Editorial Team Key Takeaways
- We boosted ROAS by 12% on the “Autumn Harvest” campaign simply by targeting urban centers where specialty food stores were clustered.
- Our A/B tests proved that mobile-first ads with strong calls to action beat desktop-optimized versions, getting a 2.5% higher CTR.
- Tapping into partner APIs for real-time foot traffic data let us adjust bids in specific zip codes which cut our cost per conversion by 18% in the campaign’s second half.
- We put 15% of the initial budget into local influencer deals, and it paid off with a 30% higher engagement rate than our standard social ads.
- The campaign’s real secret was dynamic ad copy that changed with local event calendars, giving us a 7% conversion lift on certain product lines.
We recently ran a fall campaign for a gourmet food distributor that really shows what happens when you get your geo-targeting right as part of a larger market strategy. The “Autumn Harvest” campaign which we launched in September 2025, is a perfect example of how a tight focus can deliver big returns in a crowded market. The debate isn’t about whether geo-targeting works anymore. It’s about how carefully you execute it.
Campaign Overview: “Autumn Harvest”
For the “Autumn Harvest” campaign, our goal was to move a lot of seasonal specialty foods, artisanal preserves, organic baking mixes, gourmet coffee blends, in major Northeast metro areas. A generic national campaign would have just burned money on these niche products. So we concentrated our firepower on areas where demographic data pointed to people willing to buy premium food, specifically targeting consumers inside a 5-mile radius of specialty grocers and farmers’ markets. The whole thing ran for eight weeks, from September 1st to October 26th, 2025. We started with a $150,000 budget spread across paid social, search, and local display ads. Our targets were a 3:1 return on ad spend (ROAS), a 1.5% click-through rate (CTR) on display, and a cost per lead (CPL) under $12.
Strategy: Hyperlocal Focus and Dynamic Creative
The entire strategy hinged on hyperlocal targeting. We picked out 15 specific zip codes in Boston, New York City, and Philadelphia that screamed “target audience”: household incomes over $90k, tons of college-educated residents, and a clear pattern of searching for food-related content. We got this granular because 2025 data from eMarketer confirmed that the shift toward quality and local sourcing in food wasn’t just a fad. Inside those zip codes, we segmented even further by online behavior, going after people who’d recently searched for things like “artisan food delivery,” “gourmet gifts,” or “local produce.” We also layered on interests for known foodies, home bakers, and organic shoppers. The creative had to be just as smart. We built a set of dynamic ads that could pull in local landmarks or event info (like, “Perfect for your [Local Farmers Market Name] haul!”) and even change based on the weather. If it was a rainy day, the ads might push cozy, warm drinks, while a sunny day would get ads featuring picnic-friendly foods. This kind of real-time adjustment requires a pretty sophisticated ad-serving platform, but it’s worth it.
Implementation and Data Metrics
We attacked this from multiple channels. Over on Meta Ads Manager, we used carousel ads to show product pairings and recipes, targeting custom audiences we built from lookalikes of our best past customers. In search, we went after long-tail keywords with city names or neighborhoods, like “gourmet coffee beans brooklyn” or “artisanal preserves rittenhouse square.” Our display ads ran on the Google Display Network, placed on food blogs and local news sites. Here’s a look at how the performance numbers shook out:
| Metric | Target | Actual (Week 1-4) | Actual (Week 5-8) |
|---|---|---|---|
| Budget Spent | $150,000 | $70,000 | $80,000 |
| Impressions | 5,000,000 | 2,800,000 | 3,500,000 |
| CTR (Display) | 1.5% | 1.2% | 1.8% |
| CPL (Search) | $12.00 | $14.50 | $10.80 |
| Conversions | 12,500 | 5,800 | 9,200 |
| Cost Per Conversion | $12.00 | $12.07 | $8.70 |
| ROAS | 3:1 | 2.8:1 | 3.5:1 |
What Worked Well
The dynamic creative strategy was a huge win. Tying ad copy to local stuff like the Fenway Fall Festival (“Hosting? Grab our spiced cider mix!”) or just the day’s weather made the ads feel relevant. We watched the CTR on these dynamic display ads climb from a sluggish 1.2% to 1.8% by the second half, beating our goal. That personalization made the ads feel like a helpful tip, not just marketing noise. Our hyperlocal search strategy also came through, especially after the first couple of weeks. The CPL started a bit high, but once we got aggressive with our negative keywords and tweaked bid multipliers for specific neighborhoods, it dropped nicely. We found that search queries with both a product and a specific street or district converted 25% better than broader city-level searches which told us people were ready to buy from a local source that our D2C model could serve. On top of that, using our own first-party data (customer purchase history, email lists) to build lookalike audiences on social was incredibly effective. Those audiences always gave us a lower cost per click and a better conversion rate than the standard interest-based targeting.
