Audience Targeting with Display & Video 360 (DV360)

Despite the complexity of today’s digital landscape, your goal likely hasn’t changed much: find your ideal audience, and reach that audience at scale — all while allowing for optimizations to reach maximum efficiency and ROAS. Historically, this was a guessing game based on large geo data sets and less-than-accurate “journal” data. Digital changed the game, but DSPs like Google’s Display and Video 360 (DV360) are adding new audience features regularly. So what does this mean for marketers in 2022?  

Enter the hyper-empowered customer. She is incredibly tech savvy and, more importantly, understands what tech can do for her. This has heightened her expectations for assistance.  She expects experiences that are assistive, useful, personal, and frictionless. 

And how do you assist this hyper-empowered customer? You understand her intent.

DV360 has many programmatic audience targeting features to not only find the right audience, but to understand and take action based on their intent. 

Types of DV360 Audiences and Use Cases

Affinity audiences are useful to advertisers looking to raise awareness and drive consideration among groups that have a strong interest in their products. These are based on user browsing patterns as well as content consumption. This allows marketers to show ads to people who are likely to be enthusiasts – examples include Health & Fitness Buffs, Foodies, and Sports Fans. There are 200+ signals to choose from when selecting an affinity audience in the DV360 platform. 

There are several best practices and performance considerations to keep in mind when using affinity audiences.

• Engagement metrics are the best measure of success. Use CTR or CPCV (across the entire video buy). 
• CPA can decrease with affinity, when compared to run of network.
• Combining with demographics can help achieve reach against the target audience (example: layer Car Enthusiasts with household income).
• Combining with Viewable Impression Bidding can achieve only viewable impressions against a target audience.
• Reach could increase when layering with third-party data, but performance could suffer if the third-party list is not of sufficient quality.

Custom affinity audiences are, well, just what they sound like: an opportunity for marketing to build a custom audience that leverages Google’s affinity audience data. This lets you target an audience by the media they consume online or by the products and services they may be interested in. This type of audience is best for the consideration or purchase phase. 

In-market audiences help marketers find users based on their intent to purchase online.

The active intent of users is updated in real time, based on behavior across the Google-owned properties. These customers have demonstrated behavior similar to purchasers. 

Designed for the consideration and purchase phase, an example would be a combination of the following user behaviors:

• User searched for shoe retailers “near me” in Maps and
• User researched specific shoe styles on Google Search and 
• User watched YouTube videos of shoe reviews

There are a few specific considerations for using in-market audiences.

• CPA or conversions are the best measure for successful performance.
• Do not expect CPAs to be the same for remarketing campaigns – they will likely be higher since you are reaching net-new customers.
• In-market segments are smaller subsets of affinity segments.
• Do not combine with affinity audiences — this will greatly increase reach, but performance will suffer.

Custom intent audiences are specifically designed to help acquire new customers in the purchase phase. This is great for marketers looking for increased granularity from in-market categories. This works with DV360 to build an audience of users who show signs of intent in the products or services. However, this is driven by your described inputs. Keywords allow DV360 to have a focal point, while URLs and apps provide additional context to build a complete audience. 

Proximity targeting is perfect for building an audience of potential buyers that are near to your physical retail locations. This feature allows marketers to use geolocation data to:

• Target users near a specific location or store.
• Taylor ads or creative elements to be more relevant to the user’s location.
• Deliver offers to those near a store to encourage offline purchase.

First-party audiences, in the age of deprecated cookies, are even more important and powerful than ever before. DV360 offers a wide variety of first-party targeting options that allow you to use what you know about your customers to reach them with the right message. This also means excluding current customers to only reach new customers. First-party audiences can be customers from your CRM, DMP, Floodlight Remarketing, or Analytics 360 lists. They can also come from engagements on Youtube, Google Ads, and Search Ads 360. 

Google Analytics 360 audiences are available in DV360 for remarketing based on many parameters. Synchronizing DV360 and Google Analytics 360 (GA360) also provides unified reporting, automation, and data activations across multiple channels. This is most useful in the form of engagement or conversion data for use in remarketing or prospecting (similar audience). At its most basic level, funnel segmentation or custom GA data can be shared. However there are more advanced features such as Sessions Quality, Smart Lists, and Conversion Probability.  

Lookalike modeling or similar audiences are a great way to reach individuals who share interest profiles similar to your existing customers. By targeting similar audiences, you can reach more potential customers without all the guesswork. This works by combining first-party audience data, Google’s audience targeting data, and the lookalike modeling algorithm to build a custom audience of users who are likely to engage, click, and convert with your ads. There are limitations, however.

• The first-party list must have active users in the past 30 days.
• The first-party list must contain more than 500 users.
• Similar audience sizes are not taken into account in the DV360 reach widget.

Third-party audiences, as the name implies, allow you to create audience targeting in Google using a wide variety of rich third-party data. DV360 has access to over 100,000 audience segments from many trusted data sources, including data from 35+ GDPR-compliant providers. 

Third-party data gives marketers the ability to build audiences with additional data points that may not be available with their known user base (first-party data). You can create an audience for lifestyle segments — this could include geo-location data, such as users who visit the gym or grocery shop at least once a week. It could also include purchase data, including sales data from Mastercard and Visa that identifies shoppers of a specific category or store. It can also include very specific research data. For example, if a client ran research with Experian and found their target consumer is Mosaic segment G28 – Modern Parents, they can target that segment online.

Third-party audiences can add tremendous benefit, but there are some best practices and considerations to note. 

• Depending on your objectives, some data sources are better than others. Testing and constant optimization are the best methods here. It’s best to try as many third-party audiences as possible and fine-tune based on effectiveness. You’ll have a short list of favorites in no time!

