Less Targeting, Better Results: Auto Retailer Case Study

The Challenge

A major automotive retailer was hitting their targets but suspected they could do better. Their streaming campaigns used heavy demographic and behavioral targeting—standard practice, but maybe overkill.

The Test

We stripped out most targeting filters and let the algorithm find efficiency. The client allocated 40% of their streaming budget to test broad reach vs. precision targeting for six weeks.

The logic: machine learning might be better at finding potential customers than our assumptions about who they are.

The Results

Performance improved across the board:

• 12% higher conversion rate
• 31% increase in reach (same budget)
• 19% lower CPMs
• Search traffic also increased 8%

The broad approach caught people earlier in their buying journey, creating lift for other channels too.

Why It Worked

This retailer had solid creative and a good website—the algorithm had quality data to optimize against. In streaming's fragmented landscape, sometimes our "smart" targeting just limits scale.

The client's take: "We stopped overthinking who to target and started focusing on creative, messaging, and efficient delivery."

Bottom Line

Six months later, this simplified approach became their default strategy. They still use targeting for specific campaigns, but their always-on streaming runs broad.

Sometimes the best optimization is knowing when to stop optimizing.