Challenges and Optimization Steps
We definitely hit some turbulence in the first few weeks. Our display CTR was lagging, especially in the big metro zones where it felt like we were just shouting into the wind. Our initial CPL on search was also too high because we started with some keywords that were just too broad. To fix it, we made several quick changes:
- Refined Geo-Fencing for Display Ads: This was a critical adjustment. We tightened our geo-fences for display, focusing only on a 2-mile radius around known specialty food stores and fancy apartment buildings. We also plugged into a third-party API that gave us anonymized mobile location data, letting us pump more ads into those zones during peak shopping hours.
- A/B Testing Creative Variations: We ran tests on different calls-to-action and visuals. What we found was that lively, autumnal photos with direct CTAs like “Improve your fall baking!” did way better than boring product shots with a generic “Shop Now” button, giving us a 2.5% higher CTR on mobile. That discovery made us shift our entire creative process to be mobile-first.
- Adjusted Bid Strategies: We switched our search campaigns to automated bid strategies set to maximize conversions at our target CPL. We also told the system to bid higher for mobile users in dense urban areas, since that’s where people are doing their research (and buying) on the fly.
- Localized Influencer Partnerships: Around week five, we took 15% of our remaining budget and gave it to micro-influencers in Boston’s North End, NYC’s West Village, and Philly’s Rittenhouse Square. They just did sponsored posts and stories with our products in their neighborhoods. This simple move got us a 30% higher engagement rate than our main social ads during that time, proving that an authentic local voice is powerful.
- Dynamic Product Feed Integration: We connected our product catalog directly to the ad platforms. This meant we could run dynamic ads that only showed items that were in stock and relevant to that area’s seasonal demand, which stopped us from annoying people with out-of-stock products and improved our conversion rates.
Outcomes and Key Learnings
When all was said and done, the “Autumn Harvest” campaign pulled in a 3.5:1 ROAS which we were thrilled with. Our cost per conversion fell to just $8.70, well below our goal. We hit 15,000 total conversions, which was 20% more than we’d projected, off 6.3 million impressions. The campaign definitely got seen by the right people. The biggest lesson from this whole campaign is that specificity breeds success. You can’t just slap a zip code onto a national campaign and call it geo-targeting. To do it right, you need:
- Deep Audience Understanding: You have to know more than just where your people are. You need to know who they are and what their local buying habits look like.
- Dynamic and Relevant Creative: Ads have to feel local. When they can adapt to the local culture, events, and even the weather, they just perform better.
- Continuous Optimization: You can’t set it and forget it. We were in the accounts daily, ready to make fast, data-backed changes to bids, targeting, and creative when something wasn’t working.
- Integration with Local Presence: Connecting your digital ads to the real world, whether that’s a physical store, a local event, or an influencer, is what really makes the impact stick.
The “Autumn Harvest” campaign is solid proof that a well-planned and executed geo-targeting strategy works. It showed that if you understand and react to the specific character of a place, you can get amazing results, even when the market is tough. This kind of geographical precision is what marketing requires now. Generic, one-size-fits-all campaigns are just getting less and less efficient.
What is geo-targeting in a fall campaign?
In a fall campaign, geo-targeting means you’re aiming your ads at people based on their physical location. This could be as broad as a city or as narrow as a few blocks around a store. For our “Autumn Harvest” campaign, we focused on specific zip codes where we knew there were a lot of high-income shoppers with an interest in gourmet food.
How can dynamic ad creative enhance geo-targeting?
Dynamic creative makes your geo-targeting smarter by changing the ad content based on where the user is, what’s happening locally, or even what the weather is like. With “Autumn Harvest,” our ads would mention local events or show different products on a rainy day versus a sunny one, which made them feel much more personal and effective.
What metrics are most important to track for a geo-targeted campaign?
For any geo-targeted campaign, you have to watch your Return on Ad Spend (ROAS), Click-Through Rate (CTR), and Cost Per Conversion. Tracking impressions is also good for gauging reach. These numbers tell you if your strategy is working in a particular area and show you exactly where you need to optimize.
How did the “Autumn Harvest” campaign use first-party data?
We took our own first-party data, lists of past customers from purchase histories and email sign-ups, and used it to build lookalike audiences on social media. This let us find new people who behaved just like our best existing customers, which gave us much better conversion rates than just targeting general interests.
What is a practical tip for optimizing geo-targeting efforts?
A great practical tip is to feed real-time data into your ad serving. Think about foot traffic data for shopping centers or local event schedules. This lets you automatically adjust when and where your ads run to hit people at the perfect moment. We did this in the “Autumn Harvest” campaign by targeting specific areas during peak shopping hours, and it worked really well.
