Movie and Event Ticketing for Alexa+

Overview

Solution

Movie and event ticketing started as two separate initiatives, but auditing both flows showed the same core steps: discovery, showtimes, seats and checkout. So I designed one flexible template for both, handling everything from a neighborhood movie to a stadium concert.

Better discovery, smarter seat selection, checkout that gets out of the way.

Alexa+ uses location, event data and past preferences to suggest relevant shows and pick great seats automatically, taking people from discovery to checkout in seconds.

Impact & Key Achievements

• Nearly 3% organic conversion, on par with Ticketmaster's own web and mobile benchmarks.
• Proved that high-friction purchases like seat selection can move out of a mobile app and into conversation.
• Became a reusable framework as new ticketing partners come on board.

Event ticketing is live with Ticketmaster, and movie ticketing launched to beta customers with 2 partners in September 2026.

Background

Voice handles simple requests well, but ticketing is detail-heavy: showtimes, prices, venue maps. The challenge was making a high-stakes checkout feel fast and natural.

Research and Testing

Methodology and Approach

Partners initially pushed for a fully automated seat pick to speed up checkout. My hypothesis: seat preferences are too personal to automate completely. To test it, I led a three-part study:

Part 1: How people pick seats

• Groups beat views: over 50% of patrons choose adjacent back-row seats over split front-row upgrades.
• Context shifts choices: people spread out in empty theaters, but prioritize aisles once a venue is over 80% full.
• Tiered pricing didn't help: AMC's abandoned Sightline program added friction without changing behavior.

Part 2: Testing a live prototype

I built a conversational web prototype and ran an unmoderated study with 64 moviegoers, comparing AI-only seat picks against AI picks plus a seat map, with plenty of seats and with few left:

What we learned

• The map is a safety net: with the AI's pick shown on a map, 72% bought exactly those seats.
• People reject a black box: without the map, 68% rejected the experience and satisfaction fell from 4.44 to 2.64 out of 5.
• People want the final say: about 1 in 4 changed their seats, no matter how good the pick was.
• They expect the map to listen: people tried steering it by voice, like "Get us closer to the screen."

Part 3: Ticketmaster RITE testing

Three RITE cycles with 8 frequent event-goers on next-gen Alexa, focused on large stadium maps and live event seating.

What changed

• Section dots caused "view anxiety": we added price-by-location overlays and a mid-range price sort.
• Cart edits felt risky: half of participants feared losing held seats, so held seats now stay locked while adding more.
• Voice steering needs visual feedback: "Show me sections closer to the stage" now instantly highlights the map.

The takeaway

People don't want AI to decide for them; they want AI to help them decide. Pairing a smart default with a visible, editable map is what earned their trust.

Design Process

Cross-Org Collaboration

I partnered with product, engineering, design systems, legal, business development, marketing and QA, and led regular syncs with Fandango and Ticketmaster on scope, data sharing and live inventory.

Key Decisions

01

Seat Recommendation Engine

A custom algorithm scores seats on view angle, distance, party size and past preferences, instead of relying on partner feeds that surface one pick and randomize the rest.

Pros

• Consistent, high-quality picks across Fandango and Ticketmaster.

Cons

• Complex scoring logic to build and maintain.

02

Interactive Seat Maps

A full venue map with the AI's picks highlighted: lightweight for movie theaters, guided for stadiums.

Pros

• A visual safety net that drove 72% acceptance in testing.

Cons

• Takes real screen space, and needs logic for words like "closer" or "better."

04

Unified Commerce Pattern & Architecture System

One interaction pattern for movies and live events, adapting the details to each.

Pros

• A familiar flow for users and consistent logic for design and engineering.

Cons

• Layouts had to scale from a 100-seat theater to a 60,000-seat stadium.

Final Design

Core Customer Experience

Recommendation engine

Behind the picks is a deterministic scoring system with a conversational fallback.

Priority order

1. Explicit requests win: "I need aisle seats" overrides everything else.
2. Past habits anchor the search, for example someone who picks aisle seats over 70% of the time.
3. Accessibility is a hard requirement, never a point adjustment.
4. New users start with center seats 60–75% back from the screen.

Seat scoring

To evaluate available inventory programmatically, seats are calculated against a multi-variable value matrix:

Tie-breaker rule: When two options yield equal scores, the engine selects the row closer to the back of the auditorium.

Conversational fallback

• Explains every pick: "Center seats were full, so I found the next best option in row J."
• Backs off when rejected: after two rejections, Alexa hands control back to the user.

Next Steps

Next, I'm exploring price tracking across vendors, an event-day companion with parking and entry info, and sending tickets to friends by voice.