Why we built AI into ticketing, and how we did it

When vivenu's board first pushed the company toward AI, the CEO wasn't sold. Simon Hennes told a room of sports and entertainment executives at the vivenu AI Summit in New York that the first use cases he saw were customer support bots. Software replacing the people who talk to customers. "We didn't want to do that," he said. "We want to be close to our customers. We want to learn from them, be consultative, be proactive." For a company that flies out to run workshops with its organizers, automating the conversation away felt like the wrong direction.

AI does not have to replace the parts of the work you value. It can take over the parts nobody wanted in the first place and give you your time back. "It can elevate what you're doing well," Simon said. For vivenu, that was proactive customer consulting. For an organizer, it is the fan relationship.

This is the long version of why vivenu built AI into ticketing, what the team refused to compromise on, and how the vivenu AI Assistant works.

Every entertainment organization has more ideas than hands

Two or three years ago, the most common reason a prospect gave for not switching to a platform where they own their data had nothing to do with the platform. Organizers liked the data ownership and the open ecosystem. Then came the sentence that ended the conversation: we don't have the internal team to do anything with it.

That was true, and it is still true. "Every entertainment organization is understaffed. Full stop," Simon said. "So many great ideas in those organizations, so much creativity. We just can't live it out because we don't have the resources." Meanwhile, live entertainment competes for disposable income with Netflix, travel, and every other night out, and it competes on relevance. A fan who gets a newsletter with nothing in it for them notices, the same way anyone notices a home screen full of films they would never watch.

AI helps both. It gives you the necessary speed of analysis and execution, and in doing so it allows you to focus more on winning the experience game. Customer segmentation, campaign copy, reporting, pricing adjustments: the work that needed a department can now be done by a small team asking questions. Simon was explicit that this is a growth story. Entertainment is growing, and the goal is higher revenue for organizers through higher value for fans. "Higher revenue, but in a meaningful and highly relevant way."

Four rules to make sure vivenu AI works to solve those problems right away

vivenu has been API-first from the start, and the first customer connected their own AI agent to the vivenu API in early 2025. This means, technically speaking, we could have left solving these problems to our organizers and the Anthropics of this world. So why did we still build AI into the platform at all?

Because API access gets an agent to the data. Understanding what the data means is the hard part. Jens Teichert, CTO and Co-Founder of vivenu, put it this way at the summit: "This industry is so nuanced, so complex. You can only come up with reasonable insights if you understand the business, the ticketing industry, and how the product works under the hood. it needs very, very specific context." Hence, before the team implemented a single feature, it set four rules:

1. Grounded in your business. Language models are brilliant and clueless at the same time. The assistant has to know ticketing, the vivenu platform, and the organizer's own setup inside and out.

2. Useful on day one. AI capability is racing ahead of AI adoption, and most organizations don't have the budget for a dedicated context engineering team, helping them shape a knowledge/context database. Whatever feature we build, it has to be prompt-ready.

3. Secure by inheritance. The assistant respects the roles and permissions each user already has.

4. Your data stays yours. It is never used to train a model, and it never becomes someone else's competitive advantage.

How the vivenu AI Assistant works: a ticketing knowledge base first, then a data engine

The first months of the project produced no visible feature. The team built a context corpus instead: how every part of the vivenu product works, how ticketing works as an industry, and how a given organizer runs its business, from the way events are configured to the order history, whether it is an attraction with recurring shows or a club with a season.

On top of that sits a data engine that lets the assistant query all of an organizer's live data, well beyond revenue by week. The difficult part was definitions. "Simple things like the definition of revenue vary from organization to organization," Jens said. "We had to build a context layer so that when an agent asks for a metric like revenue, we can guarantee it is calculated the right way." Handing an agent the database and letting it write its own queries produces confident answers on the wrong scope, which in ticketing is how a good-looking report ends up wrong.

The assistant also remembers. Your definition of "VIP," the metrics you check most, how your team names things. Every organizer gets its own assistant working only on its own data, so nothing learned about one business is shared with or learned from another. And it runs inside the same security perimeter as the rest of vivenu: SOC 2 Type II, PCI-DSS, GDPR, and ISO 27001.

It launched in July 2026 as the vivenu AI Assistant, the live entertainment industry's only AI assistant trained specifically for ticketing.

What the AI Assistant does today: analyze, segment, campaign, repeat

In practice it closes a loop that used to take a week. Ask how the renewal campaign is performing. Ask for every customer who bought a VIP season ticket last year and hasn't renewed. Ask it to build an email campaign for the top 500 of them, ready to send. After the send, ask how many renewed.

Spenser Ayres, Associate Athletics Director for Ticket Sales & Operations at Stanford Athletics, was among the first to use it: "I can ask questions in plain language about ticket sales, revenue, and customer behavior, and get answers pulled directly from our live data in seconds. It has changed how quickly I can turn data into decisions."

The most used case so far is the least glamorous: customized reporting. Every stakeholder wants a slightly different cut of the same data, and teams spend hours producing it by hand. The second is product knowledge, from "where is the setting that does this" to "which feature would help me grow revenue on this event series." The assistant answers because it knows the platform, its API, and its changelog.

What comes next: actions, the vivenu MCP server, and autonomous workflows

Reading data is the first layer. The entire vivenu API is becoming a set of tools the assistant can use, more than 100 of them: update events, set prices, reallocate inventory between channels, hold seats for a sponsor, create promo codes. The agent executes, you approve. "In ticketing, a wrong input in a price field on the wrong event can easily cost you millions," Jens said. "So we want to make sure it's safe to use."

The same tools are available outside the platform through the vivenu MCP server, so teams can connect their own AI, whether that is Claude, ChatGPT, or a custom stack, directly to their ticketing data. Ask for five adjacent VIP seats for the next three home games and let the agent find them, move inventory from the box office allocation to the web shop, and hold them. The MCP server has the same capabilities as the in-product assistant. Which of them your organization allows, updating pricing for example, is a policy you set.

The step after that is autonomous workflows inside the platform: the daily wrap report that lands in the CFO's inbox at 9 a.m. as a PDF, described once in a conversation and scheduled from then on. Whole workflows that monitor, decide, and act inside the guardrails you set.

Where to start with AI in ticketing

Simon's sequence from the summit stage is the same one we argued for in our piece on personalization.

Own your data in a connected ecosystem. AI can only reason about what it can see, so the CRM record, the ticketing record, and the marketing record have to resolve to the same fan.

Make the signals clean and actionable. Every ticket, scan, and renewal, live and ready to act on inside the platform, without a data project in between.

Then let AI do the orchestration at scale: the segments, the campaigns, the reports, the price moves no team has the hours for. Stanford Athletics is already there on one of those levers, with a 24% increase in revenue per basketball ticket from seat-level dynamic pricing set by AI, season over season.

The summit was held in vivenu's SoHo office, on purpose. vivenu came into ticketing without a ticketing background and learned the business by building alongside its customers. The AI work follows the same pattern: ship, watch how an athletics department or a theater uses it, build the next layer. If there is a question about your own business you have always wanted answered, bring it. We'll run it on your own events. Book a demo