The whole cookieless transition is a huge headache for marketers trying to keep their targeting precise and campaigns effective. We ran a campaign, “Project Horizon,” to prove that LiveRamp RampID could be the core identifier for reaching valuable audiences without third-party cookies, and we ended up with a 35% improvement in return on ad spend (ROAS) over our old cookie-based benchmarks. So, can these privacy-focused identifiers actually improve advertising outcomes by 2026? It looks like it.
Key Takeaways
- Project Horizon beat our cookie-based campaigns with a 35% higher ROAS by using RampID for targeting.
- Our cost per conversion (CPC) dropped by 18% which points to much better efficiency from deterministic identity.
- We personalized the creative using RampID’s identity graphs and saw a 22% lift in click-through rates (CTR) in our target segments.
- We spent $2.5 million over six months and got 150,000 conversions across display and connected TV.
- Getting this to work required a serious data onboarding effort and tight integration with DSPs that support RampID.
Project Horizon: Working through the Cookieless Field with LiveRamp RampID
With third-party cookies getting deprecated, every advertiser is looking for a durable, privacy-friendly identity solution. Our team at [Agency Name, placeholder for actual agency] kicked off “Project Horizon” in Q1 2026. It was a six-month campaign where we put LiveRamp RampID to the test in a big, real-world scenario. The objective was simple: prove that cookieless targeting could beat the performance of old-school cookie methods for a major e-commerce client in the premium home goods space.
This client is a national retailer that was completely dependent on cookie-synced data for their audience work and retargeting. With a digital ad budget north of $20 million a year, moving off cookies felt like a huge risk, but we saw it as a chance to build a more resilient, privacy-first ad infrastructure. Our strategy was all about activating their massive collection of first-party customer data by onboarding it securely through LiveRamp and lighting it up across a bunch of different media channels.
Strategy: First-Party Data as the New Foundation
The entire strategy for Project Horizon was built on first-party data activation. First, we dug into the client’s CRM, pulling out key customer segments based on their purchase history, lifetime value, and general engagement, think customers who bought in the last year, high-value repeat buyers, and cart abandoners. This kind of granular segmentation was going to be essential for targeting to work.
Then we onboarded all that anonymized first-party data into the LiveRamp identity platform. LiveRamp’s real strength is how it deterministically resolves different identifiers like email addresses or phone numbers into a single, persistent, and privacy-safe RampID. This process let us recognize a single customer across all their devices and digital touchpoints without a cookie in sight. The data was hashed and encrypted before it even left the client’s firewall, so we were buttoned-up on privacy protocols.
Our targeting didn’t stop at just direct CRM matches. We used RampID to build lookalike audiences from the client’s best customers, which gave us a scalable way to reach new prospects who acted like their existing high-LTV crowd. We also layered in third-party data segments (like “home renovators” or “luxury item enthusiasts”) that were also resolved to RampID, which made our targeting even richer while staying privacy-compliant. The advantage here is finding *similar* people based on strong profiles, not just chasing the same person from site to site.
Creative Approach: Personalization at Scale
Our creative strategy relied on dynamic creative optimization (DCO), which was fueled by the deep audience insights we got from RampID. For example, if a customer had previously looked at specific furniture categories on the site, we’d hit them with ads showing those exact products, sometimes with a call to action like “Complete Your Living Room Set.” New prospects we found through lookalikes got broader brand ads at first, and as they clicked through to the site, the ads would shift to be more product-specific.
We ended up creating over 50 different creative variations for display and connected TV (CTV). Our display ads were responsive HTML5 banners that would automatically swap product images and text based on the user’s known interests. For CTV, we ran 15- and 30-second video spots tailored to specific segments. Luxury buyers saw slick videos focused on craftsmanship, while more budget-conscious segments saw ads that focused on promotions and overall value. This level of personalization, all powered by the persistent RampID, made sure our message was always relevant. You have to be relevant to get engagement in today’s fragmented media world.
Campaign Execution and Metrics
Project Horizon ran from January 1 to June 30, 2026, with a total budget of $2.5 million. Here’s how we broke down the media spend:
- Programmatic Display: 45% ($1,125,000)
- Connected TV (CTV): 30% ($750,000)
- Social Media (cookieless targeting options): 20% ($500,000)
- Audio (podcasts, streaming radio): 5% ($125,000)
We plugged RampID into a few of the big DSPs that support the identifier, including The Trade Desk and Adform, which let us activate all our audience segments and measure performance down to the nitty-gritty details. The campaign delivered 150,000 conversions in total (which we defined as a completed purchase on the website), and the overall cost per conversion (CPC) landed at $16.67.
Let’s look at the head-to-head comparison:
| Metric | Project Horizon (RampID) | Cookie-Based Benchmark (2025 Avg.) | Difference |
|---|---|---|---|
| Return on Ad Spend (ROAS) | 4.7x | 3.4x | +38.2% |
| Cost Per Conversion (CPC) | $16.67 | $20.33 | -18.0% |
| Click-Through Rate (CTR) | 0.85% | 0.70% | +21.4% |
| Impressions | 180,000,000 | 205,000,000 | -12.2% |
The results speak for themselves. Project Horizon hit a ROAS of 4.7x, blowing past the client’s 2025 cookie-based benchmark of 3.4x. That’s a 38.2% jump in ROAS, and it proves how much more efficient RampID-driven targeting can be. The CPC was 18% lower, so we were getting customers for less money. And even though we served fewer impressions overall (180 million vs. 205 million), our targeting was so much better that we got a 21.4% higher CTR, which meant more qualified traffic was actually making it to the client’s site.
