In her 2019 Internet Trends Report, Mary Meeker highlighted several key statistics that impact digital marketing strategies. As Americans spend more time than ever — 6.3 hours a day — consuming digital media, advertisers are taking advantage by investing more in digital advertising. Internet ad spend rose 22% in 2018, propelled by more data for better targeting and higher relevancy, enhanced AI capabilities, and more innovative ad formats. But at the same time, Meeker cautioned marketers on the rising cost of customer acquisition — a rate that won’t be able to exceed lifetime value for much longer.
Customer acquisition has always represented a push and pull tension: How to reach the right potential customers, but do so at scale, while keeping costs at bay. The perfect balance of achieving both efficiency and effectiveness in prospecting new customers has eluded marketers.
For companies in sectors like banking, credit and insurance, this added nuance is managing risk: improve the rate of gaining lower risk customers while managing acquisition costs.
Lookalike audiences to stimulate demand
With this risk-management mindset, one way to approach audience creation is by targeting new customers who look like high-performing or high-value current customers. A lookalike audience can help you stimulate demand from new prospects who have the same detailed profile characteristics as your existing best customers. By targeting prospects who look like existing customers, you also reduce wasteful spending on people who aren’t a good fit, an approach that helps tip the efficiency scales in your favor.
Lookalike-audience techniques are a very powerful addition to your marketing mix. However, the construction of the lookalike audience can mean the difference between success and failure. There are three important components to consider when evaluating your lookalike strategy:
- Seed — A sample of users who represent the desired target. Examples include recent converters, high-scoring leads and customers with a high lifetime value.
- Data — The types of data used to determine how much one user looks like members of the seed.
- Modeling — The methodology used to create the lookalike audience. In other words, exactly how is meaning extracted from the data and made useful?
Since lookalikes are based on what you already know about your customer, they tend to perform better than an off-the-shelf audience segment. That seed file becomes the basis by which you can expand your footprint using an audience that has already proven valuable to you — refilling the top of the funnel with prospects who are more likely to be responsive and loyal.
The data science that is then applied to model your seed is critical to success. For marketers in highly regulated sectors like financial services, one roadblock with activating lookalikes is that the modeling is often done in a black box. That is, without knowing the data that fuels the modeling algorithms, financial services marketers risk unintentional use of restricted attributes. Couple this with growing measures around data privacy, and it’s no wonder that transparency becomes of the utmost importance — even if you don’t operate in a historically regulated environment.
Find the right lookalike partner
Find a lookalike audience partner who has transparency built into their modeling process with the flexibility to use only the modeling attributes that meet your company’s compliance standards.
As companies continue to vie for consumer attention in a highly fragmented media ecosystem, don’t expect the pressure, barriers and associated costs to capture new customers to let up. Smart marketers will continue to test and learn ways to apply accurate data and sophisticated data science via lookalikes or other custom audience solutions to find their best customers and drive both efficient and effective acquisition.
The information contained in this blog is for informational purposes only and does not constitute legal or financial advice. You should consult your own attorney or financial adviser regarding your particular situation. TransUnion does not guarantee the accuracy or completeness of the information provided. The appearance of external hyperlinks does not constitute an endorsement by TransUnion. TransUnion LLC and its subsidiaries and affiliates (“TransUnion”) make no endorsement, recommendation, or representation with respect to the information provided herein. For complete details of TransUnion products, visit www.transunion.com.