In June, a shopper scrolls past a video ad showing how a new moisturizer looks on the skin. A few weeks later, another ad brings her to the brand’s website, where she browses several products but leaves without purchasing. By December, she needs a refill. She remembers the brand, searches for it by name, and finds a holiday promotion offering 20% off.
The promotion might’ve closed the sale, but it didn’t create the demand on its own. In a standard report, the shopper’s branded search might receive most of the credit, even though the brand had been building familiarity for months.
We’ve previously explored how Demand Gen and YouTube can build audiences that convert in Q4. Creating early demand is only the beginning. Marketers also need a way to distinguish a passing impression from developing interest and, eventually, purchase intent.
Audience Layering Distinguishes Interest From Intent
An “all website visitors” list is technically an audience, but it’s not much of an audience strategy.
Someone who watched part of a video shouldn’t automatically receive the same message as someone who returned to the website twice and abandoned a cart. The video viewer has shown interest. The cart abandoner has shown intent. Combining them into one group ignores the most useful difference between their behavior.
Audience layering separates users according to their relationship with the brand:
New prospects haven’t meaningfully interacted with the brand.
Engaged users might’ve watched a video, interacted with an ad, or visited the website.
High-intent users have returned to the site, explored products, or abandoned a cart.
Customers include previous purchasers, high-value customers, and lapsed customers.
Q2: Build First-Party Audiences Early
Demand Gen, YouTube, and other upper-funnel campaigns can introduce a brand before someone is actively searching for it. A shopper watching an ad in June might not be ready to buy, but her engagement offers an early indication of what exactly captured her attention.
Not every video viewer will become a customer, and audience size alone doesn’t indicate value. Separating video viewers, website visitors, category and product visitors, repeat visitors, leads, and purchasers allows marketers to see which groups continue engaging as the year progresses.
Audience duration also requires judgment. Google Analytics allows membership durations of up to 540 days, but someone who briefly visited six months ago might be less relevant than someone who returned last week. The appropriate duration should reflect the brand’s buying cycle and typical research period.
That research period can extend further than the final conversion date suggests. The National Retail Federation found that 42% of 2025 holiday shoppers planned to begin browsing or buying before November, while 60% expected to finish in December. This extended timeline gives early familiarity time to develop into consideration.
Q3: Refine Audience Segmentation Before Holiday Competition
By Q3, advertisers can begin separating people who briefly encountered the brand from those who continued engaging with it. Did video viewers eventually visit the website? Did category visitors return to individual product pages? Which messages attracted people who kept exploring, rather than those who produced an inexpensive click and disappeared?
These patterns also guide creative strategy. A person encountering the brand for the first time might respond to a lifestyle, problem, or use case. Someone comparing products might need more specific information: benefits, demonstrations, reviews, or proof that the product is worth its price.
Q3 is also the time to refresh Customer Match lists, confirm that acquisition campaigns are excluding existing customers where needed, and check whether key audiences are large enough to activate. By the time Q4 begins, the window for discovering which audience differences actually influence performance is already narrowing.
Q4: Use Audience Targeting Across Google Ads
By December, the shopper knows the brand exists, and she’s deciding whether to purchase from it. Another broad introduction does little to answer the questions that remain.
High-intent users might respond to product-specific creative, promotions, availability, or urgency. Engaged users who have not yet explored products might still need reviews and differentiators. Existing customers might be better suited for gifting, complementary products, or reactivation messaging.
The audiences can also serve different functions across Google Ads:
In Search, audiences can be added in Observation mode to compare performance without limiting keyword reach.
In Performance Max, first-party audiences can be used as signals that help guide Google’s optimization. They don’t restrict delivery exclusively to those users.
In Demand Gen, website visitors, video viewers, and customer data can support re-engagement. Advertisers can also create Lookalike segments to reach new users who resemble their first-party audiences, a capability currently unique to Demand Gen. Beginning in March 2026, Google started shifting Lookalikes from a firm targeting constraint toward an AI-powered audience signal that guides optimization rather than strictly limiting reach.
In YouTube, video ad sequence campaigns can deliver creative in a defined order, introducing the brand first and following with more specific messaging as the viewer progresses through the sequence.
Through Customer Match, advertisers can distinguish acquisition from retention and create more relevant campaigns for current or lapsed customers.
The platform mechanics matter, but the underlying principle is simple: the follow-up should reflect what the customer has already done. Audience layering loses most of its purpose when everyone receives the same holiday promotion in the same way.
Measure the Progression, Not Only the Final Click
Nielsen recommends measuring upper-and lower-funnel activity separately because each stage serves a different purpose. In Q2, engagement and audience growth can show whether a campaign is attracting attention. By Q4, revenue, conversion rate, CPA, and ROAS become more important.
Attribution reporting can identify the interactions that preceded a conversion, but it can’t prove that every Q4 purchase was caused by a Q2 impression. Marketing platforms cannot reconstruct every customer’s journey with perfect certainty, and advertisers shouldn’t waste time pretending they can. Audience layering provides a practical way to recognize that different behaviors carry different levels of intent and respond with greater relevance.
Build the Handoff Before It’s Needed
A branded search or holiday promotion might produce the final sale, but neither explains how the brand entered the shopper’s consideration. Building audiences in Q2, refining them in Q3, and activating them in Q4 connects those earlier interactions to the conversion strategy. The final click is the conclusion of that work, rather than the entire customer journey. Ready to put your Q2 and Q3 audience data to work in Q4? Let’s talk.
Sources
Beyond Bottom-Funnel: How Demand Gen and YouTube Build Audiences That Convert in Q4
Nielsen: Why Focusing on the Entire Marketing Funnel Is Key for Long-Term Brand Growth
This blog was edited with the help of Claude