DV360 Audience Targeting: A Practical Guide for Media Teams

DV360 Audience Targeting: A Practical Guide for Media Teams

Audience targeting is where most of the performance — and most of the wasted spend — in Display & Video 360 (DV360) lives. Get it right and you concentrate budget on the people most likely to convert. Get it wrong and you either scale too narrow to matter or too broad to move the needle.

This guide breaks down how audience targeting actually works in DV360, the data sources you can use, and how to structure lists so campaigns stay clean, measurable and scalable.

The Audience Sources Available in DV360

DV360 pulls audiences from several places. Understanding the origin of each list matters, because it affects scale, freshness and how much control you retain.

First-party audiences

These are lists you own and bring into the platform:

• Website and app visitors captured via Floodlight tags or the Google tag.
• Customer Match lists built from hashed emails, phone numbers or addresses.
• CRM segments onboarded through your own data pipeline or a connected data provider.

First-party data is your most valuable asset because it reflects real customer behaviour and intent. It's also increasingly central as third-party cookies decline. If you're still building out your first-party foundation, our managed services team can help structure onboarding and Floodlight architecture.

Google audiences

Google provides ready-built segments derived from search and browsing signals:

• Affinity audiences — interest-based, useful for upper-funnel awareness.
• In-market audiences — users actively researching a product or category, strong for consideration and lower-funnel.
• Demographic segments — age, gender, parental status, household income.
• Detailed demographics and life events — milestones like moving home or graduating.

These offer instant scale but less differentiation, since every advertiser can access them.

Third-party audiences

Data providers offer segments through the DV360 marketplace. Availability and pricing vary by region and provider, and their long-term reliability depends on identity signals. Treat them as a supplement, not a foundation.

Combined audiences

This is where DV360 becomes powerful. You can layer multiple lists with AND / OR / NOT logic to build precise, intent-rich segments — more on this below.

Comparing Audience Types

Audience type	Typical funnel stage	Scale	Differentiation	Data control
First-party (site/CRM)	Mid–lower	Limited by your traffic	High	Full
Customer Match	Lower / retention	Depends on list size	High	Full
In-market	Mid–lower	Large	Low	None
Affinity	Upper	Very large	Low	None
Demographic	Any (layer)	Very large	Low	None
Third-party	Varies	Varies	Medium	Limited

Building Effective Combined Audiences

Single lists rarely give you the precision you want. Combined audiences let you express real targeting logic.

Use AND to narrow

Stack signals to increase relevance. For example:

• In-market for your category AND a specific income demographic.
• Site visitors AND an affinity segment that matches your brand positioning.

Each AND condition reduces reach, so monitor that you retain enough scale to spend the budget.

Use OR to broaden

Group related segments when any one of them signals interest:

• Multiple in-market segments across adjacent categories.
• Several first-party lists representing different high-value behaviours.

Use NOT to exclude

Exclusions are the most under-used lever in DV360. They prevent waste and protect the customer experience:

• Exclude recent converters from prospecting campaigns.
• Exclude existing customers from acquisition budgets.
• Exclude bounced or low-quality traffic from retargeting.

A disciplined exclusion strategy often improves efficiency faster than any new prospecting list.

Structuring Audiences Across the Funnel

Think of audience targeting as a system, not a set of isolated lists. A clean funnel structure keeps reporting honest and prevents overlap between line items.

Prospecting

• Google in-market and affinity segments layered with demographics.
• Lookalike-style expansion using Google's similar-audience and optimised targeting features.
• Third-party segments where they genuinely add signal.

Retargeting

• Segment by depth of engagement: product-viewers vs cart-abandoners vs newsletter sign-ups.
• Apply recency windows so messaging matches how recently someone interacted.
• Always exclude recent purchasers unless you're running a deliberate cross-sell.

Retention and cross-sell

• Customer Match lists built from your CRM.
• Segment by product ownership, lifetime value or churn risk.

Mapping list logic to funnel stage keeps your account readable and makes optimisation decisions obvious. If you'd rather have this architected and maintained for you, explore our co-managed services.

First-Party Data Is the Long-Term Advantage

As signal loss reshapes programmatic, first-party data becomes the durable differentiator. Practical priorities:

• Instrument thoroughly. Ensure Floodlight and the Google tag fire on every meaningful action, not just conversions.
• Build a membership-duration strategy. Match list durations to your buying cycle — short for impulse categories, long for considered purchases.
• Refresh Customer Match regularly. Stale lists decay; automate uploads where possible.
• Consolidate identity. Bring CRM, site and app signals into consistent segments rather than fragmented one-off lists.

First-party audiences also feed better modelling. DV360's optimisation features perform best when they're learning from clean, high-intent seed data.

Activation Best Practices

Once lists are built, activation quality determines results.

• Give each audience enough scale. Over-layering AND conditions can starve delivery. Check estimated reach before launch.
• Avoid audience overlap between line items. Overlapping lists bid against each other and distort attribution. Use exclusions to keep line items mutually exclusive.
• Match creative to audience. A cart-abandoner and a cold prospect should never see the same message.
• Let optimised targeting expand cautiously. It can find incremental users, but monitor that it doesn't drift into irrelevant inventory.
• Respect frequency. Even a perfect audience underperforms if you fatigue it. Set frequency caps at the campaign or insertion-order level.

Common mistakes to avoid

• Running prospecting and retargeting from the same list.
• Forgetting to exclude converters, inflating apparent conversion rates.
• Relying solely on third-party data with no first-party foundation.
• Building narrow combined audiences that never spend their budget.

Measuring Audience Performance

Audience targeting only improves if you measure it properly. In DV360:

• Compare performance by audience list to identify your most efficient segments.
• Watch for audience overlap that skews cost-per-action reporting.
• Use incrementality thinking — retargeting a hot audience will always look efficient, but that doesn't prove it drove the sale.
• Reallocate budget toward the segments delivering genuinely incremental results.

Strong measurement discipline turns audience targeting from guesswork into a repeatable process. To review your current setup with a specialist, talk to an expert.

Bringing It Together

Effective DV360 audience targeting comes down to a few principles: build on first-party data, use combined audiences to express real intent logic, apply exclusions relentlessly, and measure incrementality rather than surface-level efficiency. Structure your lists to mirror the funnel, keep line items mutually exclusive, and give each audience enough scale to matter.

Done well, audience targeting is the difference between a DV360 account that quietly wastes budget and one that compounds performance over time.

Ready to Sharpen Your Targeting?

Whether you're building a first-party data strategy from scratch or refining an established account, our team can help you design an audience framework that performs. Get in touch to discuss your goals and how we can support your DV360 activation.