Cases
Case Study: How Beymen achieved a 393% ROAS during its 50% discount IBQML campaign
Challenge:
- For luxury retailer Beymen, the main goal was to move beyond broad outreach and efficiently identify users with a high probability of making a purchase from its vast online audience.
- The challenge was to proactively predict future buying behavior to maximize return on ad spend (ROAS) during key sales periods.
393%
50% Discount Campaign
125%
increase in ROAS
Approach:
- In partnership with SEM, Beymen used Google Analytics 4 (GA4) and BigQuery Machine Learning (IBQML) to launch a predictive analytics strategy.
- A custom IBQML model analyzed granular behavioral data to identify conversion patterns and scored the entire user base by purchase probability.
- Users with the highest scores were automatically segmented into a high-value audience and exported to Google Ads for direct engagement in campaigns.
Partnering with SEM:
- SEM is a Google Premier Partner Award winner and one of the leading digital marketing agencies in Turkey.
- The agency specializes in performance marketing, brand marketing, search engine optimization (SEO), data intelligence, and web analytics.
Results:
- 50% Discount Campaign: Performance Max campaigns achieved a 393% ROAS (a 393% increase in ROAS).
- 60% Discount Campaign: Achieved a 125% increase in ROAS
- Overall Benchmark Improvement: ROAS improved by 21.69% over standard benchmarks.
- Key Takeaway: High-intent predictive modeling significantly outperforms traditional segments in driving efficient growth.