Loyalty Data Drives 2.5x ROAS in 2026 Video Ads

The idea of using granular loyalty data to personalize video ads is that you move past broad demographics and speak to individual customer preferences. This should lead to deeper engagement and more efficient ad spend, but we wanted to see if the numbers actually backed that up.

Case Study: Enhancing Customer Lifetime Value with Personalized Video

We just wrapped a video advertising campaign for a major direct-to-consumer (DTC) apparel brand. Their goal was to reactivate dormant customers and get their active loyalty program members to increase their average order value (AOV). Our hypothesis was that personalized video creative, built from their rich customer relationship management (CRM) data, would easily outperform standard video campaigns aimed at similar demographic groups.

Campaign Strategy and Objectives

Our plan was direct: use the company’s existing loyalty program data to create highly relevant video ads that would hit home with individual customers, in the end driving repeat buys and making them feel better about the brand. We defined our targets very specifically: a 2x increase in ROAS for the loyalty segments, a 15% uplift in CTR compared to our non-personalized control ads, and a 10% reduction in cost per conversion. We ran the campaign for three months (September to November 2026) to line up with important shopping seasons. The total budget for the project was $250,000, split between Google Ads (we used YouTube TrueView for Action) and Meta Ads across Instagram and Facebook.

Data Integration and Segmentation

The entire campaign was built on the brand’s loyalty program data, which lived in their CRM and was much more than just a list of emails. We had purchase history (product categories, average spend, frequency), browsing behavior from their site (viewed products, abandoned carts), engagement with past marketing (email opens and clicks), and even preferences they shared when they signed up. We used this to slice the loyalty database into a few key groups:

• High-Value, At-Risk: Customers with 3+ purchases in the last 12 months but zero activity in the last 60 days.
• Mid-Tier, Dormant: Customers with 1-2 purchases in the last 18 months and no activity for 90 days.
• New Loyalty Members: People who joined in the last 30 days but hadn’t bought anything yet.
• Category Enthusiasts: Customers who consistently bought from specific product lines (like activewear or casual wear).

Having these granular segments meant we could create extremely specific messaging and product suggestions. For example, a “High-Value, At-Risk” customer who always bought activewear would see a completely different video than a “New Loyalty Member” who’d only browsed the casual wear section.

Creative Strategy and Dynamic Personalization

Our creative was designed from the start to take advantage of the loyalty data. Instead of producing hundreds of unique videos (which would be a nightmare), we created modular video templates with dynamic elements that we could swap out based on the target segment. These included:

• Product Overlays: Showing specific items or collections based on a customer’s past purchases or what they’d looked at on the site.
• Call-to-Action (CTA) Language: Tailoring the offers, like “15% off your next activewear purchase” for the enthusiasts or “Free shipping on orders over $50” for new members.
• Background Music and Voiceover Tone: Making small adjustments to match the feel of the different demographic profiles in our segments.
• Urgency Messaging: Hitting dormant segments with a “Limited-time offer” while showing active customers a more relaxed “Explore our new arrivals” message.

We used a creative automation platform, Adobe Creative Cloud, connected to a dynamic ad platform to manage all these combinations. This approach let us produce personalized videos at scale without the creative costs spiraling out of control. Every segment received at least three different video versions so we’d have plenty to A/B test. If we hadn’t done it this way, the cost of making individualized videos would have killed the strategy before it even started, especially with this budget.

Targeting and Campaign Execution

The execution was straightforward. We uploaded our segmented CRM lists into Google Ads and Meta Ads as custom audiences, which let us target these exact groups with our personalized video creative. On the Google side, we focused on YouTube placements, using “TrueView for Action” to drive direct sales. For Meta, we ran a mix of in-stream video and story placements across Instagram and Facebook. To have a solid baseline for comparison, we also ran a control group. This group was a lookalike audience built from the brand’s active customer base, and they saw only generic video ads. This was the only way to know for sure if our personalization efforts were actually paying off.

What Worked: Unpacking the Performance Data

The numbers we got back from the loyalty-driven segments blew our initial projections away.

