Project Overview

In this project, we aimed to enhance audience targeting and improve marketing effectiveness by combining behavioral data with demographic insights. Through advanced data processing, metric-driven evaluation, and persona development, we identified the most responsive audience segments and generated actionable recommendations for future campaigns.

Data Collection & Preparation

To prepare the data, I applied the regular expression matching function to parse and categorize the audience segment column in our dataset. This allowed us to classify user IDs according to the type of content they were engaging with during ad interactions.

Next, we merged the cleaned behavioral dataset with a comprehensive demographic dataset, enabling us to analyze not only how users behaved but also who they were. This detailed view of the audience formed the foundation for more granular and actionable insights.

Metrics & Scoring System

Our analysis centered on three key performance metrics:

• CPA (Cost Per Acquisition)
• CVR (Conversion Rate)
• CTR (Click-Through Rate)

We established clear performance benchmarks:

• Maximum CPA of $250
• Minimum CTR of 0.015%

To evaluate audiences holistically, we developed a scoring system that normalized each metric to a 0–1 scale:

(Value - MIN(Range)) / (MAX(Range) - MIN(Range)) = Normalized Value

We then combined the normalized metrics using the formula:

(Normalized CVR + Normalized CTR) - Normalized CPA = Audience Score 

Key Findings

Using this scoring method, we identified the top 10 performing audience segments. The following categories consistently stood out:

• Real Estate
• Health
• Sports
• Fashion & Beauty
• Automotive
• Travel
• Finance

These segments demonstrated an optimal blend of low CPA and high CVR/CTR, making them ideal for focused marketing efforts.

Additionally, we observed significant variance within broad categories. For example, the “Sports” segment included both casual readers and high-intent enthusiasts with markedly different behaviors. This finding emphasized the importance of granular segmentation over broad category targeting.

Persona Development

We then created audience personas based on the highest-performing segments. These personas humanized the data, offering a relatable lens through which marketers could craft more resonant messaging and creative strategies.

Data-Driven Recommendations

Based on our findings, we developed several actionable recommendations:

• Prioritize college-educated parents in their 20s-30s.
  This demographic showed strong interest in financially significant areas such as real estate and finance. Campaigns targeting this group should emphasize themes of financial empowerment and investment readiness.
• Use lifestyle content selectively for older demographics.
  Fashion and sports content can perform well among older segments, but should only be prioritized if aligned with the product or service being promoted.
• Leverage educational segmentation.
  College graduates consistently engaged with more complex topics and responded well to advanced content. Tailoring messaging based on education level can significantly enhance personalization and improve conversion rates.

Conclusion

By integrating behavioral data with rich demographic insights and applying a thoughtful scoring methodology, we successfully identified and characterized the highest-performing audience segments.

Our work delivered actionable insights into:

• Demographic preferences
• Cost efficiency
• Content engagement patterns

This data-driven approach ensures that marketing strategies are not only more precisely targeted, but also optimized for maximum ROI.