Agentic AI: Smarter Personalization, Stronger Relationships

Agentic AI: Smarter Personalization, Stronger Relationships

For a decade, "personalization" in ticketing has mostly meant one thing: a better email. Segment the database, pick a subject line, send it Tuesday at 10am, hope for a 2% click-through. The fan gets a message. Then the fan gets a login screen, a seat map, a cart, a checkout form, a captcha. Somewhere in that journey most of them leave.

That model is running out of road. Not because the data got worse — teams and venues know more about their fans than ever — but because the last mile never changed. Personalization stopped at the point of sending. Agentic AI moves it to the point of doing.

The numbers behind the shift

This isn't a forecast anymore. The behavior has already changed on both sides of the transaction.

Fans are already starting their journey with AI. McKinsey found that roughly half of consumers — across every age group, boomers included — now deliberately use AI-powered search when making purchase decisions. Adobe reported that 56% of US consumers used generative AI during the 2025 holiday season, up from 11% the year before, and Reuters put AI-influenced online sales on Black Friday 2025 alone at over $14 billion. In a July 2026 survey by Fractl, 70% of consumers said their use of AI for search had increased over the past year; only 4% had never used it.

And when AI sends them, they're ready to buy. Adobe's retail data shows traffic referred from generative AI sources to US retail sites grew 693% year-on-year over the 2025 holidays. More telling than the volume is the quality: by March 2026, AI-referred shoppers were converting 42% better than non-AI traffic — a full reversal from a year earlier, when they converted 38% worse. The shopper who arrives from an assistant has already done the comparison. They land pre-qualified.

The enterprise side is moving just as fast — but unevenly. Gartner expects 40% of enterprise applications to embed task-specific AI agents by the end of 2026, up from under 5% in 2025. Yet McKinsey's data shows that while 88% of organisations use AI somewhere, only 23% are scaling an agentic system — and Gartner warns that over 40% of agentic AI projects could be cancelled by 2027 where governance, observability, and ROI aren't nailed down.

Read those together and the picture for ticketing is clear: fans are already asking AI where to go and what to buy. The question is whether the club or venue has an agent on the other end that can actually finish the job.

From recommending to acting

The distinction that matters is simple. A recommendation system suggests. An agent completes.

When an agent is embedded in the channel a fan already uses — SMS, RCS, WhatsApp — the personalization isn't a nudge toward a website. It is the transaction. A season-ticket holder gets a message the morning of a match: two seats have opened up in the section they upgraded into last November, and the agent already knows their price ceiling from the three previous upgrades. They reply "yes." Payment runs on the card they've used before. Confirmation lands in the same thread. Total time: under a minute, no app, no password.

That's what "smarter" personalization looks like in practice. It's not a more accurate guess about what a fan wants. It's the removal of everything standing between the guess and the outcome.

Why this builds relationships rather than just conversions

There's a reasonable worry that more automation means less human connection. In ticketing, the opposite has tended to be true, for three reasons — and the retail conversion data above hints at why: fans who complete a task inside a conversation trust the next one more.

The channel is personal. A text thread is not a marketing inbox. Fans treat it more like a conversation with a friend or a concierge — and they hold it to that standard. An agent that operates there earns trust by being useful and disappearing when it isn't, not by being loud.

Memory compounds. Every completed interaction teaches the agent something a segment never could: this fan buys late, this one always adds a parking pass, this one returned tickets twice last season and would probably like a heads-up when a big game gets close. Over a season, the agent isn't personalizing against a persona. It's personalizing against a person.

Two-way beats one-way. Traditional CRM talks at fans. An agent talks with them. A fan can ask "can I move closer to the aisle?" or "what's the cheapest way to bring my kids on Saturday?" and get an answer that resolves into an actual ticket. That loop — ask, act, confirm — is the basic unit of a relationship, and until recently it required a human on the phone.

What "agentic" actually requires

The word is being used loosely, so it's worth being precise about what separates an agent from a chatbot with a nicer tone. An agent that can genuinely serve a fan needs four things working together:

• Intelligence — the reasoning layer that understands intent, context, and when not to act.
• Tools — live, write-capable connections into the ticketing system, payment rails, and CRM. Read-only integrations produce read-only personalization.
• Skills — the specific, well-tested workflows: upgrade, return, transfer, group booking, waitlist. Each one is a product, not a prompt.
• Knowledge — venue rules, pricing policy, inventory holds, the fan's own history. Without it, the agent is confidently wrong.

Missing any one of these and you get an experience that looks personalized until the fan tries to complete something. The relationship damage from a failed "yes" is worse than never having asked. That gap — between adopting an agent and actually scaling one — is exactly what the McKinsey and Gartner numbers describe, and it's why so many pilots stall.

The guardrails are part of the product

Autonomy without boundaries isn't a feature. The teams and venues doing this well set clear rules: what the agent can do without confirmation, where it must ask, and what it will never do. A fan should be able to tell the agent "don't offer me upgrades on weeknights" and have that stick.

The same architecture that lets an agent act on behalf of a real fan also needs to be able to tell when the "fan" on the other end isn't one. Personalization built on verified identity, device-native channels, and transaction history is inherently harder for bots to exploit than a public web checkout — which is one of the less-discussed reasons rights holders are paying attention.

What changes for teams and venues

The practical shift is from campaign thinking to conversation thinking. Instead of asking "what should we send this segment this week," the question becomes "what should this fan be able to ask for, and can we let them get it in one step?"

That reframing tends to surface a lot of previously invisible demand. Fans who never opened the app but would have said yes to a text. Upgrades that were never offered because the email went out before the seats released. Returns that turned into no-shows because the process was too much friction for a Tuesday night.

Agentic AI doesn't invent this demand. It just stops losing it.

The relationship is the moat

Anyone can buy a data platform. What can't be bought is a fan who has completed twenty frictionless transactions with your club through the same thread, who trusts that when a message arrives it's worth reading, and who has never once been asked to reset a password.

Smarter personalization gets you the first transaction. The relationship is what happens when the agent gets the next nineteen right.

Pogoseat builds vertical AI agents for sports and live entertainment, enabling full ticket transactions inside SMS and RCS.

‍Sources: McKinsey (State of AI, 2025; consumer AI search research, October 2025); Adobe Digital Insights (2025 Holiday Shopping Report; March 2026 retail traffic analysis); Reuters (Black Friday 2025); Fractl (July 2026); Gartner (enterprise agent forecasts, 2025).