How Beam unlocked a 4.5x ROAS acquisition channel

Beam is a subscription supplements business. That means a first purchase only matters if the customer stays subscribed long enough for the economics to work - which makes acquisition channel quality as important as acquisition channel volume.

Garner was a strong fit for exactly that reason. Rather than optimizing for clicks, the network is designed to find the upstream traffic most likely to convert into long-term customers. For Beam, that meant not just a new channel, but a better class of customer coming through it.

Most acquisition channels are optimized for the wrong outcome

Paid social and search can generate volume. What they can't do is filter for shoppers who are likely to stay subscribed for several months. For a subscription business, low CAC and good customers are not the same thing. You can buy plenty of first orders without moving the underlying economics at all.

The right traffic is nearly impossible to find manually

Beam's best customers don't come from supplement buyers in general - they come from specific contexts where a shopper's existing habits and purchase behavior already align with what Beam offers. Identifying those contexts by hand would have meant testing dozens of potential brand partners, managing creative across all of them, and reallocating spend every time the signal shifted. It wasn't a practical path.

Garner automatically tested the network for fit

Garner ran Beam across more than 50 brands, learning which upstream traffic sources produced the strongest response. Rather than asking Beam to guess where they belonged, the network tested potential brand partners continuously and concentrated distribution where the offer resonated most.

Creative improved inside the system

Finding the right brand environments was only half the equation. Garner's AI model also continuously refined how Beam appeared inside each of them - adjusting campaign creative based on shopper response so that the message matched the context, without Beam's team manually managing variations across dozens of partners.

SKU-level targeting improved relevance

Garner also learned which Beam products fit best in different partner contexts. Menopause shoppers were more likely to see Beam's hormone balancing blend, while sleep-focused shoppers were more often shown Beam Dream. That SKU-level targeting made the offer feel more relevant to the shopper and helped Beam convert a higher-quality customer.

Better-fit traffic produced better customers

Because Garner was finding Beam's highest-fit upstream contexts rather than the cheapest available clicks, the customers coming through the network were materially stronger. They spent more over time, and more than half stayed subscribed for over a year.

Beam achieved a 4.5x ROAS on Garner as an acquisition channel, with a 12% average conversion rate on Garner traffic. The customers acquired through the network outperformed Beam's broader average: lifetime value was 22% higher, and 50% remained subscribed for more than 12 months.

The CTR trend confirmed the same story. As Garner tested more partner environments and refined Beam's creative, click-through rate rose steadily - a clear signal that the network was learning where Beam fit best and how to present the offer effectively.

Key results

• 4.5x ROAS on Garner as an advertiser channel
• 12% conversion rate on Garner traffic
• 22% higher LTV than Beam's average customer
• 50% of acquired customers stayed subscribed for more than 12 months