The personalized video ads generated a 3.8x ROAS from the loyalty segments, which absolutely crushed the 1.5x from the control group. That 153% improvement shows exactly what happens when you deliver a relevant message. The CTR hit 2.85%, confirming that the tailored creative was getting people to actually click. Our cost per conversion for the loyalty segments was $17.36 compared to $30.00 for the control group, and this efficiency is where we saw the real budget savings. The dynamic product recommendations were a huge win. For example, the “High-Value, At-Risk” segment, which saw ads showing new arrivals in their preferred categories along with a limited-time discount, gave us a 4.1% CTR and an AOV of $110. By showing them exactly what they were interested in, we kept them from tuning out and nudged them to buy something immediately.

What Didn’t Work and Optimization Steps

Of course, not everything worked perfectly on the first try. Our “Mid-Tier, Dormant” segment was a problem at the start, with a low CTR of about 1.1% and a CPL hitting $45. We quickly realized our first creative for this group was too soft, focusing on brand messaging instead of giving them a clear reason to come back. Optimization Steps:

• Refined Creative for Dormant Segment: We A/B tested new videos for this group that dropped the brand fluff and went straight for a strong, time-sensitive offer (like, “Welcome back! Enjoy 20% off your entire order for the next 48 hours”). We also mixed in some social proof by showing user-generated content from other loyal customers.
• Adjusted Bid Strategy: The “Maximize Conversions” bid strategy we started with was too aggressive for the low-engagement “Mid-Tier, Dormant” segment. We switched to “Target CPA” and set a slightly higher target, giving the algorithm more room to find conversions that were there.
• Cross-Channel Frequency Capping: We saw that some users were getting hit with ads on both Google and Meta. We put a combined frequency cap in place (no more than 7 impressions per user per week across both platforms) to stop annoying people, a lesson backed up by Nielsen’s 2023 report on ad frequency.
• Refreshed Creative: After about six weeks, we rolled out new creative for our best-performing segments (“High-Value, At-Risk” and “Category Enthusiasts”). Even winning ads get old. Swapping in new visuals and copy kept engagement up and prevented ad blindness.

These adjustments, especially the creative fix for the dormant segment, boosted their CTR by 35% and cut their CPL by 28% within two weeks. I’ve learned that assuming your initial creative will be a hit across all segments is a rookie mistake. You have to be watching the data and be ready to adapt all the time.

Editorial Insight: The Overlooked Power of Negative Data

You learn as much from what bombs as what succeeds. For our “New Loyalty Members” segment, we tested a video with a celebrity endorsement, thinking it would build instant trust. It tanked. The CTR was a pathetic 0.9% and the CPL was sky-high. This showed us that for this group, practical value (a direct discount on their first purchase) was far more important than an aspirational message. Don’t just chase every marketing trend without checking if it actually fits your audience. The data, especially the bad data, will tell you what they really want.

Budget Breakdown and Final Metrics

Here’s how the $250,000 total budget was spent:

• Google Ads (YouTube): $110,000
• Meta Ads (Facebook/Instagram): $90,000
• Creative Production & Dynamic Personalization Tools: $35,000
• Analytics & Reporting Tools: $15,000

After the three-month campaign and all our optimizations, here are the final aggregate metrics for the loyalty-driven segments:

• Total Impressions: 15,800,000
• Overall CTR: 2.68%
• Total Conversions: 9,150
• Average CPL: $20.50
• Overall ROAS: 3.2x

The final ROAS of 3.2x dipped a bit from the 3.8x peak of our best segments, but it still blew past our 2x target and showed a fantastic return on the investment in these loyalty program members. The budget for creative and tools was clearly money well spent.

Conclusion

The lesson here is clear: plugging loyalty program data directly into a video ad strategy works. It makes your campaigns more efficient and gets customers to actually engage. Once you understand what your customers have bought and browsed, you can stop shouting generic brand messages and instead deliver relevant video content that leads to better conversion rates and a much healthier return on ad spend.