Marketing Analytics ROI: Sweet Stack’s Comeback

Marketing is no longer a guessing game. With the right marketing analytics, you can pinpoint what’s working, what’s not, and how to improve your ROI. Are you ready to stop wasting money on ineffective campaigns and start seeing real results?

I want to walk you through a real-world example of how we used marketing analytics to turn around a struggling campaign for a local Atlanta-based business. This isn’t just theory – it’s a deep dive into the trenches.

### The Client: Sweet Stack Creamery

Sweet Stack Creamery is a popular ice cream shop located in the heart of Little Five Points. They specialize in custom ice cream sandwiches and unique flavor combinations. They came to us in early 2026 because their summer marketing efforts were falling flat. Despite having a strong local following, their online campaigns weren’t driving enough foot traffic.

### The Challenge

Sweet Stack’s existing campaign was a broad, untargeted effort on Meta Ads Manager Meta Ads Manager. They were spending $5,000 per month with very little to show for it. Their cost per lead (CPL) was a staggering $45, and their return on ad spend (ROAS) was a dismal 0.8. They were essentially losing money on every dollar spent.

Here’s the thing: throwing money at a problem rarely solves it. You need data.

### Our Strategy: Data-Driven Optimization

Our approach was simple: dive deep into the data, identify the leaks in their marketing funnel, and plug them with targeted solutions. We broke down our strategy into three key phases:

1. Audience Analysis: Understanding who we were targeting and where they were spending their time online.
2. Creative Optimization: Crafting compelling ad creatives that resonated with our target audience.
3. Performance Tracking and Iteration: Monitoring key metrics and making data-driven adjustments to improve campaign performance.

### Phase 1: Audience Analysis

The first step was to understand Sweet Stack’s existing customer base. We used Google Analytics 4 to analyze website traffic, identify demographics, interests, and behavior patterns. We also looked at their social media followers to gain additional insights.

What we found was eye-opening. While Sweet Stack assumed their primary audience was young adults (18-25), the data revealed a much broader demographic. Parents with young children, tourists visiting the area, and even local business professionals were all regular customers.

We also discovered that a significant portion of their website traffic was coming from mobile devices, particularly from users searching for “ice cream near me” or “dessert in Little Five Points.”

Based on these findings, we created three distinct target audiences within Meta Ads Manager:

• Young Adults (18-25): Targeted with trendy, visually appealing ads showcasing Sweet Stack’s unique flavor combinations.
• Parents with Young Children (25-45): Targeted with family-friendly ads emphasizing Sweet Stack’s kid-friendly options and welcoming atmosphere.
• Local Professionals (25-55): Targeted with ads promoting Sweet Stack as a convenient and delicious after-work treat.

We also implemented location-based targeting, focusing on residents within a 5-mile radius of their shop on Euclid Avenue, and tourists visiting popular attractions like the Jimmy Carter Presidential Library and Museum.

### Phase 2: Creative Optimization

With a clear understanding of our target audiences, we moved on to creative optimization. Sweet Stack’s original ads were generic and uninspired, featuring low-quality images and bland copy.

We decided to A/B test different ad creatives to see what resonated best with each audience. We created three different ad variations for each target audience, each with a unique headline, image, and call-to-action.

For example, for the “Young Adults” audience, we created ads featuring vibrant, eye-catching images of Sweet Stack’s most popular ice cream sandwiches, along with headlines like “Level Up Your Instagram Feed” and “The Ultimate Atlanta Dessert Destination.”

For the “Parents with Young Children” audience, we created ads featuring heartwarming images of families enjoying ice cream together, along with headlines like “The Perfect Family Treat” and “Make Memories at Sweet Stack.”

We used Meta Ads Manager’s A/B testing feature to run these ads simultaneously, tracking key metrics like click-through rate (CTR), cost per click (CPC), and conversion rate.

