Dynamic Pricing for Events & Attractions: How AI Is Transforming Ticket Sales

For 25 years, I’ve worked at the intersection of digital marketing, revenue, and technology. During that time, I’ve watched hotels adopt revenue management systems, OTAs evolve pricing algorithms, and marketing automation transform how rooms are sold. But there is one segment of hospitality that has historically lagged behind in revenue optimisation:

Events, attractions, shows, and ticketed experiences.

In Phuket and across Southeast Asia, I see it constantly. A theatre show, theme park, cabaret, island attraction, or cultural performance sets fixed ticket prices months in advance. Agents receive seat allocations. Pricing rarely changes. Demand fluctuations are handled manually, if at all.

Meanwhile, airlines, hotels, and OTAs have been using demand-based pricing algorithms for decades. That gap is now closing. Dynamic ticket pricing, powered by AI, is transforming how events and attractions think about revenue. And if you operate in this space, the shift is not optional.

What Is Dynamic Ticket Pricing?

Dynamic ticket pricing means adjusting ticket prices based on demand patterns, rather than setting one static price for all dates and all seat categories. At a basic level, this includes:

• Increasing prices as a show approaches sell-out
• Discounting early to stimulate demand
• Adjusting prices by seat type based on historical performance
• Releasing or restricting agent blocks based on pickup pace
• Modifying pricing based on seasonality and country mix

At a more advanced level, it involves:

• Forecasting final capacity weeks in advance
• Predicting demand curves per showtime
• Identifying underperforming segments early
• Reallocating inventory between direct and agent channels
• Optimising add-on revenue (transport, F&B, VIP upgrades)

This is where AI becomes powerful.

Why Static Pricing Leaves Money on the Table

Let’s look at a typical example. Imagine a 600-seat show running nightly in Phuket.

Ticket pricing:

• Standard seat: 1,200 THB
• VIP seat: 1,800 THB
• Child discount fixed
• Agent commission fixed
• Price unchanged year-round

In the high season, certain dates sell out 5–7 days in advance. In the low season, some shows operate at 40–60% capacity. 

Static pricing means:

• You underprice high-demand nights.
• You fail to stimulate demand early in slower periods.
• You cannot optimise seat mix dynamically.
• Agent blocks may sit unused.
• Upsell revenue is not forecasted or optimised.

Hotels would never run like this. Yet many attractions still do.

The Demand Curve Model: The Core Logic Behind AI Pricing

Dynamic pricing is not guesswork. It is built around demand curve modelling. A demand curve for a showtime typically looks like this:

• Strong early sales for peak dates
• Moderate mid-window bookings
• Last-minute spikes from agents or walk-ins
• Distinct patterns by country of origin
• Different pacing for weekday vs weekend shows

With historical data, we can model:

• Days-to-event vs cumulative tickets sold
• Pickup velocity compared to last year
• Expected final capacity at any given point in time

For example:

If 10 days before a show, historical data indicates:

• On average 65% capacity is sold
• Final average sell-out rate is 95%

But today, you are already at 80% capacity 10 days out, that is a strong indicator of excess demand. An AI system would flag: “Showtime likely to exceed historical final capacity by 8–12%. Consider price increase for remaining premium seats.” Without modelling, that opportunity is invisible.

AI Ticketing System vs Traditional Ticketing Software

Most ticketing systems focus on:

• Seat maps
• Agent allocation
• Payment processing
• Reporting
• QR scanning

All important. But they are operational tools, not revenue intelligence tools.

An AI ticketing system adds:

1. Demand forecasting per showtime
2. Agent performance scoring
3. Dynamic price recommendations
4. Add-on purchase probability modelling
5. Revenue risk alerts
6. Channel efficiency tracking 

This transforms ticketing from an administrative function into a revenue engine. At The Percentage Company, when we design ticketing infrastructure, we are not just thinking about selling seats. We are thinking about:

• Yield per seat
• Yield per showtime
• Yield per agent
• Yield per country segment
• Yield per add-on

That mindset shift changes everything.

Real-World Logic: How Demand-Based Pricing Works in Practice

Let’s break down a simplified example.

Scenario A: High-Demand Show

• 14 days before event
• 75% seats already sold
• Historical average at this point: 60%
• VIP seats almost gone
• High percentage of bookings from Australia & Europe (higher-spend markets)

AI Insight:

• Demand running 15% ahead of normal pace
• High-spend country mix
• Low cancellation history for this show type

Suggested Action:

• Increase remaining VIP seat price by 10%
• Reduce agent commission incentive on last 50 seats
• Promote premium add-ons (e.g., dinner, backstage package)

Revenue Impact:

Even a 10% price adjustment on 100 remaining seats can significantly increase revenue, without increasing marketing spend.

