How TravelJoy achieves 80% AI resolution, saving 193 hrs each month

TravelJoy had a problem.

They needed a way to manage customer service for their ever-growing customer base, 24/7, across the world, without hiring more people.

Their Zendesk account rep got in touch and offered a trial of their new Zendesk Advanced AI.

After trying it for a few months, they didn’t see the impact they had hoped for and so turned to My AskAI.

Since integrating their My AskAI customer service agent into their Zendesk Ticketing and Messaging platforms it has resolved over 24,500 support tickets and is currently deflecting 80% of their tickets, saving their team 193 hours every month.

Now you’re probably asking yourself: “how does TravelJoy do this?!” and “how can I do this for my SaaS?”.

Let’s take you through their setup and show you how.

TravelJoy AI customer service results on Zendesk at a glance

What does TravelJoy do?

First, we should probably explain a little about who TravelJoy are and what they do…

TravelJoy is an all-in-one platform for travel advisors that combines CRM, itinerary building, client messaging, and integrated payments to streamline trip planning from lead to booking.

It’s built for ease of use and time-saving workflows, helps advisors stay organized, automate routine tasks, and deliver professional, shareable itineraries to clients.

It is used by thousands of travel entrepreneurs and has facilitated over $1bn in annual spend.

Which helpdesk does TravelJoy use?

First off, their helpdesk - TravelJoy uses Zendesk as their helpdesk platform.

More specifically, They use Zendesk’s Ticketing and Messaging features when providing support over email and chat and had previously looked at Zendesk’s own AI solutions but weren’t impressed (see our comparison here).

So they installed My AskAI’s approved Zendesk Ticket app and Zendesk Messaging app.

Play video

The results were instant.

How did they train their AI customer service agent?

TravelJoy started off by just connecting website to give their AI agent a base knowledge of their common FAQs.

They supplemented this with further knowledge by connecting to their internal Notion workspace.

When did they decide to turn on ‘direct replies’ to customers?

They decided pretty much from day 1 that the fastest way for them to learn how their new agent would perform in their Zendesk Messaging would be to put it in front of their users and start gathering feedback.

In the worst case scenario, there figured there were always lots of ways the users could request to speak to someone and hopefully in some cases it would give a number of users faster answers.

They were pleasantly surprised to see that from go-live, due to their extensive collection of articles and tutorials, that they were able to see a ~76% resolution rate, beating the 24% they previously achieved with Zendesk’s AI agent comfortably. Alan Pugh, their Head of Customer Service, put the switch like this: "In comparison to Zendesk's AI agent, we're now achieving an impressive 76% AI resolution rate, versus just 24% before. The dramatic improvement has elevated the overall level of service."

As well as having their AI agent reply directly, they also have it drafting notes to tickets received via email.

Due to the nature of queries received via email, they decided it made the most sense to use My AskAI as a copilot there instead, giving their agents quick response on first contact using our “reply to only the first message” feature.

What was the biggest thing they did to improve their AI agent’s resolution?

But TravelJoy didn’t want to stop at 76%, they wanted to see how much further they could take their AI customer service agent.

So they regularly checked in on their Insights to see what people were asking, reviewed and fixed their knowledge gaps and created custom answers to make sure their agent had all the knowledge it could to answer questions.

Through this continual improvement process, they have managed to increase their average AI resolution to where it sits today, at around 80%.

How do they customize their AI agent setup to work for their business?

Every business manages support in their own way and needs a solution to flex to their needs.

Here are a couple of ways TravelJoy make My AskAI work for them:

Controlling their communication style

TravelJoy use our Communication style guidance feature to make sure that their AI agent responds in exactly the way they want, following their reply structures and using their brand tone of voice.

They also use Handover guidance to ensure conversations are handed over when they need to be.

Using AI to auto-tag and categorize their tickets

Previously TravelJoy were manually assigning reasons for contact to each ticket. Now they use our AI Tagging Zendesk app to auto-classify tickets on receipt saving their team the monotonous job of (inconsistently) categorizing each month.

What impact is their AI customer service agent having now?

Good question.

As a result of TravelJoy’s continual efforts and focus on improving their AI customer service agent, their AI agent is currently:

• Resolving 80% of tickets automatically each month across over 2,500 tickets (that’s over 2,000 chats per month their team doesn’t have to deal with directly).
• Saving their customer service team 205 hours per month (assuming a 5 minute per ticket handle time).
• Achieving a 86% AI CSAT score across those 2,700 tickets.

Where do they go from here?

TravelJoy are keen to increase their AI resolution still further, with a focus on auto-resolving tasks that currently take time out of their (human) team’s day.

Things like, sharing more information on card decline reasons and extending free trials.

They’re now looking at how our tools and tasks features can help them do just this, automating both simple and complex actions.

Watch this space!

View all case studies or get started with My AskAI.

Learn more about building an AI-first support operation in our complete guide.