Luca: a performance-first CDN with a latency wedge

I led Luca from positioning through pre-seed — a scalable CDN with native and global server-switching for media-heavy applications. My job was product strategy and the technical narrative early customers and investors needed to believe. We weren't selling "cheaper infrastructure." We were selling a number teams could benchmark: time to first byte, global switchover under load, cost per gigabyte at scale.

Media-heavy apps in emerging markets pay latency tax twice: slow uploads and slow delivery. Incumbent CDNs won on brand and ecosystem, not on measurable latency wins for teams without enterprise budgets. The trap was building a feature matrix to match incumbents page-for-page. We'd lose on sales slides and win nowhere on the metric that actually moved apps.

I bet on latency as the wedge — native and global server-switching that moved a number teams could benchmark. Pre-seed positioning was "performance alternative with proof," not "cheaper clone." Every roadmap item had to answer: does this improve the benchmark story? If not, it waited.

[NEEDS YOUR INPUT: Add the specific trade-off moment — e.g., what feature you said no to, or the benchmark that won the first customer conversation.] Until then: the fight was resisting feature parity with incumbents and holding the line on measurable latency wins as the product story. Sales wanted parity decks; I wanted benchmark scripts customers could run themselves.

Luca raised a $300K pre-seed round. I owned the performance narrative in the data room — latency benchmarks, server-switching architecture, and the story investors could repeat without a live demo. Early customers bought the number, not the logo comparison slide.

I'd publish benchmark methodology publicly before sales asks for it. Transparency builds trust faster than claims. I'd also cap v1 to one media type and one region until the number is undeniable — breadth kills latency stories.