Urban Roots: 4.2x ROAS With 2026 Identity Data

In the 2026 digital ad world, consumer attention is fragmented to an almost impossible degree, so effective identity resolution is the bare minimum for delivering personalized content that people actually notice. Our recent work with “Urban Roots Garden Supplies,” an e-commerce retailer in the organic gardening space, is a case in point. We hit a 4.2x return on ad spend (ROAS) by getting their user journey mapping right. This is how sophisticated data orchestration can transform PPC targeting.

Campaign Overview: Urban Roots Garden Supplies

Urban Roots Garden Supplies had a problem we see all the time: their online presence was growing, but conversion rates were flat. Their ad strategy was stuck in the past, depending on third-party cookies and broad demographic buckets, which led to a lot of wasted ad spend and a generic experience for users. Our job was to rip out their old PPC targeting, plug in an identity resolution platform, and start delivering personalized ads that would finally move the needle on ROAS and customer acquisition cost (CAC).

Initial Metrics & Budget

• Campaign Duration: 6 months (January 2026 – June 2026)
• Total Ad Budget: $150,000
• Baseline ROAS: 2.1x
• Baseline CPL (Cost Per Lead): $18.50
• Baseline Conversion Rate: 1.8%
• Baseline CTR: 1.2%

Strategy: The Identity-First Approach

Our entire strategy was built around a central identity graph. We pulled all of Urban Roots’ scattered data sources, CRM records, website analytics, email lists, purchase history, into one place to build unified customer profiles. This meant stitching together fragmented signals from different devices and channels with both deterministic and probabilistic matching. For instance, we could finally see that the user browsing “organic soil amendments” on their laptop was the same person who later used their phone to sign up for the newsletter. This insight allowed for actual personalization, way beyond basic retargeting.

We were very specific in choosing a platform with strong server-side tagging. This move was to future-proof the entire campaign against the death of third-party cookies and all the new browser privacy controls. Collecting data from our own server instead of relying on the client’s browser gave us much more accurate and resilient data streams for identity matching, which was the whole point.

Creative Approach: From Generic to Granular

Our creative strategy had to shift completely. We stopped running generic “gardening supplies” ads and instead built a modular creative system. This let us pull together ad copy and images on the fly, based on what we knew about the user from their resolved identity. For example:

• Segment 1: “New Gardeners” (people who read beginner blog posts but hadn’t bought anything): They saw ads for starter kits, easy-to-grow plants, and content about basic gardening.
• Segment 2: “Experienced Organic Growers” (identified by past purchases of organic fertilizers or browsing advanced tools): We showed them ads for new specialized nutrients, eco-friendly pest control, and high-end tools.
• Segment 3: “Urban Balcony Enthusiasts” (people browsing container gardening and small-space products): Their ads featured compact planters, vertical gardening kits, and plants that don’t need much water.

This specific approach carried over to ad formats, too. We used dynamic product ads (DPAs) to show users the exact items they’d viewed but didn’t buy, sometimes with a small discount to nudge them over the line. For prospects who were earlier in their journey, we ran video ads with customer testimonials that were relevant to their specific interests.

Targeting: Precision at Scale

The identity resolution platform was the brain of the operation. We pushed our resolved audience segments straight into Google Ads Custom Audiences and Meta’s Advanced Matching. This let us do a few things really well:

1. Exclude Irrelevant Audiences: We stopped wasting money on existing loyal customers who just made a purchase by temporarily excluding them from acquisition campaigns. Simple, but effective.
2. Prioritize High-Value Segments: We identified users with high lifetime value potential (based on patterns in the data) and gave them higher bid adjustments to make sure they saw our ads.
3. Cross-Device Retargeting: This was a huge win. If someone abandoned a cart on their laptop, we could hit them with a reminder ad on their phone, complete with a personalized discount code. The old system could never connect those dots.

Here’s a perfect example: we found a group of users who bought a lot of heirloom seeds but almost never bought tools. So we started serving them ads for high-quality hand tools, positioning them as the perfect companion for their hobby. This kind of micro-segmentation, all powered by a single view of the customer, was a world away from their previous, generic “gardening enthusiast” targeting.

