Case Study: Realtor.com achieves a faster, better customer experience and conversion lift with Optimizely Full Stack

How Realtor.com Delivered a Faster, Better Customer Experience with Experimentation

Realtor.com, a digital-first real estate marketplace, faced a tradeoff between site speed and ad revenue: digital ads were slowing page performance and potentially hurting user engagement and conversions. The analytics team needed to prove whether reducing ads and improving speed would lift conversions enough to offset lost ad income, but standard A/B testing was unsuitable because they had to measure long-term user growth, SEO effects, and limit advertiser impact by testing at a geographic level.

Using Optimizely Full Stack, the team created a geo-based experiment by generating billions of parallel randomized combinations of U.S. states to form three balanced state buckets that moved together seasonally and covered 30–35% of users each. They analyzed results with a difference-in-differences relative change method, ran the program over six months, and found meaningful increases in user growth and conversion — enabling a broader rollout, faster iteration on experience improvements, and stronger monetization and engagement outcomes.

Realtor.com