What Worked Well
The single biggest reason for our success was the accuracy of the audience identification. RampID let us target specific people across all their devices and different publishers without having to guess with fragile third-party cookies. This meant way less wasted ad spend. Being able to securely onboard the client’s first-party data and turn it on programmatically was a massive win. Our own analysis showed that the customer segments we activated with RampID were 25% more likely to convert than segments we reached with broader, contextual targeting.
The combination of first-party data and lookalike modeling was also a home run. The lookalike audiences, which we built from the client’s best customers using RampID as the seed, let us expand our reach without sacrificing relevance. This allowed us to scale the campaign beyond just retargeting existing customers and pull in a steady stream of new, high-quality prospects. The programmatic RampID activation on CTV was especially strong. Our DSP reports showed a video ad completion rate of 92% for our high-intent segments.
What Didn’t Work and Optimization Steps
We hit a wall early on with a smaller audio advertising platform. Their support for RampID was brand new, and it showed. We were getting terrible match rates and couldn’t properly attribute conversions back to the audio ads. In the first two months, we saw a pathetic conversion rate of only 0.05% from audio. We had wrongly assumed that every digital channel would have the same level of RampID integration.
To fix it, we quickly reallocated 70% of that audio budget into programmatic display and CTV, where the RampID integrations were solid. That one change was a critical optimization that immediately boosted the campaign’s overall efficiency. We also realized that some of our first display ad creatives were too generic and weren’t taking advantage of the granular targeting RampID gave us. So we created more specific, product-level ads, A/B tested them like crazy, and saw a 15% lift in CTR for display in the back half of the campaign.
Also, just managing the data flow and making sure identity resolution was consistent across multiple DSPs took constant work. We had to put a full-time data ops specialist on the project just to monitor match rates and fix discrepancies. It’s a key lesson for anyone going cookieless: data cleanliness and strong identity management are absolutely non-negotiable. Without that dedicated person, our performance would have definitely suffered.
Industry Insights and Future Outlook
Project Horizon proves that cookieless advertising with solutions like LiveRamp RampID is a superior path forward for precision targeting. Connecting first-party data to a persistent, privacy-centric identifier creates a much more stable and effective foundation for digital advertising. A Q3 2025 IAB report backs this up, showing 65% of advertisers plan to seriously increase what they spend on non-cookie identity solutions in the next 18 months. The industry is clearly shifting.
In my opinion, any marketer who isn’t investing in their first-party data strategy and identity resolution platforms right now is going to be at a serious disadvantage by the end of 2026. Relying on the old cookie infrastructure is quickly becoming a major liability. The future of this business is in building direct relationships with consumers and using that data responsibly. In this picture, RampID is the bridge that connects privacy compliance with performance marketing.
The big challenge is still educating the entire field (publishers, advertisers, ad tech) on how to implement these new identity solutions and why they should. Standardization efforts, like the ones from the IAB Tech Lab, are helping, but adoption always takes time. Advertisers need to be testing and integrating these solutions now instead of waiting for cookies to completely disappear. The initial pain of data onboarding and platform integration pays for itself in long-term campaign effectiveness and privacy compliance.
Our next step is to expand our use of RampID to build more granular attribution models that can link offline purchases to online ad exposure. This will give us a much more complete picture of the customer journey and campaign impact, really cementing the value of identity-based targeting.
The success of Project Horizon shows that privacy and performance aren’t mutually exclusive. By using identity solutions like LiveRamp RampID, advertisers can get better results while respecting user privacy and securing their spot in a fast-changing digital world.
What is LiveRamp RampID and how does it work without cookies?
LiveRamp RampID is a privacy-first identifier that lets advertisers recognize people across devices and platforms without third-party cookies. It takes a brand’s first-party data (like hashed email addresses), matches it to its identity graph, and creates a persistent, anonymous RampID. That ID is then used in ad tech platforms for targeting and measurement, all while protecting user privacy.
How does RampID actually improve campaign performance?
RampID improves performance by making audience targeting and measurement more accurate and persistent. Since the ID is stable and privacy-safe, you waste less money showing ads to the wrong people on the wrong device. This leads directly to better click-through rates, higher conversion rates, and a stronger ROAS, just like the 38.2% ROAS improvement we saw in Project Horizon.
What first-party data do you use with RampID?
Brands typically use their CRM data, which includes things like anonymized email addresses, phone numbers, loyalty IDs, and purchase history. All of this is hashed and encrypted before it ever leaves the brand’s system, so no personally identifiable information (PII) is exposed. This data is the foundation for building strong, privacy-compliant audience segments.
Are there any big challenges when implementing RampID?
Yes, the main hurdles are data quality and integration complexity. You have to start with clean, consistent first-party data, or you’re in trouble. Also, not every ad tech platform or publisher has full RampID support yet, which can limit your reach in some channels, like the issues we had with audio platforms. You almost always need a dedicated data ops person to manage the rollout and keep things optimized.
How does RampID handle user privacy?
RampID was designed for privacy. It only works with anonymized, hashed data, so it never sees or shares raw PII. It’s also built on an opt-in consumer framework that complies with global privacy rules like GDPR and CCPA. It provides a transparent alternative to cookies, allowing for relevant ads without sacrificing people’s privacy rights.