What happens when you optimize lookalike audiences on the wrong metrics?


  • Increase of high-value customers by 19% (customers that generate 16x in sales)
  • Uplift in high-value customers by 19% generates a minimum revenue increase of 15%
  • Increase of orders for medium LTV customers by 46%
  • Decrease of orders for unprofitable customers by 51%

The Challenge

Growing profitability will require more sophistication and creativity to navigate the slowing economy and tectonic shifts in privacy.

One key way to do this is by leveraging the "80/20 rule," which states that a small percentage of high-value customers generate a disproportionate share of sales.

It’s hard to target high-value customers when you’re only focusing on the first purchase. Yet it takes months to know if a customer generated from a given campaign will buy once or again and again. In contrast, marketing decisions are made on a daily and weekly basis.

Optimizing acquisition channels such as lookalike audiences (LALs) is getting harder and less predictable for eCommerce operators, due to changes in privacy regulations and the power of existing platforms’ algorithms. 

Regardless of the technology, it's hard to get more juice out of lookalike audiences.

Many brands only measure the effectiveness of LALs based on the first purchase, which does not take into account the long-term value of the customer.

One of our clients, Good Clean Love, experienced this firsthand when they used a seed audience of predicted high lifetime value (LTV) customers and saw minimal uplift in return on ad spend (ROAS) compared to using historic purchase data to build audiences. They realized that by only looking at the first purchase, they were missing out on the full potential of high-value customers.

The Solution

Good Clean Love is a fast-growing consumer packaged goods (CPG) brand, that realized they could get more out of their ad spend by optimizing on better metrics. By predicting sales in real-time and before a transaction with 90% accuracy, Good Clean Love evaluates offers, content/creative, and influencers against the most important key performance indicator: LTV.

Ocurate analyzes campaigns from UTMs and makes recommendations to optimize ad spend based on predicted LTV (updated in real-time).

Instead of measuring the performance of Meta, TikTok, or Google from 3 to 6 months ago, brands are able to measure the day-to-day changes in LTV with 90%+ confidence.

Ocurate's insights helped the brand divest from low-performing customers, campaigns, and channels, accounting for 30% of its budget. They reallocated those investments to ads with higher predicted LTV.

The results: In less than six months, Good Clean Love increased the number of high-value customers by 19%. For any 80/20 company, this increase in high-value customers means a minimum revenue gain of 15%.

If 20% of customers generate 80% of sales, this means high-value customers generate 16 times more sales than everyone else.

Get in Touch

Our team would love to hear from you!

Let's Talk