I've been running both YouTube and Connected TV campaigns for a few months now. I kept getting surprised by the cost differences, so I finally built a dashboard in Looker Studio to track CPV side-by-side.
My initial data shows Connected TV CPV is consistently 2-3x higher. Is that typical? I'm wondering if my CTV targeting is too broad or if this is just the normal premium. Also, how are others handling the attribution gap between the platforms in their reporting? My dashboard feels a bit simplistic because of that.
I'm Caroline, a senior data scientist at a mid-market DTC brand where we run about two dozen concurrent video experiments across YouTube, CTV, and social platforms, using a combination of Looker Studio for dashboards and a dedicated experimentation platform for causal analysis.
Your observed 2-3x CPV delta for Connected TV is directionally correct in my experience, but the drivers and reporting implications are critical. Here's a breakdown structured for a buyer comparing these channels.
1. **Channel Objective & Target Audience Fit:** YouTube is intent-driven (search, browsing) with a lower-funnel, skippable ad format. Connected TV is ambient and interruptive, targeting household-level demographics and interests for upper-funnel awareness. A typical CTV CPM in my environment is $35-50, while YouTube TrueView is $8-15. The CPV difference stems from this premium for guaranteed, large-screen attention in a lean-back environment.
2. **Real Cost Structure & Hidden Costs:** Beyond the direct media CPV, you must factor in data and measurement costs. CTV often requires integrating a third-party attribution partner (like Measured or Innovid) to bridge the clickless gap to downstream conversions, which can add $2-5 CPM. YouTube's attribution, while imperfect, is native and directly pipes into Google Analytics, making incremental measurement costs near zero.
3. **Deployment & Integration Effort:** Building the dashboard is the easy part. The core integration effort is creating a unified user journey key. For YouTube, you rely on Google's pseudonymous IDs. For CTV, you're typically matching IP-based household graphs from your attribution partner. Merging these into a single identity spine in your CDP or data warehouse is a multi-week engineering project to ensure you're not double-counting reach.
4. **The Honest Attribution Limitation:** Your dashboard's simplicity is unavoidable without a randomized controlled test (RCT) or geo-lift measurement. Both platforms use last-click/views-based models that ignore cross-channel influence. In our last incrementality study, we found YouTube's reported conversions were inflated by ~40% and CTV's by over 60% when compared to a geo-split test. A dashboard can only show platform-reported performance, not true causal impact.
I would recommend continuing to run both, but with a strict test-and-learn budget allocation. For pure performance marketing with a direct conversion goal, shift budget to YouTube. For launching a new product or driving branded search lift, accept the CTV premium and immediately implement a geo-based measurement plan to track its real contribution. To give a cleaner recommendation, tell us your primary campaign goal (awareness vs. conversion) and whether you have the resource budget to implement geo-lift or matched-market tests.
Nullius in verba