A festival that had done this before, and a year that felt different
The operator runs several festivals a year. This one is a major multi-day music event with tens of thousands of guests, and a lean team behind it. Guest support was effectively one person's responsibility, layered on top of everything else that goes into producing a festival.
They had run this event before. They knew the shape of the week. The days before gates and the first day on site are when guest questions arrive in waves, the same questions over and over, faster than a small team can keep up with.
This year was different. With ASQR live on guest support, the volume that used to define that week mostly handled itself. The team got through the season without adding a single person to support, and the operator reported a clear drop in support volume compared to the prior year's edition of the same festival. Same event, fewer fires, same lean team.
The AI carried the volume that used to carry the week
A festival like this fields thousands of guest inquiries across a single event cycle. Most of them are not complicated. They are the where-is-it, when-does-it-start, can-I-bring-this questions that repeat by the hundred and quietly consume a support team's week.
On live chat, ASQR took on that volume directly. Around 84 percent of chat conversations were handled end to end with no human involved at all. The team's attention was freed for the conversations that actually needed a person.
The AI was strongest exactly where guest volume is highest and the questions are factual. Event details, order status, ticket transfers, parking, day-of logistics. Across those high-volume topics it answered correctly the vast majority of the time, often in the 85 to 95 percent range. These are the asks that, left to a small team, turn into a wall of repetitive replies during the busiest week of the year.
It held up when the weekend hit
The real test of any support setup is the surge.
During the busiest 48 hours of the event, the day before gates and opening day, the AI's resolution rate did not dip. It edged up on the chats it handled on its own. Guest sentiment across the run stayed calm. The system did not crack at the exact moment a live event could least afford it to.
It closed real gaps on the fly, mid-event
No two festivals ask exactly the same questions, and you cannot pre-write answers to questions you have not seen yet. What matters is how fast you can close the gap once it appears.
Days before gates, questions about parking lot addresses, nursing stations and re-entry policy started to climb. Because ASQR classifies conversations by intent as they come in, the trend surfaced right away instead of waiting for a post-event report. The answer went into the bot the same day, and from that point these questions were answered straight from the AI.
A question the system could not answer in the morning became one it answered confidently by the afternoon. That is the difference between a static FAQ and a system that learns while the event is still in front of you. (More on how that closed-loop learning works.)
Support that became operational intelligence
The most valuable thing the platform did during event days had nothing to do with deflection.
Because every conversation is sorted by intent and outcome as it arrives, the support queue stopped being a flat pile of messages and became something the team could actually read in real time. They could see what guests were really asking, where the knowledge gaps were, and which conversations carried the most weight. A plain inbox hides all of that. It just shows you an unread count.
It even surfaced guests who were trying to spend more, asking about upgrades and transfers, so those moments stood out instead of getting buried in the rush. That is the difference between closing tickets and surfacing operational intelligence. The guest experience becomes a feedback loop the operator can act on, not a black box they clear and forget.
A lean team that never had to grow
The clearest signal in the whole season was something that did not happen. Nobody had to be hired.
The volume that used to define the week was mostly absorbed by the AI. The support pile-up stopped being something to brace for. The person who owned guest support got their time back for the rest of the job, and the operator credited the deflection for the drop they felt from the year before. For an operator deciding where to put resources next season, that is the comparison that matters.
It worked well enough that the operator and our team are already setting up next year's deployment and extending the platform to more of their events.
If you run live events, the lesson is not that AI replaces your team. It is that a helpdesk built for this industry lets a lean team carry a workload that would normally force new headcount, while turning every guest conversation into something you can learn from. That is what a guest intelligence platform is for, and it is why we believe the experience is everything.
See how ASQR handles the volume and surfaces the insights for live events. Explore the platform or book a 20-minute demo.
Ready to turn guest support into guest intelligence?
See how ASQR helps live events organizations understand their guests better.