Google AdWords Display Network Best Practices Guide

The Google Display Network is the largest global display ad network which allows advertisers to reach 90% of all internet users worldwide. To make the most of Google AdWords Display campaigns, marketers are increasingly relying upon programmatic advertising which uses automation and data to deliver the most relevant ads and efficient results by targeting the right people at the right moments, in the right context.

Programmatic advertising involves 3 objectives:

  1.  Reaching targeted audiences at the right moment
  2.  Matching the moment with the most appropriate message
  3.  Optimizing strategies for best performance  

This guide will help you understand the Display Network campaign best practices

Set the Display Campaigns Up for Success

  1. Test conversion tracking by generating conversions which should register across AdWords, Google Analytics and third party platforms, or you can confirm the tag is on the right page by viewing the page source of the conversion page.
  2. The Search Network and Display Network differ in how each identifies targets for marketers.  The Search Network uses keywords to capture explicit intention, while the Display Network uses signals to influence.
  3. Because of the differences, each is used separately for better control over bids, ads and targeting.
  4. If there isn’t enough time or resources to manage separate search and display campaigns, Google does offer the Search Network With Display Select option.
  5. Assigning credit for sales and conversions to touchpoints in a conversion path is very important. Attribution modeling refers to the rule or set of rules that determines how credit for sales and conversions is assigned to touchpoints in conversion paths. Using only last-click attribution puts advertisers at risk of underinvesting in display because their models may not account for the upper funnel conversions that a Google Display Network campaign may generate.
  6. It’s possible where viewers will see an advertiser’s display ad and don’t click on it but may remember the advertiser’s brand instead and search for the product or service at a later time, perhaps when they need it.  View-through conversions can be added to placement reports so that a conversion may be measured for a display ad which a viewer saw and didn’t click, but later caused a conversion. The view-through conversion metric can be used to optimize Display performance to identify high-performing placements.
  7. To gain a better perspective of how many conversions each channel initiated, assisted and completed, review the Assisted Conversion report in Google Analytics.
  8. A/B split testing  allows you to split users into two groups: those that were shown your display advertisements versus people who were not shown the display advertisements. A/B split testing can be used to understand how many more visits or conversions occurred as a result of the display advertisements.
  • Tip: Ensure all cross-device conversions are counted by making the AdWords conversion tracking tag fire unconditionally.
  • Tip: Smart Goals in Google Analytics is an easy way to allow you to use your best visits as conversions.

Reach the right audience at the right moment

  1. On average, consumers will visit mobile websites 6 times in the purchase process. It’s important to re-engage with them when a user is most likely to buy. Remarketing keeps advertisements in front of people who have previously visited your site or app in order to accomplish this goal.
  2. The most common sources for building a remarketing list consist of desktop and mobile website visitors. The top lists include all site visitors, category page viewers, cart abandoners, product page viewers and past converters.
  3. There are two main use cases for mobile app remarketing. The  first is to drive app re-engagement, the second is to drive mobile app conversions.  Re-engagement remarketing should focus on targeting lapsed or inactive users. To target conversions for mobile apps, advertisers should include product viewers or abandoners in their remarketing lists.
  4. Creating  video remarketing lists based off the YouTube viewers and specific actions they’ve taken on your YouTube channel is another way marketers can re-engage past site visitors and audiences. In order to re-engage with YouTube video viewers, the AdWords account and YouTube channel must be linked.
  5. Dynamic remarketing is best for businesses with a large, diverse portfolio of products. Dynamic remarketing provides viewers the most tailored advertisement possible, which is most likely to perform best based on a business’s product inventory,
  6. The similar audiences feature allows advertisers the ability to find people who share similar characteristics with their site’s current visitors. Using similar audiences can grow remarketing conversions by up to 38% compared to using remarketing alone.
  7. To find additional customers, advertisers can use contextual targeting to show ads  to viewers based on the content they are currently or recently viewed. Contextual targeting involves the use of keyword targeting and or in-market audiences. To choose keywords to use on the Google Display Network, advertisers can use top performing keywords from their Search campaigns, they can also use their keyword taxonomy to guide campaign structure, or add keywords based on more broad concepts and structure them into tighter themed ad groups.
  8. There are two targeting options to automate targeting with display targeting optimization, conservative targeting and aggressive targeting. Conservative targeting is used when advertisers want to drive additional conversions based on their business goals or Target CPA. Conservative targeting is the default setting. The other option is the use aggressive targeting which will allow advertisers to drive more conversions with some flexibility on the cost of the Target CPA. Aggressive targeting tends to perform best when an ad group has a significant and consistent amount of volume of conversions.
  • Tip: Consider creating similar audiences based on past purchasers or best-selling products and services.
  • Tip: Set up a separate campaign for similar audiences. It will allow for better management of bids, ads and targeting.
  • Tip: Avoid stacking multiple targeting methods for audience lists. The ads will not serve due to the restrictions of too many audience lists overlapping one another.

Match the message and bid to the moment

  1. Use both text and image ads in multiple sizes and formats. Advertisers who have actively used both text and image ads, on average see 40% more conversions without increasing cost-per-acquisition.
  2. Upload image ads in high-definition (HD) sizes. HD sizes are typically double sized so a 300*250 ad may be double sized to a 600*500 ad.
  3. Google will automatically resize the five most popular image sizes to fit on 95% of the placements on the Google Display Network. The five most popular ad sizes are:
    • 300*250
    • 728*90
    • 160*600
    • 320*50
    • 300*600
  4. Google recommends creating a separate display campaign focused on Gmail advertisements. Gmail advertisements can be found by selecting the specific placement for Gmail, mail.google.com. Gmail ads are created specifically for Gmail inboxes. 
  5. Optimize campaign bids using Target CPA or ROAS for increased performance. Automated bidding saves time, allowing advertisers to focus more on strategic optimization strategies. Automated bidding also allows advertisers to deliver better, more informed bids based on the user’s context as close to auction-time as possible.

Optimize for Better Display Performance

  1. Automated bidding will automatically optimize placements for advertisers.
  2. Use exclusions to avoid content that target customers may not find appealing.
  3. The first step in excluding underperforming site categories and placements is to audit the site categories. Most brands will desire to exclude sensitive content, error pages, parked domains, or forums.
  4. Optimize placements by identifying irrelevant and underperforming placements. Use statistically significant data to base decisions.