What Worked and What Didn’t

Successes:

• Personalized Creative Impact: Serving highly relevant ads was the biggest win, period. Our average CTR jumped to 2.8%, a 133% lift from the baseline, and some ad groups targeting the “Experienced Organic Growers” hit CTRs of 4.1%.
• Reduced Wasted Spend: Because we could accurately identify and exclude people who weren’t interested or had just bought something, we saw a huge drop in impressions that went nowhere. This directly lowered our CPL.
• Improved Conversion Rates: The conversion rate climbed to 3.2% by the end of the campaign (an 80% improvement). This was driven by better ads, but also by dynamically changing the landing page. A “New Gardener” clicking an ad for a starter kit would land on a page that was all about ease of use, not the full, overwhelming product catalog.
• Cross-Device Consistency: The platform bridged the gap between devices beautifully. This meant we weren’t sending repetitive or annoying messages which led to fewer unsubscribes and better engagement with the email campaigns triggered by ad clicks.

Challenges:

• Initial Data Onboarding Complexity: Getting Urban Roots’ old CRM and website analytics to talk to the new identity platform was a real pain. We spent most of the first month just on data cleaning and mapping, and we definitely underestimated how many inconsistent data entries we’d have to fix.
• Platform Learning Curve: The new platform was powerful, but it wasn’t intuitive. It took a lot of training to get the marketing team comfortable with the advanced segmentation logic and audience activation workflows.
• Maintaining Data Hygiene: We had to be vigilant about monitoring data inputs. Every time a new data source was added, like a new email form, it had to be integrated and checked right away to stop new data silos from forming.

Optimization Steps Taken

Over the six months, we were constantly tweaking things. Here are the main optimizations we made:

1. Iterative Segment Refinement: We started with broad segments and then continuously broke them down based on performance. For instance, our initial “New Gardeners” segment got split into “Container Garden Beginners” and “Raised Bed Beginners” when we noticed they had different product preferences and that more specific content got higher engagement. Weekly data dives on product views and conversion paths drove this process.
2. Bid Strategy Adjustments: The identity graph gave us predictive scores on how likely different user profiles were to convert, so we used that information to adjust bids dynamically. High-intent users got higher bids, while we used cheaper impression-based campaigns for lower-intent folks. That level of control was impossible before.
3. Negative Keyword Expansion: We got aggressive with our negative keyword lists, especially for acquisition, to stop showing ads to people just looking for information. We did this by cross-referencing search query reports with our identity data to find patterns where certain searches got tons of impressions but zero conversions for a specific segment.
4. Creative Refresh Cycles: To keep ads from getting stale, we set up a bi-weekly creative refresh for our high-volume ad groups. The identity platform made this easy, we could quickly spin up new ad variations for our existing segments by just swapping out product photos or testimonial text.

Results

The campaign blew past its initial goals, and the precision we got from identity resolution was the main reason.

Final Metrics:

• Final ROAS: 4.2x (a 100% increase from baseline)
• Final CPL: $11.80 (a 36% reduction from baseline)
• Final Conversion Rate: 3.2% (an 80% increase from baseline)
• Final CTR: 2.8% (a 133% increase from baseline)
• Total Impressions: 12.5 million
• Total Conversions: 3,840
• Cost Per Conversion: $39.06

This whole project showed that investing in an identity resolution platform isn’t about collecting data for its own sake. It’s about generating real, actionable insights that let you create better personalized content and spend ad money more efficiently. Yes, the initial data integration can be a headache, but the long-term payoff in customer engagement and profit is huge. You’re building a coherent story for each customer, no matter where they interact with your brand.

Conclusion

For any marketer trying to get noticed and connect with consumers, identity resolution is a must-have. It’s the data plumbing you need to run a truly personalized content strategy, which leads to better campaign performance and stronger customer relationships. You need to build a solid first-party data strategy and connect it to a capable identity platform to get the most out of your PPC targeting.