Public-facing customer evidence. Honest about what is live data and what is scaffold. Cohort sample size N=4 (our own brands) until merchant data flows through the benchmark pipeline.
Pauses out-of-stock ad sets, reallocates Meta budget when blended ROAS dips, surfaces true ROAS vs reported daily.
Live data Q3 2026 post BB-01 + 30 days. The current page renders against stub data; figures update once the BB-01 connect lands.
Read case studyDetects rank-proximity content gaps weekly, drives 2x organic share over three months on the content-gap algorithm.
Watch tier, one of our own brands. Organic-only brand; no paid spend in the picture.
Read case studyReads weekly content trend signals from Reddit and Google Suggest to inform the next month of family lifestyle topics.
Watch tier, one of our own brands. Creator brand; reads signal, decides editorial.
Read case studyCatches off-intent search terms, validates conversion tracking daily, drives CAC down quarter over quarter.
Eating our own dog food on Google Ads. The recursive proof point.
Read case studyWhy scaffold pages ship before live data. A case study with placeholder numbers is more honest than one with fabricated numbers. The B&B scaffold renders structure now and fills with live data after the BB-01 connect completes its first 30 days. The status pill on each card tells you which state the underlying page is in.
Cohort sample size. All current case studies are oddly's own brands. The N=4 disclaimer stays visible until merchant cohort data flows into the public surface via the benchmark aggregation pipeline. We do not fabricate. If a metric is not ready, we say so.