Case Study: Octopart achieves higher website conversions and revenue with Split

Octopart Increases Website Conversion via Experimentation

Octopart, a leading parts search engine with a database of 50 million electronic components, needed to scale its decision-making as traffic and revenue grew. Because revenue depends on many small cost-per-click transactions, even minor UX or copy changes could materially affect conversions. Early-stage, ad-hoc changes became too risky at scale, so the team needed a reliable, data-driven way to measure impact without building and maintaining an in-house experimentation platform.

Octopart adopted Split for A/B testing and feature flags, valuing its transparent statistics, documentation, and integrations. Within two months they found new CTA copy that significantly increased conversions and revenue, and Split’s experiments also uncovered a search UX change that unexpectedly reduced conversions — allowing the team to roll back and learn. Today they run multiple concurrent experiments, use Segment and their data warehouse for deep analysis (including regional segmentation), and gate launches with “do no harm” tests to protect conversion and growth.

Octopart

Sam Bobb

Sr. Director of Data Science