Lume Wearables — Stabilizing ROAS With Full-Funnel Creative and Attribution

Lume had product-market fit but volatile ROAS. Our job was to stabilize performance and build a repeatable playbook. Here’s what we did.

The Problem — Volatile ROAS

Spikes and drops made it hard to plan spend and forecast. Creative fatigue, limited creative variety, and unclear attribution across Meta and TikTok were the main suspects. We started with a full-funnel creative audit and attribution setup.

Creative Overhaul and UGC

We shifted from a small set of hero creatives to a continuous UGC pipeline: real users, varied use cases, and clear hooks. Testing cadence increased; we killed underperformers quickly and scaled winners. Creative refresh became a weekly discipline.

Multi-Touch Attribution

We implemented a multi-touch attribution model so Lume could see which touchpoints led to conversions, not just last-click. That informed budget allocation between Meta and TikTok and between top-funnel and retargeting. ROAS became a stable metric rather than a daily surprise.

Influencer and Seeding

We added a structured influencer and seeding program to feed UGC and social proof. Creators were selected for fit and reach; content was repurposed into ads. This improved creative diversity and lowered cost per piece of performing creative.

Outcome

ROAS stabilized around 4.8x on a blended basis, with less week-to-week variance. The brand could scale spend with confidence and plan inventory and offers around more predictable performance. Full-funnel creative and attribution were the foundation.