DV360: Unlocking 4.0x ROAS in 2025 E-commerce

Getting a good return from programmatic execution in DV360 isn’t about the basic setup. It’s about how you use audience signals, bidding, and creative together to hit real business goals, like when you have to squeeze every last sale out of a Q4 budget. So how do you get past the platform’s surface-level features and start running campaigns with serious efficiency?

Campaign Teardown: Driving Q4 E-commerce Sales with DV360

In Q4 2025, we had to maximize sales for a direct-to-consumer electronics brand and its new smart home device line. We were targeting a return on ad spend (ROAS) of 4.0x or higher, with a hard cost per conversion (CPC) ceiling of $35. We had a total budget of $250,000 to spend over eight weeks, from October 1st to November 26th, and we leaned heavily on DV360’s advanced tools to get there.

Strategy and Planning: Beyond Basic Retargeting

We built a full-funnel strategy right inside DV360, covering awareness, consideration, and conversion. For awareness, we went after broad interest segments and lookalikes built from existing customer data. The consideration phase zeroed in on people who had already checked out competitor content or were showing high-intent signals in the smart home category. Then the conversion phase was all about retargeting website visitors, cart abandoners, and anyone who’d engaged with our previous ads.

A huge piece of this was getting deep into the brand’s first-party data. We integrated their customer relationship management (CRM) data, including purchase history and customer lifetime value (CLTV) segments, directly into DV360. This allowed us to build audience lists with incredible granularity, moving from simple behaviors to potential future value. For instance, we created a “High-Value Lapsed Purchaser” segment for customers who had spent over $500 in the past two years but hadn’t bought in six months. Targeting this specific group, instead of a generic “past purchaser” list, is what a platform like DV360 is really for.

Creative Approach: Dynamic Messaging for Each Funnel Stage

Our creative had to speak to people at each stage of the funnel. For top-of-funnel awareness, we ran short, snappy video ads showing the lifestyle benefits of the smart home gear, focusing on solving problems instead of listing tech specs. For consideration, we used carousels and display banners that highlighted key features and positive customer reviews. The conversion-focused creatives were blunt, pushing specific product offers, “Limited Stock” warnings, and direct calls to action (CTAs).

We also turned on dynamic creative optimization (DCO) in DV360. This let the system automatically mix and match headlines, images, and CTAs based on user signals like browsing history or location. For example, a user who just looked at a specific smart thermostat on the brand’s site would see an ad for that exact thermostat with a headline about energy savings. A user in a cold climate might see a headline about comfort. This kind of personalized ad assembly gave us a 25% higher click-through rate compared to our static ads.

Targeting Precision: Beyond Demographics

Our targeting strategy was a mix of different audience types available in DV360:

• First-Party Data Segments: As mentioned, CRM data was key. We sliced it into actionable lists like “High Intent Abandoners” (added to cart but bailed) and “Recent Product Viewers” (viewed a product page multiple times).
• Custom Intent Audiences: We built these from scratch by feeding the system keywords and URLs related to smart home tech and competitors. Think keywords like “smart thermostat reviews” or “home security systems.”
• In-Market Audiences: For broader reach in the consideration phase, we used DV360’s pre-built in-market segments for “Consumer Electronics,” “Home Automation,” and “Small Appliances.”
• Geographic Targeting: We concentrated our spend in major US metro areas known for higher tech adoption and disposable income, like Atlanta, Georgia, and even drilled down to specific neighborhoods like Buckhead and Midtown where the ideal customer was likely to live.

Layering these targeting methods let us start broad and then progressively shift budget toward the most qualified prospects as they moved down the funnel. We found that combining our own first-party data with our custom intent audiences consistently delivered a much higher conversion rate than relying on third-party segments alone, a detail that often surprises teams who are over-reliant on purchased data.

Bidding Strategy and Optimization: Algorithmic Prowess

We let DV360’s algorithmic bidding do most of the heavy lifting. For awareness, we started with “Maximize Impressions” and then moved to “Maximize Clicks.” For our mid-funnel and bottom-funnel campaigns, we went straight to “Maximize Conversions” with a target cost per acquisition (CPA). We set the initial target CPA at $45, giving the algorithm enough breathing room to learn before we tightened the screws.

We were in the account daily, monitoring performance and making adjustments. Here’s a quick look at what worked, what flopped, and what we changed.

What Worked:

• First-Party Data Segments: Our “High-Value Lapsed Purchaser” segment was a rockstar, pulling in a 6.2x ROAS. The CPC for this group was just $28, way under our $35 goal.
• Dynamic Creative Optimization: Ads using DCO saw a 25% higher click-through rate (CTR) than static versions, proving the personalization paid off.
• Video Ads for Awareness: The 15-second video spots had an average view-through rate (VTR) of 78%, which was great for building brand recall cheaply.

What Didn’t Work:

• Broad Interest Segments with Generic Creatives: These burned cash. We got tons of impressions (over 50 million) but a pathetic CTR (0.15%) and a CPC of over $60. They were completely inefficient for conversions.
• Single-Image Display Ads without Clear Offer: In the conversion phase, these just didn’t work. People needed a stronger reason to click, like a discount or a richer ad format.

Optimization Steps Taken:

After the first two weeks, we looked at the data and made some big changes:

1. Budget Reallocation: We pulled 30% of the budget from the failing broad interest segments and pushed it into our high-performing first-party and custom intent audiences. This is a point where being ruthless with the budget is non-negotiable. If a segment isn’t working, the funds have to move.
2. Creative Refresh: We killed the generic display ads. For the conversion phase, we built new creatives that included specific discount codes and customer reviews. We also tested new video hooks for the awareness campaigns.
3. Bid Strategy Adjustment: For our best-performing segments, we started walking the target CPA down to $32, forcing the algorithm to find even more efficient conversions. For segments that were struggling, we either paused them or temporarily bumped the target CPA up to see if we could find a pocket of performance before cutting them loose.
4. Landing Page A/B Testing: We ran a simple A/B test on our landing pages. The product-specific page delivered a 12% higher conversion rate than the generic category page for users coming from a product-specific ad. No surprise, but good to prove.
5. Frequency Capping: We started seeing signs of ad fatigue in some segments. We implemented stricter frequency caps (like 3 impressions per user per day) on our retargeting campaigns to avoid annoying people and keep the brand perception positive.

Results and Metrics

We ended up beating our initial goals across the board. Here’s the final scorecard:

The campaign drove almost 9,000 conversions at an average CPC of $27.80, smashing our $35 target. That final 5.1x ROAS was a direct result of a strategy built on good data, dynamic creative, and non-stop optimization. A recent IAB Digital Ad Revenue Report H1 2025 noted that programmatic channels keep growing, which just means that if you want to get good results, you have to get good at platforms like DV360.

My biggest lesson from this campaign? Programmatic is never “set it and forget it.” The space moves too fast. What worked last quarter might be useless now, so the only way to get these kinds of results is through continuous monitoring and iteration, like when we shifted 30% of the budget away from broad segments after just two weeks. It’s a non-negotiable part of the job.

Yes, the upfront work of integrating CRM data and building out granular audience segments is a pain, but it’s what pays for itself when you can turn a $250,000 budget into a 5.1x ROAS. When you really dig into DV360, you find that focusing on granular data, dynamic creative, and constant optimization lets you find efficient conversions at scale, like we did by bringing our CPA down to $27.80 on a $35 target. For more on this, check out how CX-driven ad creative can significantly boost CTR.