### Phase 3: Performance Tracking and Iteration

The final phase was all about monitoring performance and making data-driven adjustments. We set up detailed tracking in Google Analytics 4 to measure website traffic, bounce rate, and conversion rates. We also tracked in-store sales to determine the true impact of our online campaigns.

Here’s where the rubber meets the road.

After the first week, we saw a significant improvement in campaign performance. The CTR for our ads increased by 12%, and our CPL dropped to $30. However, the ROAS was still below our target of 2.0.

We dug deeper into the data and discovered that while our ads were driving traffic to Sweet Stack’s website, many users were abandoning their carts before completing a purchase. We hypothesized that the website’s checkout process was too cumbersome and confusing.

To address this issue, we simplified the checkout process, reduced the number of required fields, and added a progress bar to guide users through the steps. We also implemented a retargeting campaign to re-engage users who had abandoned their carts, offering them a small discount to incentivize them to complete their purchase.

### The Results

After one month of optimization, the results were nothing short of remarkable.

• CPL: Decreased from $45 to $18 (a 60% reduction)
• ROAS: Increased from 0.8 to 2.5 (a 212.5% increase)
• Website Traffic: Increased by 45%
• In-Store Sales: Increased by 20%

Here’s a comparison table:

| Metric | Original Campaign | Optimized Campaign | Change |
| —————— | —————– | —————— | ———- |
| Budget | $5,000 | $5,000 | No Change |
| Duration | 1 Month | 1 Month | No Change |
| CPL | $45 | $18 | -60% |
| ROAS | 0.8 | 2.5 | +212.5% |
| CTR | 1.5% | 2.8% | +86.7% |
| Website Conversion | 1.0% | 1.8% | +80% |

Sweet Stack Creamery was thrilled with the results. They were finally seeing a return on their investment and were able to use the additional revenue to expand their menu and hire more staff.

### Lessons Learned

This case study highlights the power of data-driven marketing. By carefully analyzing our target audience, crafting compelling ad creatives, and continuously monitoring performance, we were able to turn around a struggling campaign and deliver exceptional results for our client. If you want to see similar results, understanding smarter performance marketing is key.

I’ve seen this pattern repeatedly: campaigns fail not because the product is bad, but because the targeting is off, the messaging is weak, or the tracking is nonexistent. Here’s what nobody tells you: marketing analytics isn’t just about numbers; it’s about understanding people.

The IAB (Interactive Advertising Bureau) publishes a wealth of industry reports. According to an IAB report, digital ad spending continues to climb, but the pressure to demonstrate ROI is also intensifying. You can’t afford to guess. You need to know.

Another anecdote: I had a client last year who was convinced that their target audience was on TikTok. They spent a fortune on TikTok ads, only to discover that their actual customers were primarily on Instagram and Facebook. A little bit of upfront research would have saved them a lot of money. To avoid this, you need a solid marketing attribution strategy.

### The Power of Multi-Touch Attribution

One more point that’s crucial: understanding attribution. Sweet Stack was initially using a first-click attribution model in Google Analytics. This meant that all the credit for a conversion was given to the first ad a user clicked on. This gave us a skewed view of which ads were truly driving sales.

We switched to a multi-touch attribution model, which distributes credit across all the touchpoints in a customer’s journey. This gave us a more accurate picture of which ads were most effective at driving conversions. Specifically, we implemented a data-driven attribution model in Google Analytics 4, which uses machine learning to determine the optimal attribution weights for each touchpoint. Another key point is that you need actionable insights from your analytics.

The Fulton County Department of Revenue needs businesses to thrive. If you’re not getting a return on your marketing investment, you’re not just hurting your business; you’re hurting the local economy.

Ultimately, the success of Sweet Stack’s campaign came down to a willingness to embrace data-driven decision-making. We didn’t rely on gut feelings or assumptions. We let the data guide our decisions, and the results spoke for themselves.

Don’t let your marketing budget go to waste. Start using marketing analytics to make smarter decisions and drive real results.