Scenario B: Underperforming Show

• 7 days before event
• Only 35% capacity sold
• Historical average: 55%
• Agent block utilisation low
• Strong domestic segment bookings (price-sensitive)

AI Insight:

• Weak pickup trend
• Low conversion on premium seats
• Excess unsold standard inventory

Suggested Action:

• Offer early bird-style temporary discount for specific seat zone
• Release additional seats to top-performing agents
• Bundle with transport or add-on incentive
• Trigger remarketing to previous attendees

Instead of panicking 24 hours before the show, the system detects risk early.

Agent Management: The Overlooked Revenue Lever

In many Phuket attractions, agents control significant seat inventory. The problem? Most operators do not track:

• Revenue per agent
• Cancellation rate per agent
• Seat block utilisation rate
• Late release patterns
• Add-on conversion per agent

An AI layer can score agents based on:

• Revenue yield
• Deposit reliability
• No-show frequency
• Upsell performance

Over time, you can:

• Allocate larger blocks to high-performing agents
• Reduce blocks for underperformers
• Adjust commission tiers based on yield contribution

That is event revenue management, not just ticket distribution.

The Add-On Opportunity

Dynamic pricing is not only about ticket price. In many shows and attractions, add-ons include:

• Transport
• Dinner packages
• VIP upgrades
• Merchandise
• Drinks
• Photography

If historical data shows:

• Guests from certain countries purchase transport at 70% rate
• Others at only 25%
• Certain seat types correlate with higher F&B spend

AI can rank add-ons per booking context. Instead of showing all extras equally, the system prioritises:

• High-probability add-ons first
• Premium packages for high-value segments
• Bundles for price-sensitive markets

Increasing add-on revenue by 10–15% often has a greater margin impact than raising ticket prices.

Why This Matters for Hotel Owners

You might ask: “I run a hotel. Why does this matter to me?”

If your property includes:

• A dinner show
• A beach club
• A theme experience
• A ticketed attraction
• A cultural performance
• A spa package with fixed time slots
• An activity program

You are operating event inventory. The same dynamic logic applies. Hotels have embraced revenue management for rooms. Event operators are now catching up.

The operators who adopt AI-based event revenue management early will outperform competitors significantly in the next 3–5 years.

Data Requirements: What You Actually Need

Contrary to popular belief, you do not need massive infrastructure to begin.

You need:

• Historical ticket sales (at least 1–2 years ideally)
• Seat type categorisation
• Days-to-event tracking
• Agent performance data
• Add-on purchase data
• Cancellation and no-show data

From there, forecasting models can be built using:

• Demand curve modelling
• Gradient boosting classification
• Seasonality analysis
• Pickup velocity comparison

The intelligence sits on top of your existing system.

Avoiding Hype: What AI Does NOT Do

AI does not:

• Magically sell empty seats
• Replace commercial judgment
• Eliminate seasonality
• Guarantee sell-outs

What it does is:

• Remove guesswork
• Surface early risk
• Quantify demand signals
• Improve timing of decisions
• Increase average yield per seat

It enhances commercial discipline.

The Bigger Shift: From Ticketing Software to Revenue Infrastructure

This is the core difference. Most providers offer ticketing systems. Very few offer revenue intelligence systems for events. At The Percentage Company, our philosophy is simple: We do not want to be your software vendor. We want to be your revenue partner and we have the team and the tech infrastructure to be able to deliver what others cannot. 

That means:

• Monitoring demand daily
• Forecasting show performance
• Optimising seat allocation
• Managing agent yield
• Increasing direct booking contribution
• Integrating ticketing into overall marketing strategy

When ticketing, marketing, forecasting, and financial reporting sit in separate silos, revenue opportunities are lost. When integrated, they compound.

The Future of Event Revenue Management

Over the next decade, dynamic pricing in events will become standard. Airlines already do it. Hotels already do it. OTAs already do it. Event operators who resist will face:

• Margin compression
• Higher distribution dependency
• Reduced pricing power
• Greater vulnerability in low season

Those who adopt AI-driven demand modelling early will:

• Increase yield
• Stabilise performance
• Improve forecasting accuracy
• Make more confident pricing decisions

Final Thoughts

After 25+ years in hospitality marketing and technology, I have learned one consistent lesson: The operators who win are the ones who treat pricing as a living system, not a static number.

Dynamic ticket pricing is not about charging more. It is about charging smarter. It is about aligning price with demand. It is about understanding your seat inventory the way hotels understand room inventory. And it is about building infrastructure that turns data into decisions, consistently.

If you operate an event, show, or attraction and want to explore how AI-based demand modelling can increase your yield, I would encourage you to start with your data and instead of using a basic booking platform, use one that is setup for the AI revolution. The opportunity is already there. You just need to activate it.