Edlitera works best in modern browsers that support HTML5 video. Click to learn more.

Consulting Case Study

40% reduction in marketing expenses.

25% reduction in cost to acquire new members.

Faster and more efficient marketing campaign testing and evaluation.

Better targeting for acquisition campaigns using machine learning.

The Client

Our client is a well-established e-commerce company, specializing in household goods.

The Challenge

Our client had recently launched a new category of products under a different brand. While business was steady, it was not doing as well as they had projected, and they were looking for ways to speed up the customer acquisition process, while keeping costs low.

The company measured the effectiveness of their marketing campaigns every four months, and these evaluations showed that the strategies they were using to acquire new customers were simply not performing as well as expected. The company acquired large volumes of leads, but these leads were increasingly lower quality, and were not converting. Also, the targeting models used in their acquisition campaigns were losing their effectiveness, and increased competition in their market niche was also a problem.

Our Solution

Specifically, the company needed to accelerate the rate at which they learned from and optimized their marketing activities in order to find new ways to reverse the decline in performance and lower their customer acquisition costs. The company reach out to us through a referral, and we started working with them towards these goals.

First, we designed and implemented a new testing framework for marketing activities, which enabled the company to test for key marketing components, like audience, offer, and creative assets, in combination, while lowering their testing volume and cost.

In addition, we did a multidimensional segmentation of the company’s database of prospects, and evaluated possible growth opportunities for each segment using recent campaign performance data.

Finally, we analyzed the company’s existing customer pool, and created new campaign targeting and lookalike models based on the characteristics and customer personas we uncovered. We then used machine learning to automate this analysis, so the company could run it on their own as often as needed after our engagement.

Results & Outcomes

The more flexible testing framework allowed the company to test the effectiveness of their marketing programs much more frequently, instead of only 3 times a year. In addition, our research of the existing customer pool and the lookalike models we built allowed the company to identify segments of high-value prospects, as well as to pinpoint which marketing channels these prospects preferred to be contacted through. This increased precision enabled our client to better reallocate marketing resources from underperforming initiatives to higher value opportunities.

Overall, the company’s marketing expenses dropped by more than 40%. Also, thanks to the more granular segmentation of their prospects, and to more precise targeting, the company lowered its cost of acquiring new customers by over 25%.

Do you have a similar challenge? Get in touch with us at [email protected], and let’s see how we can help.