B2C E-commerce

Increased subscribers by 3X
Decreased CPA by 10%

Trendy Butler is a diverse men's apparel subscription service that gives members $150+ worth of men's clothes for only $65 per month. Their technology personally caters to each user's style preferences, sizes, and overall personality in order to ensure that customers love each package more than the last.

Visit Client Site
B2C E-commerce case study featured image
Industry Retail Subscription
Vertical B2C E-commerce

The Performance Proof

Takeover Date: January 2017

Challenges & Goals

Trendy Butler was facing the challenges of increases subscriptions and revenue without relying solely on their brand name. Internal attempts to build out non-branded campaigns were unsuccessful, and relying on their branded name for all sales would be detrimental to growth. Our primary goals were to ramp up their non-branded efforts, increase sales, and capture customers with a high lifetime value at a profitable CPA.

Approach, Strategy & Solution

Trendy Butler required a very complex account structure and strategy. We first developed a funnel strategy (mid and low) that utilized categorical keyword pools. This structure separated keywords by their category and which part of the sales cycle they most likely fit in to. All broader keywords that contained “clothing” in them, but did not include a specific gender or focus on a subscription fell under mid-funnel. Keywords that contained the specificity of gender or a clothing subscription fell under low-funnel. The low-funnel campaigns & keywords were given an increased budget and higher bids, since these searchers were more qualified and interested in our specific offering.

We also conducted an in-depth analysis on where branded sales were originating from to ascertain that we were utilizing the best possible attribution model. Our analysis determined that using a first-click attribution model would be most conducive to the user scenario we were looking for. The conversion path involved users searching for either men’s clothing or subscription based clothing services. They were not specifically searching for our brand, but since we were bidding on non-branded terms like “men’s clothing” or “clothing subscription” the user got exposure to our brand name. This user would then search our brand name, “Trendy Butler” click our ad and then convert. Using traditional, “last click” attribution all of the credit would go to our Branded campaign and the investment into our non-branded keywords like “Men’s clothing” would seem like a bad choice on our part. But, in all reality we achieved this “branded” sale, due to our non-branded efforts. After analyzing 6 months of data we discovered 85% of all branded sales, originating from non-branded keywords. Properly attributing non-branded keywords in the sales cycle gave us the flexibility to grow this element of the business.

We expanded our non-branded campaigns to capture all relevant traffic for both men’s clothing and subscription services. This also provided our account with massive swaths of data that could be used for remarketing lists. Combining remarketing lists, detailed demographic, gender, and income targeting allowed us to use decreased CPAs to leverage higher CPA campaigns and broad targeting.

As a result, we saw tremendous increases in revenue and subscriptions in both branded and non-branded campaigns while maintaining overall CPA goals.


Increased subscribers by 3X
Decreased CPA by 10%