Blog

How a CPG Brand Used Analytics to Discover New Customers

This is the second post in a series on building an analytical framework for marketing. For more on the series, start with the intro, "Analytics Are Key to Unlocking Marketing Success."

In an August 2016 interview,  Procter & Gamble CMO, Mark Pritchard said that the CPG titan had made a mistake with its Facebook campaigns by “targeting too much and going too narrow.” Now, he said, they were asking, “What is the best way to get the most reach but also the right precision?”

The balance between how broad of an audience to reach and how precisely to target particular customer subgroups’ unique wants and needs is one of the fundamental challenges facing large national brands. Striking this balance has become even more complex in the digital age. While marketers have targeted consumers in print, TV, outdoor or radio, using demographic, geographic and interest/lifestyle heuristics for decades, the advent of programmatically-enabled digital buying has brought with it the promise of addressing individual consumers with theoretically limitless precision.

However, while the potential for targeting may be limitless, the law of diminishing returns still applies. The marginal benefit of delivering a customer a more persuasive message or offer declines exponentially with greater precision. Further, not all products and brands are the same; the utility of targeting depends on the breadth and diversity of the relevant customer base.

As with all things, it comes down to costs and benefits. Differentiating between individuals so you only reach those who are in-market for your product and reaching them with a message tailored to their specific wants and needs has obvious benefits (e.g. serving a Spanish language ad to a Spanish speaker, or serving a frequent flier program benefit for a business traveler versus a low-price offer to a leisure traveler.)

Unfortunately, the costs of achieving this mythical marketing nirvana are far greater than just data costs required to execute it. Even if you know everything about everybody, it would be prohibitively expensive to develop creative designed to appeal to each individual’s specific needs and wants and to set up the campaign to appropriately allocate those creatives.

The right balance lies somewhere in the middle. By distinguishing users on a limited  set of key dimensions -- such as age and gender, or income and location -- a marketer can design a campaign that profitably chooses to send messages tailored to their unique wants and needs.  

The question becomes, “Which segments of consumers are worth targeting in a proprietary way, and how should I message to them?” This is where Turn’s data management platform (DMP) provides a fast, powerful way to discover what the most important customer segments are and the technology to reach them deterministically.

In an example of targeting done right, one of our DMP clients, a major CPG brand, asked us to help them design a targeting strategy for a self-improvement hygiene product. They provided us with site data on users visiting the section of their website devoted to the specific need and asked us to group those visitors into segments that would be useful to target to.

While a plurality of the visitors to the site were in the brand’s target demographic, female millennials, more than half of users were over 50, either parents or retirees. It seemed reasonable to think these folks should be engaged with different imagery and messaging than millennial women. Turn worked with the brand insight team and agency to develop and implement a campaign targeted based on age and gender. Since targeting by age and gender is one of the least expensive, most accurate, and scalable dimensions for targeting available, the campaign was able to achieve some precision with a high reach and at a low cost.

When targeting ads, you have to strike a balance. At Turn, we don’t just have the tools to target campaigns; we are trusted advisors to brand marketers and are dedicated to understanding your customers and to achieving your goals.

See the intro post on building an analytical framework

Eric Mason

Analytics Lead