the customer support. But let's say you want to assign it to high priority whenever there is someone with a different kind of expectation or like a plan or let's say you want to assign to enterprises when you got the kind of different level of support. Okay, this is going to be uh a simple assignment but if we want to
[Assign based on Subscription type]
switch it taking in into consideration the kind of plans they've got now we can click on plus and then logic adding either a condition or a switch. Okay. So, switch is like you're testing a property against different values. So, over here I'm going to select plan. Remember that the plan type is going to be selected at that stage whenever the user start a conversation. Now, there is a link below in the description on how to understand the properties and how they work. But as for the sake of this lesson, the plant type would be something that differs from customer to customer. And let's say here the value would be pro or it would be enterprise or you could have I don't know basic or free.
Okay. Now when you save it you will get several different options. Now the pro if it's important for you you can assign them to the inferiority. Then enterprise would be assigned to the enterprises. That's correct. then basic and everything else can be assigned to support. So if you do want to select only these two, you can get rid of the basic and the free because then it will be just the end the else of this logic and you will be able to assign it to support as a back solution over here. So this would work out perfectly to create an assignment based on the type of the plan.
very same thing you can do it with the owner ID or uh the owner person name the property will have different name of the owners or the owner's ID and then you can assign it directly to that person now we're going to call this escalation route and we're going to publish it now why using a separate chunk flow is important for escalation because now for every single thing that can be assigned to one of these three teams based on these rules. Now you can instead of assign it directly to one of the teams, you can now use the launch chat flow over here and then you can select escalation rules.
So as you can see this is quite powerful because it's connecting the main flow to another flow that is going to escalate based on your route and the route might be super tailored with your own rules. Okay. So now we're going to save this. Now this is very specific for support. But whenever uh Aura is getting into a sales intent for example or a plan upgrade then this is going to be completely different because the escalation rules over here would be specific to this very intent. So there is not much more to do over here because already this intent is about upgrade and if this requires to go directly to the sales team your escalation will work flawlessly towards your conversation.
So this is the way to manage in an advanced way your escalations based on the plan or the the owner type and so on. Of course, you can have different shot flows, one for each kind of escalation routes and reuse them in different situations. So like for example, if you do have different flows to that needs to
[Assign based on Customer intent]
be redirected to the sales or the success team, then what you would do is basically creating another chat flow with that intent switching by for example account to owner ID, whatsoever is working for you. This keeps your workflow clean and saves hours of repetitive setup. Amazing. Now, we saw it that not all escalations are about plant types. Sometimes you need to prioritize based on the intent of the conversation. For example, a sales inquiry intent could be routed directly to the sales team as we've been seeing over here. Or an upgrade intent could go to an account manager. In these cases, the chat flow can directly assign the conversation to the right team or person without additional switches. This keep things simple when there is no need for complex logic. Of course, finally, let's talk about direct escalation to
[Assign to account's owner]
specific people. With the same concept of using properties, you can set up the escalation system to assign conversation directly to the owner ID of the customer or the owner name in your system. This is perfect for situation where each customer has dedicated account managers or support agent. When the escalation happens, the assistant assign the conversation directly to that person, ensuring personalized and efficient support. Now you can create one by clicking on plus and if you scroll down you will be able to find one of our templates and this is super simple. Um actually this is when a new conversation starts. So uh this will lead every new conversation to the switch and understand which is the contact owner of this person and then of course you can assign it to the very specific person in the team.
So this is quite simple again and it's a template you can use of course or you can make your own depending on your uh needs. With this system in place you can handle all these scenarios seamlessly. So with a smart
[What’s Next & CTA]
priorization system your AI system becomes more than just a support tool. It becomes a strategic part of your customer experience. By creating a dedicated escalation chat flow, using properties to define routing and leveraging owner assignments, you can handle escalations at scale without breaking sweats. Seriously. All right, that's it for this lesson on building a prioritization and escalation system. Now it's your turn to set it up, customize it for your needs, and watch your support team run. If you found this helpful, hit like, subscribe, and let's keep building smarter support systems together. I'll see you in the next one.