• For most lists, the CPM price averages $1, but this could be as high as $3-4. 
• Use common sense – if a list says it’s comprised of Rolex owners in the US, but the list has 500M uniques in the US alone, then something isn’t right.

DV360 Audience Targeting Best Practices

Layering is a great way to expand or refine audiences with multiple strategies or features. You’ll want to keep your campaign objectives in mind when layering. For example, awareness and performance goals should be handled differently. For awareness, you may want to exclude known customers, but for some performance segments, you may want to only include known purchasers. Adding all segments related to your target audience isn’t always a best practice. 

There are three common layering types you should consider when building or optimizing audiences. 

• Exclusions
  • An often overlooked layering type, this will exclude audience segments from your targeting. This is great for awareness campaigns when you’re trying to reach net-new customers and don’t want to waste impressions on your current customer list. There are no limitations with this layering type – all audiences are supported.  
• “OR” combination
  • This allows you to target users who match multiple audience segments. This is also a great layering type for awareness campaigns, since users only need to match one target segment to be active. This will expand reach, but could negatively impact engagement, depending on the quality of chosen segment.  
• “AND” combination
  • This type offers the ability to build hyper-niche audience segments for pinpoint messaging. For example, you could build an audience of known purchasers of a specific product who also visited your website in the past week. This will affect scale, as a user must meet all conditions to be active in the audience segment. Affinity and in-market audiences are available in this layering type. 

With layering, keep in mind that data providers typically sync lists every 30-90 days. This lag could be an issue for performance campaigns in time-sensitive categories.  

Demographics in DV360 are a combination of user signals from across Google’s properties like Google Search, Youtube, Google Maps, and more. Demographic audiences are best measured using reach and engagement success metrics — they are not designed to drive hard conversions like transactions.

Some tips to keep in mind:

• Demographics layered with any other audience defaults to an AND combination. For example, if you layer Car Enthusiasts AND Male, you’ll target only Male Car Enthusiasts. 
• Demo data is not perfect, so you may want to include “unknown” (age or gender) in order to increase reach. “Unknown” can be a large percentage of the total segment, so don’t ignore it. This segment typically has a lower CPC since it’s not as popular as targeting a specific age/gender.

Combined audiences allow marketers to create a single audience from multiple first-party and third-party data sources, as well as DV360’s affinity and in-market audiences. These are easily saved and reused across different campaigns, making them super handy for viral or time-sensitive campaigns. This feature takes the form of the Combined Audience Builder, which is an interactive interface providing real-time feedback based on your segment selections.

There are several key best practices and limitations for combined audiences.

• Include and Exclude first-party, third-party, and Google audiences. 
• Combined audiences cannot include YouTube lists.
• Combined audiences cannot be targeted for TrueView campaigns.
• Only existing line-item level AND/OR/NOT combinations are currently available.
• At the line-item level, combined audiences cannot be mixed with other audiences. 
• You can only positively target combined audiences (exclusions are managed at the audience creation level).

Activity-based audiences allow marketers to build audiences based on audience activity across campaigns. These audiences have two types: campaign activity and frequency cap.

• Campaign activity audiences are based on the number of clicks, conversions, and impressions that led to conversions of first-party lists from Floodlight activities within a given advertiser in DV360.  
  • This is great for reaching an audience that has previously engaged with a display or video campaign.
  • These audiences are not available for YouTube campaign activity (for now).
  • Strong use cases include sequential messaging or story campaigns. For example, a smart thermostat company could use this tactic to introduce potential customers to money-saving messaging before getting into technical details about its products.  
• Frequency cap excludes users based on the number of impressions they were served across media, channels, and screens.
  • Control the frequency of exposure to a given user, across campaigns, deals, and inventory types in a few clicks.
  • These audiences are not available for YouTube impressions (for now).
  • Use case: You’re a shoe brand with many campaigns and products. You may want to limit the frequency across all campaigns so you don’t oversaturate (annoy) your loyal customers. 

DV360 Audience Targeting: Putting It All Together for Full-Funnel Marketing 

The above audience features are best when used in a full-funnel or customer journey marketing strategy. While your funnel or journey may have more segments, most can be broken down into three main phases: awareness, consideration, and purchase. 

In the awareness phase, it’s best to rely mostly on Google audience targeting data in the form of demographics and patterns. With this audience targeting, you should be focused on reach and frequency metrics that align with your lower-funnel goals. You can start to bring in affinity audiences and third-party audiences during the awareness phase, depending on your industry and goals. 

In the consideration phase, affinity audiences are still a great fit and can bridge the gap between awareness and consideration targeting. Third-party data is also still a great strategy during this phase. There are a ton of great third-party audiences from many providers like Bluekai, VisualDNA, Nielsen, Kantar, and more. Moving further into the consideration phase, many marketers start to build custom affinity audiences based on past performance or niche targeting. This is a great way to build custom solutions that fit your exact targeting needs. As we move closer to the purchase phase, it’s a good idea to start looking at similar audiences and in-market audiences. Both provide a great crossover between a consideration user and a purchaser, depending on how they are set up and used. 

Once you’re in the purchase phase, custom intent is your primary Google audience. Outside of that, you’ll want to start leveraging first-party data you’ve collected during the purchase process — things like CRM audience, Floodlight tags, DMP audiences, Analytics 360 audiences, and Google Ads remarketing audiences. Since these can be built from previous purchasers, you’re able to target them directly or expand for a broader reach.

Want to see the DV360 platform in action? Schedule a live demo with our experts.