GEO vs. AEO Video Ads: $75,000 ROAS in 2026

Deciding between GEO (Geographic) and AEO (Audience-centric) video ad strategies requires a clear understanding of campaign objectives and target audience behavior. In 2026, with increasing platform sophistication and data privacy shifts, the choice between these two approaches significantly impacts return on ad spend and overall campaign effectiveness. Which approach delivers superior results for your next video ad push?

I recently oversaw a video ad campaign for a client, “Urban Eats,” a burgeoning chain of fast-casual restaurants specializing in locally sourced ingredients. They had three new locations opening concurrently across the Atlanta metropolitan area: one in Midtown, another near Emory University in Druid Hills, and a third in the bustling Buckhead Village district. Their primary goal was to drive foot traffic and initial online orders during the first three months of operation for each new spot. The total budget allocated for this video ad push was $75,000 over a 90-day period. Our target cost per lead (CPL), defined as an online order or a confirmed store visit via loyalty app signup, was $8.00. We aimed for a return on ad spend (ROAS) of 3:1.

The Strategy: A Hybrid Approach with Phased Rollout

We opted for a phased, hybrid strategy, starting with a strong GEO-focused approach for the initial launch period, then gradually incorporating AEO elements as we gathered more first-party data. This wasn’t a choice between one or the other, but an understanding of how they complement each other. The core idea was to saturate the immediate vicinity of each new restaurant first, then expand outward to relevant audiences. This allowed us to build local awareness quickly.

Phase 1: Hyper-Local GEO Targeting (Days 1-30)

For the first month, our focus was almost exclusively on geographic proximity. We defined precise geofences around each restaurant location, typically a 2-mile radius, and targeted users within these zones on platforms like Google Ads (YouTube) and Meta Business Suite (Facebook/Instagram Reels). We used Foursquare’s location intelligence to refine our understanding of foot traffic patterns around these areas. This wasn’t just about drawing a circle on a map. It involved layering in data about residential density, office buildings, and public transport hubs.

Creative Approach for GEO Targeting

The video creative for this phase was highly localized. For the Midtown location, we filmed shots featuring the skyline, people walking through Piedmont Park, and quick cuts of the restaurant interior with “Midtown’s New Flavor” overlays. The Emory location’s ads highlighted campus life, students studying, and the convenience of a quick, healthy meal between classes. Buckhead’s creative emphasized its upscale environment, featuring shoppers and local landmarks like the Atlanta History Center. Each video was 15-seconds, designed for quick consumption on mobile devices, and ended with a clear call to action: “Order Now for Pickup or Delivery” or “Visit Us Today.” We also included a unique QR code for each location, linking directly to their respective online ordering pages and offering a first-time customer discount.

Our budget allocation for Phase 1 was $30,000, split evenly across the three locations. We ran these campaigns with a daily budget of $333 per location. Impression volume was high, averaging 1.2 million impressions per location over the month. The click-through rate (CTR) for these GEO-targeted ads was surprisingly strong, hitting an average of 1.8%. This indicates the relevance of the localized creative. Our CPL during this initial phase came in at $9.20, slightly above our target of $8.00, but still acceptable given the newness of the locations. The immediate ROAS was 2.5:1, driven primarily by online orders and initial loyalty sign-ups.

Phase 2: Expanding with AEO (Days 31-60)

Once we had a month of data, including customer demographics from online orders and loyalty program sign-ups, we began to integrate Audience-centric (AEO) targeting. We used the first-party data to create custom audiences and lookalike audiences on both Google and Meta platforms. For instance, customers who ordered vegetarian options frequently became the basis for a lookalike audience interested in healthy eating and plant-based diets. We also targeted broader interests such as “foodies,” “local dining,” “healthy lifestyles,” and “university students” (for the Emory location). The geographic boundaries were expanded slightly, moving from a 2-mile radius to a 5-mile radius, allowing us to reach a wider, yet still relevant, audience.

Creative Approach for AEO Targeting

The creative for Phase 2 shifted to focus more on the product and lifestyle benefits, rather than just the location. We produced a series of 30-second videos showing the freshness of ingredients, the speed of service, and the lively atmosphere of the restaurants. Testimonials from early customers (with their explicit permission, of course) were also incorporated. One ad might feature a busy professional enjoying a quick, healthy lunch, while another highlighted friends gathering for dinner. We still maintained a subtle nod to the Atlanta area in the background, but the emphasis was on the universal appeal of good food. The call to action remained consistent: “Experience Urban Eats, Order Now!”

