Skip to content
Michael MancusoProduct & technology executive
Projects & case studies

Complex work.
Concrete results.

Product launches, operating-model changes, and complex AI-enabled builds. Explore the business problem, the decisions I led, and what changed.

Built with agents

Product strategy and a repository-based second brain.

Phase 3 Benefits brings eight health, commuter, and wage-law account types onto one payment card. I built a repository-based second brain—product strategy, specifications, verified facts, and agent memory—to keep AI development aligned.

Specification-driven development at platform scale.

Shared FHIR-based contracts coordinated parallel development across clinical, operational, and financial workflows.

Leadership case studies

Showing 6 of 38

Wendy's

Deployed Google Firebase for predictive modeling on Wendy's website, increasing conversion rates and revenue through targeted offers and enhanced user experience.

01

Challenge

Wendy’s needed to enhance its promotional strategies by delivering personalized offers that would increase user spending and drive higher conversions. The challenge was to leverage predictive modeling to identify users with a high probability of making purchases and serve them targeted offers.

02

Action

To address this challenge, I led the implementation of Google Firebase for predictive modeling on Wendy’s website. Key actions included:

  • Data Collection and Analysis: Integrated Google Firebase to collect and analyze user behavior data, identifying patterns and trends that indicated a high likelihood of spending.
  • Predictive Model Development: Developed predictive models to calculate the probability of user spending based on historical data and real-time interactions.
  • Targeted Offer Personalization:Configured the website to serve personalized offers to users identified by the predictive models as having a high probability of spending. Specific implementations included:
    • User Segmentation: Segmented users based on their behavior, such as frequency of visits, types of pages viewed, and interaction with promotions.
    • Offer Personalization: Created dynamic offers tailored to each user segment, ensuring relevance and appeal.
    • Real-Time Adjustments: Adjusted offers in real-time based on user interactions and updated probabilities from the predictive models.
  • Continuous Improvement: Continuously refined the predictive models and personalization algorithms through ongoing analysis and feedback, ensuring they remained effective and accurate.
03

Results

The implementation of predictive modeling with Google Firebase led to significant improvements in Wendy’s promotional strategies and revenue:

  • Higher Conversion Rates: The targeted offers resulted in higher conversion rates, as users were more likely to respond positively to personalized promotions.
  • Increased Revenue: The use of predictive modeling to serve offers to high-probability spenders resulted in increased revenue from higher-value transactions.
  • Enhanced User Experience: Users received more relevant and appealing offers, leading to a more enjoyable browsing experience and increased satisfaction.
  • Data-Driven Insights: The continuous analysis of user behavior provided valuable insights that informed future promotional strategies and helped optimize the predictive models.
  • Efficient Resource Allocation: The ability to target offers more effectively allowed Wendy’s to allocate marketing resources more efficiently, maximizing the return on investment.

Questions

How was predictive modeling used at Wendy's?

Google Firebase predictive modeling was used to analyze customer behavior patterns and deliver targeted offers, increasing revenue through personalized marketing based on data-driven insights.

What revenue impact did the predictive modeling deliver?

The predictive modeling initiative drove increased revenue through targeted offers that reached the right customers at the right time, improving conversion rates and customer engagement.

How did Firebase enable targeted marketing at scale?

Firebase provided real-time behavioral data and machine learning predictions that segmented customers into actionable groups, enabling Wendy's to deliver the right offer to the right person at the right moment.

What data signals did the predictive model use to identify high-value customers?

The model analyzed behavioral signals including visit frequency, page types viewed, interaction with promotions, historical purchase patterns, and real-time browsing behavior to calculate spending probability.

How were targeted offers personalized in real time?

The system dynamically adjusted offers based on real-time user interactions — as browsing behavior updated, the predictive model recalculated spending probability and served the most relevant offer for each user segment.

Need a leader who can
build with AI?

Let's talk about the product challenge, the team, and the outcomes you need—whether you're improving how your business runs or building something new.

Discuss your product challenge Connect on LinkedIn