Marketers must understand the difference between blasting impressions and deploying safe targeting tactics.
Recent news around audience buying and privacy has me thinking about a digital world without targeting, and competing philosophies on how to optimize campaign performance. I'm often asked by brand advertisers, "If I'm selling XYZ and everyone is potentially a customer, how important is targeted advertising?" This is a deceptively complex question.
It's not unusual for marketers to see the road to salvation paved with quantity over quality. To "spray and pray," or "cookie bomb," is indiscriminately buying as many ad impressions as cheaply as possible, regardless of content. After all, the advertiser is getting a lot of conversions; especially post-view conversions, and therefore this must be good return on the cheap investment, right? Not really.
The first thing to consider on this "cookie bombing" scheme is baseline performance. Baseline performance here is defined as the conversion rate of the population not exposed to display ads from the advertiser. Keep in mind, baseline performance is usually not zero since other advertising and marketing channels also influence consumers. By blasting out low-cost impressions to a large amount of people, cookie bombing gets perceived performance by claiming some credit for all baseline conversions that would otherwise occur outside the influence of a particular campaign. However, what we really should care about is the incremental lift. This can be done by reserving a small percentage of users in a control group that is not exposed to any display ads. The performance of this control group is your baseline performance. You'll soon see that cookie bombing doesn't outperform the control group by much, if at all.
Another problem with the cookie bombing logic is that it only looks at where conversions happen (everywhere with no apparent pattern) vs. how they happen (much more structured and patterned). Suppose two media buying tactics run side by side, the first uses precise audience targeting with strict frequency control and run on high quality sites; the second uses the lowest cost sites with no audience targeting and uncapped frequency (i.e., cookie bombing). In my opinion, consumers are more likely to be influenced by the high-quality impressions. However, since many consumers may not convert right away, they are very likely to run into one of the uncapped impressions from the second tactic. Therefore, the naïve "last touch win" attribution will give most of the credit to the cookie bombing method.
Additionally, blanketing the Web with lots of cheap impressions not only dilutes the brand image, but may also associate the brand with lower quality content. In a recent campaign for a major electronics advertiser, my company's data showed that Tier 1 (content that is appropriate for audiences of all ages, but with limited audience sizes) delivered an eCPA that was 43 percent lower than the average and that Tier 1 received an eCPA of 62 percent lower than the Tier 4 (deemed the riskiest content) eCPA. Additionally, CTR in Tier 1 inventory nearly doubled that of Tier 2. Even between the top two tiers, the results prove that for that extra 20 cents in inventory costs from Tier 1 to Tier 2, advertisers receive a $5 lower eCPA. Results indicate that quality and content do matter and that marketers can achieve better price per acquisitions with premium, quality placements than compared to blasting tons of impressions across cheaper inventory.
The last factor in favor of a targeted approach is what I call "stratified bidding." This method is based on a simple mathematical phenomenon that weighted average can be very different from the simple average. For example, every large enough audience segment is typically further composed of sub-segments described by different targeting rules. Just by natural variance, these sub-segments will have different response rates or levels of affinity to the brand. Stratified bidding is the real-time bidding technique that varies the actual bid based on the estimated response rate of these sub-segments. Then simply by the virtue of the weighted average you will gain on the overall ROI. The bigger the variance between sub-segments, the more you will gain by this real-time bidding technique. This technique is not just limited to rule-based sub-segments, but also any other variables that can be overlaid onto the audience bidding.
In 2011, I hope as an industry we dispatch the false prophet that is "last impression wins" in favor of fine tuning targeting principals that facilitate true performance. It's important for marketers to understand the difference between blasting impressions across tons of cheap inventory and leveraging safe targeting tactics to reach the right audiences, with the most relevant information in a brand-safe environment. Digital advertising needs to be transparent, scientifically optimized, and associated with real performance lift.
As Chief Technology Officer at Turn, Xuhui Shao focuses on the power of optimization, machine learning, and advanced analytics solutions in driving new business models, products, and services across all industries. Xuhui is responsible for architecting the machine learning and optimization technology to deliver the most effective data-driven digital advertising in the world. He is passionate about the dynamic online advertising community and works closely with industry leaders developing data transparency and consumer privacy protection.
For the last 12 years, Xuhui has practiced research and development in machine learning, statistical theory, and computational intelligence for Fortune 100 companies in various industries from banking, finance, online retailing, healthcare, insurance, marketing, and online advertising. As the lead inventor and co-inventor of three awarded patents in the areas of advanced analytics and optimization, Xuhui is a recognized expert in harnessing data and transforming analytics into actionable insights and optimization strategies.
He earned his bachelor's and master's of science degrees from Tsinghua University, Beijing, and his Ph.D. in electrical engineering from the University of Minnesota.
US Consumer Device Preference Report
Traditionally desktops have shown to convert better than mobile devices however, 2015 might be a tipping point for mobile conversions! Download this report to find why mobile users are more important then ever.
E-Commerce Customer Lifecycle
Have you ever wondered what factors influence online spending or why shoppers abandon their cart? This data-rich infogram offers actionable insight into creating a more seamless online shopping experience across the multiple devices consumers are using.
September 9, 2015
12pm ET/9am PT
September 16, 2015
12pm ET/9am PT
September 23, 2015
12pm ET/ 9am PT