Business Challenge

This client managed a large and growing portfolio of advertising assets including billboards, transit ads, and digital signage. Despite strong demand in some locations, the company faced several challenges:

• Static pricing models led to underpricing in high-demand areas and overpricing in low-traffic zones.
• Sales teams relied on manual rate cards, often disconnected from real-time market conditions.
• Inability to dynamically respond to inventory availability, demand spikes, or weather & event-based fluctuations.
• Low inventory utilization during off-peak seasons or in certain geographies.

The leadership team wanted to improve asset monetization, utilization, and competitiveness using a data-driven strategy.

Objective

Implement an AI-powered Dynamic Pricing Engine that adapts billboard pricing in real time based on multiple factors such as:

• Asset location & visibility
• Historical demand and occupancy rates
• Nearby events, weather conditions, and traffic patterns
• Seasonality and time-of-day
• Advertiser profile and industry trends

Solution

A hybrid team of AI consultants and SkyView's internal product team collaborated to design, build, and deploy the Dynamic Pricing AI Platform.

Key Components

1. Data Aggregation & Model Training
   • Integrated data from CRM, booking history, foot traffic sensors, weather APIs, Google mobility trends, and event calendars.
   • Built predictive models using historical pricing and occupancy data to forecast demand and ideal price elasticity.
2. Real-Time Pricing Engine
   • AI engine suggested optimal pricing per asset in real time.
   • Integrated with SkyView’s internal booking platform used by sales reps and clients.
   • Included override features and “floor price” settings to allow strategic manual control.
3. Recommendation Dashboard
   • Visual tool for sales and asset managers showing:
   • Pricing suggestions
   • Revenue opportunity alerts
   • Occupancy forecast per asset
   • Pricing rationale backed by AI signals
4. A/B Testing Environment
   • Rolled out dynamic pricing to 20% of digital inventory to test performance before scaling.

Results (Post 3-Month Rollout)

Key Success Factors

• Granular data labeling: Cleaned and tagged asset data by visibility, location type, traffic density
• Sales team alignment: Co-created AI rules and transparency to avoid resistance
• Built-in guardrails: AI never offered prices below minimum set by finance
• Real-time adaptability: Adjusted pricing on weekends, festivals, weather changes (e.g., during traffic jams or concerts)