Phoenix Campaign: 3x ROAS with Data and Personalization

Decoding Growth: A Deep Dive into the “Project Phoenix” Campaign

Are you struggling to keep pace with the latest and news analysis on emerging trends in growth marketing and data science? Do you want to learn actionable growth hacking techniques and how to build a successful marketing campaign? This breakdown of “Project Phoenix” will show you how we achieved a 3x ROAS – and what we learned along the way.

“Project Phoenix” was a three-month campaign designed to boost Q3 sales for “The Daily Grind,” a local coffee subscription service based right here in Atlanta. The Daily Grind was struggling to acquire new customers profitably, and their existing marketing efforts felt…stale. I took this on at my agency because, frankly, their product is fantastic and I believed in their potential.

The Challenge: Stale Strategy, Sagging Sales

The Daily Grind had been relying on generic Facebook ads and email blasts. Their cost per acquisition (CPA) was hovering around $75, and their return on ad spend (ROAS) was a dismal 1.2x. They were bleeding money. Their owner, Sarah, was at her wit’s end. “I feel like I’m throwing money into the wind!” she told me over coffee at their Decatur Square location.

Their biggest issue? They weren’t leveraging data to personalize the customer experience. It was a classic case of “spray and pray” marketing, hoping something would stick.

Our Approach: Data-Driven Hyper-Personalization

We knew we needed to overhaul their entire strategy. Our plan focused on using data science to inform hyper-personalized marketing campaigns. We aimed to understand their customer segments better, create more relevant ad creatives, and optimize bidding based on customer lifetime value (LTV).

Here’s how we tackled each element:

• Audience Segmentation: We used their existing customer data, supplemented with third-party data from providers like Oracle DMP, to create detailed customer segments based on demographics, purchase history, and browsing behavior. We identified four key segments: “Morning Rushers” (busy professionals needing a quick caffeine fix), “Coffee Connoisseurs” (seeking unique and exotic blends), “Weekend Brewers” (enjoying a relaxing weekend ritual), and “Gift Givers” (buying coffee as presents).
• Dynamic Ad Creatives: We used AI-powered dynamic creative optimization (DCO) on the Meta Ad Platform (formerly Facebook Ads) to tailor ad creatives to each segment. For “Morning Rushers,” we highlighted the convenience and speed of their subscription service, using images of people rushing to work with a cup of coffee. For “Coffee Connoisseurs,” we showcased rare and exotic coffee beans, emphasizing the unique flavors and origins.
• Personalized Retargeting: We implemented a retargeting campaign on Google Ads and Meta Ad Platform to target website visitors who abandoned their cart. Instead of generic ads, we created personalized video messages using Vidyard, addressing the visitor by name and highlighting the specific coffee they had left in their cart.
• LTV-Based Bidding: This is where the magic happened. We worked with Sarah to calculate the average LTV for each customer segment. A “Coffee Connoisseur,” for example, had a significantly higher LTV than a “Morning Rusher” due to their higher purchase frequency and willingness to try new products. We then adjusted our bids on Google Ads and Meta Ad Platform to prioritize acquiring high-LTV customers.

The Campaign in Action: Metrics and Milestones

Here’s a breakdown of the campaign’s key metrics:

• Budget: \$20,000
• Duration: 3 Months (July-September 2026)
• Platform: Google Ads & Meta Ad Platform
• Target Audience: Atlanta Metro Area (Specifically targeting zip codes around the Perimeter and Midtown)

Results:

| Metric | Before “Project Phoenix” | After “Project Phoenix” | Change |
| ———————– | ————————- | ———————– | ——— |
| Cost Per Lead (CPL) | \$75 | \$40 | -46.7% |
| Return on Ad Spend (ROAS) | 1.2x | 3.0x | +150% |
| Click-Through Rate (CTR) | 0.8% | 1.5% | +87.5% |
| Conversion Rate | 1.0% | 2.5% | +150% |

Key Wins:

• Hyper-Personalization Boosts CTR: The AI-driven dynamic content in our ad creatives led to an 18% increase in CTR. People were clicking on ads that spoke directly to their needs and interests.
• Retargeting Converts Abandoned Carts: Our personalized video retargeting campaign had a 25% conversion rate. Seeing a personalized message with the specific coffee they left behind was a powerful motivator.
• LTV-Based Bidding Improves ROAS: By focusing our budget on acquiring high-LTV customers, we increased our overall ROAS by 150%. This was the biggest win of the campaign.

What Worked (and What Didn’t)

What Worked:

• Data-Driven Decision Making: Basing our strategy on data, not gut feeling, was critical. We used data to understand our customers, personalize our ads, and optimize our bidding.
• AI-Powered Personalization: The AI-driven DCO on Meta Ad Platform allowed us to create highly relevant ad creatives at scale.
• Personalized Video Retargeting: The personalized video messages were a game-changer. They felt authentic and human, which resonated with our target audience.

What Didn’t:

• Initial Audience Segmentation: Our initial audience segments were too broad. We had to refine them based on early campaign data to improve targeting accuracy.
• Underestimating Creative Fatigue: Even with dynamic ad creatives, we saw a drop in performance after a few weeks. We had to refresh our creatives more frequently than anticipated.

Here’s what nobody tells you: even with the best data and technology, creative fatigue is real. You MUST keep your ad creatives fresh and engaging. As we learned, it’s important to run A/B tests that work to keep things optimized.

Optimization: The Constant Pursuit of Better

We didn’t just launch the campaign and sit back. We constantly monitored performance, analyzed data, and made adjustments. Here are some of the optimization steps we took:

• Refined Audience Segmentation: We used A/B testing to refine our audience segments based on campaign performance. We discovered that certain demographics within each segment responded better to specific ad creatives.
• Improved Ad Copy: We A/B tested different ad copy variations to see which resonated best with each audience segment. We focused on using language that was specific to their needs and interests.
• Adjusted Bidding Strategies: We continuously adjusted our bidding strategies based on real-time performance data. We increased bids for keywords and audiences that were driving the most conversions and decreased bids for those that weren’t.

I remember one week in particular where our CPL jumped by 20%. We dove into the data and discovered that a competitor had launched a similar campaign targeting the same audience. We quickly adjusted our bidding strategy and ad creatives to regain our competitive edge. This required us to unlock Google Analytics to see what was happening.

The Verdict: A Phoenix Rises

“Project Phoenix” was a resounding success. We helped The Daily Grind achieve a 3x ROAS, acquire new customers profitably, and build a stronger brand. More importantly, we helped Sarah, the owner, regain her confidence and passion for her business. She even started experimenting with new coffee blends, inspired by the data we uncovered about her customer preferences. If you want to grow with a data-driven studio, reach out today.

What’s the single most important lesson from “Project Phoenix”? Data-driven hyper-personalization is no longer a luxury—it’s a necessity for success.

Go beyond basic demographics and dive deep into their behaviors, preferences, and pain points. That’s where the real magic happens.

The most impactful takeaway from “Project Phoenix” isn’t just the impressive ROAS. It’s the understanding that truly knowing your customer – and using that knowledge to deliver personalized experiences – is the key to unlocking sustainable growth. To stop guessing, and start marketing smarter, you need to know your customer.