Ready to boost your digital presence?


Test

Google’s Targeting Accuracy Issues

October 2, 2024

Recently, we’ve noticed a growing number of audiences in banner campaigns falling outside the established targeting parameters. This trend has…

Google’s Targeting Accuracy Issues

October 2, 2024

Recently, we’ve noticed a growing number of audiences in banner campaigns falling outside the established targeting parameters. This trend has…

Polina Kedis

Recently, we’ve noticed a growing number of audiences in banner campaigns falling outside the established targeting parameters. This trend has been observed across both Display & Video 360 and Google Ads platforms. After a thorough analysis, we’ve discovered that the narrower we define the target audience, the higher the percentage of people outside that audience who are reached. In some cases, this non-target reach exceeds 50%, making this issue worth discussing. In this article, we’ll share the solutions we’re implementing for our campaigns.

Platform Comparison

An analysis of data from both platforms showed that the percentage of missed targeting is nearly the same between Google Ads and DV 360. This indicates that the issue is not platform-specific but rather a general problem with targeting accuracy.

Support Findings

To better understand the situation, we reached out to our managers and Google’s internal support team. The responses were consistent: ad delivery and reporting interpret demographic information differently. When showing ads with demographic targeting (whether in auction or reservation), Google uses demographic profiles based on both declared and inferred signals. However, these models are not always accurate on an individual level. A user may behave in ways that are atypical for their demographic group, leading to impressions outside the target audience. There are several reasons for this:

  1. Incorrect or incomplete demographic information: Users may misreport their data or choose not to provide it.
  2. Shared devices: Multiple people may use the same device, making it harder to accurately identify the demographics.
  3. Cookies: Not all users are logged in, making it difficult to determine their exact age or gender. This explains why, even when targeting a specific demographic group, you may still receive impressions outside the intended audience.

Our Approach to Fixing the Issue

To improve targeting accuracy, we’ve made several changes in how we plan and set up banner campaigns. We now strictly use only the client’s designated target audience, avoiding broader demographic segments. For example, if the target audience is 25-43 years old, we no longer include wider age groups, as we did in the past. We also minimize the use of optimized targeting, as it often leads to significant deviations from the target audience. When planning campaigns, we rely on the latest data on audience reach and incorporate this information into frequency settings for banner ads. Additionally, we inform clients about this issue and caution them about potential deviations when working with narrow target audiences.

Conclusions

  1. There is a real issue with the significant discrepancy between the set target audience and the actual audience reached, and it has worsened in recent months. This is particularly noticeable in banner campaigns, where deviations can exceed 50%.
  2. Video ads, on the other hand, do not exhibit such critical issues — the deviation is only around 5-10%, especially on YouTube. Support has confirmed that such deviations are within the norm.
  3. This issue affects both Display & Video 360 and Google Ads. It is likely related to increased restrictions on cookie usage and the growing number of users with less specific demographic information, such as a broader age range of 18-44 instead of more defined groups like 18-24. It can also be partially explained by technical details provided by support.
  4. The problem is not limited to age targeting but also affects gender targeting. Even when targeting women, 20-30% of impressions may go to men, as confirmed by tracking data.
  5. We expect this issue to spread to other advertising platforms outside Google. However, within YouTube, such deviations seem unlikely for now.

Share with those who need it

Get deeper into digital!

Subscribe to the newage. digital digest and receive exclusive bonus content

Leave a Reply

Your email address will not be published. Required fields are marked *