How Much Targeting Is Too Much Targeting?

Data-Driven Thinking" is written by members of the media community and contains fresh ideas on the digital revolution in media.

Today’s column is written by Frost Prioleau, CEO and co-founder at Simpli.fi.

P&G made headlines in 2016 when it announced it would scale back on Facebook targeting. At the time, P&G CMO Marc Pritchard said, “We targeted too much, and we went too narrow.”

That was misconstrued by some as a broad condemnation of narrow targeting. The real takeaway, though, is that hypertargeting has great value in some use cases but not so much in others.

For brands that advertise products that are widely bought every month, as P&G does, paying for layers of expensive data often doesn’t deliver positive ROI. For these products, it is often a better strategy to start campaigns broadly and let the optimization algorithms do their job in finding the combinations of audiences, ad sizes, domains/apps, geographies, formats and other factors that deliver desired KPIs most efficiently.

If such campaigns are overtargeted, it constrains the impression pool from which the algorithms can optimize, keeping campaigns from performing at their best.

On the other hand, there are products, verticals and categories in which using data to target clearly defined audiences can deliver the best ROI. For products bought infrequently, such as cars, houses and big-ticket items, data that defines which users are actively looking to buy helps advertisers eliminate wasteful spending on users who won’t buy, no matter how many ads they see. In-market data helps advertisers focus their spend on just the small subset of consumers who might be influenced.

Demographic targeting has great value when marketing products that may be of interest to a large portion of the population but are only relevant to a small subset. For example, there is little reason to advertise high-end cars to users with household incomes of less than $50,000. Likewise, there is little reason to advertise student loan refinancing products to consumers over 60 years old.

Some products are so specific that only a small subset of buyers even know what they are. Many industrial and B2B products fall into this category. Using search retargeting, advertisers can target users based on their recent search history and find the users searching on very specific products. Search retargeting also works well for identifying in-market buyers for products such as cars and homes.

Retailers and other advertisers have long known that site retargeting, based on first-party data that identifies users who have already visited their websites, is one of the most effective digital advertising techniques. While advertisers typically don’t have to pay for this data since they already own it, they often find it makes sense to pay for dynamic ad units that personalize ads to the actual products viewed by the site visitor.

There are also clear use cases for geolocation data. Grocery stores, brick-and-mortar retailers, restaurants, auto dealers and other advertisers know that most of their customers come from neighborhoods that are close to their locations. For these types of customers, investing in GPS-based geotargeting often delivers a strong return because the targeting enables them to pinpoint their ad placements to the best locations and see lifts in foot traffic driven by their ads.

This is a technique the CPG advertisers like P&G are using to drive users to their retail partners. Geolocation data can also be used to “retarget” users who have physically visited an advertiser’s location or users who have physically visited a competitor’s location.

As personalization capabilities become increasingly available in programmatic advertising, advertisers have had to consider whether they use data and targeting to optimize a dynamic audience that sees a static message or use dynamic ads to optimize a message that is delivered to a static audience.

Going forward, it will be more of a many-to-many approach, where multiple messages are shown to multiple audiences to deliver the best ROI.

Follow Frost Prioleau (@phrossed), Simpli.fi (@Simpli_fi) and AdExchanger (@adexchanger) on Twitter.

1 Comment

  1. Ryan Peters

    Another point to this is ensuring that agencies and brands have aligned goals. I often receive requests from agencies to overly hypertarget, such as messaging a travel rewards card to only people sitting in the international gates at JFK. So aside from only delivering a few impressions, you're also likely to hit an audience that's half international, making them ineligible for the credit card. When you ask the brand who the best target is, they have a fantastic data set based on who has already been approved for the card. Without aligning what the agency thinks is a "very cool approach" with the brand's ability to already have a good read on who they should target, neither party will be happy with performance. Being able to measure brand KPIs like applications and approvals, combined with a strong identity graph, should help answer which data is worth the investment.

    Reply

Add a comment

XHTML: You can use these tags: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>