Marketing Channel Performance

Robinhood Marketing Channel Performance Analysis

Project Overview

This project analyzes Robinhood's marketing channel performance for the first half of 2025 using MySQL 8.0. The objective is to evaluate advertising spend, user acquisition, paid subscriptions, and first-90-day revenue to identify effective campaigns and recommend a budget allocation for the next six months.

Tools Used

• MySQL 8.0
• MySQL Workbench
• SQL

Datasets

ad_spend.csv

• Daily advertising performance from January to June 2025.
• Includes date, channel, campaign, impressions, clicks, and spend.

signups.csv

• Contains 18,840 user signup records.
• Includes signup date, acquisition channel, campaign, subscription details, plan, and first-90-day revenue.

Analysis Performed

• Cleaned advertising data by identifying and removing duplicate records.
• Standardized channel names to match the advertising and signup datasets.
• Calculated campaign-level advertising spend, impressions, clicks, CTR, and CPC.
• Measured signups, paying subscribers, cost per signup, and cost per paying subscriber.
• Calculated first-90-day revenue and Return on Ad Spend (ROAS).
• Compared signup-to-paid conversion rates across paid and unpaid channels.
• Examined monthly spending, signup trends, and campaign activity.
• Developed a preliminary marketing budget recommendation based on campaign performance.

Key Findings

• Brand Search delivered the strongest results, with a cost per paying subscriber of $21.93 and a ROAS of 1.91x.
• Retargeting was the second-strongest campaign, with a cost per paying subscriber of $31.07 and a ROAS of 1.34x.
• Prospecting Video generated the highest number of signups (6,484), but its cost per paying subscriber was $159.23 and its ROAS was 0.26x.
• Display Network generated only 8 paying subscribers and recorded a ROAS of 0.03x. Its advertising records ended on May 14, 2025, requiring further investigation.
• Paid Search had the highest channel-level signup-to-paid conversion rate at 21.43%.
• Generic Search, Finance Blogs, and Prospecting Video generated less than $1 in first-90-day revenue per $1 of advertising spend.

Budget Recommendation

Prioritize Brand Search and Retargeting for controlled budget increases, as they delivered the strongest acquisition-cost and first-90-day revenue results.

Review Generic Search and Finance Blogs for opportunities to improve targeting and conversion performance. Test improvements to Prospecting Video before increasing its investment. Investigate Display Network's performance and missing activity before deciding whether to resume spending.

Budget changes should be gradual and evaluated using acquisition cost, revenue efficiency, and incremental campaign results.

Limitations

• ROAS measures revenue relative to advertising spend, not profit.
• First-90-day revenue does not represent customer lifetime value.
• Recent signups may not have a complete 90-day revenue observation period.
• Historical performance may change when advertising budgets increase.
• Organic and referral channels have no directly recorded advertising spend in the provided dataset.

Conclusion

The analysis identifies Brand Search and Retargeting as the strongest paid campaigns based on cost per paying subscriber and first-90-day ROAS. The findings support prioritizing these campaigns for further testing while investigating weaker-performing campaigns before allocating additional budget.

The project demonstrates the use of SQL for data cleaning, aggregation, joins, marketing metric calculation, and data-driven business recommendations.