Budget for Phase 2 was $25,000. Our impressions increased to 1.8 million per location, reflecting the expanded audience reach. The CTR saw a slight dip to 1.5%, which is expected when moving from hyper-local to broader targeting. However, the important metric here was the cost per conversion, which improved significantly. Our CPL dropped to $7.50, finally beating our target. This improvement came directly from the more refined audience targeting, which allowed platforms’ algorithms to find individuals more likely to convert. The ROAS climbed to 3.2:1, demonstrating the power of audience intelligence.

Phase 3: Optimization and Refinement (Days 61-90)

The final month involved continuous A/B testing of creative variations, audience segments, and bid strategies. We discovered that shorter, 10-second bumper ads performed exceptionally well for retargeting audiences who had previously engaged with our video content but hadn’t converted. We also identified specific times of day when our ads performed best for lunch and dinner rushes, adjusting our ad scheduling accordingly. We implemented a dynamic creative optimization strategy, allowing the platforms to automatically serve the highest-performing video variations to different audience segments. This iterative process is non-negotiable for maximizing ad spend.

We even experimented with a limited campaign targeting specific office buildings in downtown Atlanta, using IP address targeting (a feature available on some advanced ad platforms in 2026 for B2B contexts, though with strict privacy safeguards). This allowed us to reach professionals during their lunch breaks with relevant offers for our Midtown location. This particular micro-campaign, while small in budget, delivered a CPL of $6.00.

Budget for Phase 3 was $20,000. Overall impressions for this period reached 2.1 million per location. The CTR rebounded to 1.7% as our targeting became more precise. More importantly, our cost per conversion (CPL) dropped further to $6.80, and the final ROAS for the entire campaign reached 3.8:1. This sustained improvement was a direct result of combining the best elements of both GEO and AEO, continuously optimizing based on real-world performance data.

What Worked and What Didn’t

The hyper-localized GEO targeting in Phase 1 was instrumental in building initial awareness and driving immediate foot traffic. It established Urban Eats as a local presence quickly. The highly specific creative resonated deeply with those physically close to the restaurants. What didn’t work as well was expecting this narrow targeting to scale indefinitely. Its efficiency diminishes as the “saturation” point for the immediate vicinity is reached.

The gradual introduction of AEO targeting in Phase 2 was a key success factor. It allowed us to expand our reach to interested individuals who might not be immediately adjacent to a restaurant but were still within a reasonable travel distance or delivery zone. Relying solely on AEO from the start, however, would have likely resulted in a higher initial CPL as the platforms learned our ideal customer without the strong geographic anchor. The combination of first-party data for lookalikes proved far more effective than relying on broad interest categories alone. We initially tested a broader “food delivery app users” audience, but it performed poorly with a CPL of $12.50, demonstrating the need for more granular audience definition.

One minor misstep was underestimating the creative production requirements for such a phased approach. Developing distinct video assets for each phase and location initially stretched our resources. In hindsight, we should have allocated more budget to creative development upfront, recognizing that dynamic content is paramount for both GEO and AEO strategies. We ended up recycling some assets more than intended in Phase 3, which, while effective, limited our ability to test entirely new concepts.

Optimization Steps Taken

• Daily Bid Adjustments: Our team monitored performance metrics daily, adjusting bids for specific ad sets and audience segments to ensure we were spending efficiently.
• Negative Placement Lists: We continuously refined our negative placement lists on YouTube, excluding channels or videos that showed low engagement or brand safety concerns.
• Conversion Window Analysis: We analyzed the time-to-conversion for different ad types, finding that video view-through conversions often took longer than click-through conversions, which influenced our attribution modeling.
• Landing Page Optimization: We A/B tested different landing page designs for online ordering, finding that a simplified checkout process reduced abandonment rates by 15%.
• Geographic Exclusions: As our AEO targeting expanded, we proactively excluded areas known to have low population density or where delivery was not feasible, preventing wasted ad spend.

In the end, the choice between GEO and AEO for video ads isn’t a binary one. It’s about understanding your product, your market, and your customer journey. For Urban Eats, a multi-location business, the strategic layering of geographic and audience intelligence delivered superior results, proving that a well-executed hybrid strategy often outperforms single-focus campaigns. To master AEO algorithms for 2026 success, continuous learning and adaptation are key. Plus, the principles of local video ads played a significant role in our initial penetration. The effective use of AI creative drives ROAS, as seen in our refined targeting and dynamic optimization. For those looking to implement similar strategies, understanding how AI ad delivery boosts ROI is important.