The gap between CTV's share of consumer attention and its share of performance budgets has persisted not because the channel lacks scale, but because the measurement infrastructure to account for it has lagged.
At MAU Vegas in May 2026, Sameer Sondhi, CEO North America at Affle, moderated a session that addressed that gap from three angles: advertiser experience, attribution methodology, and supply-side accountability. The panel brought together Abby Patton, Associate Director of Performance Marketing at Fetch; Sean Galligan, CRO at Kochava; and Kunal Nagpal, Chief Business Officer at InMobi.
The full session is available below, with key points from each segment highlighted.
From Awareness Budget to 18–20% of Total Media Spend: Fetch's CTV Journey
Abby Patton opened her segment (03:44) with a trajectory that most performance teams have not yet replicated. Fetch entered CTV in late 2024 as a complement to linear TV, expecting awareness-level results. By Q4 2024, the channel was delivering users at efficiency parity with established performance channels, with comparable retention outcomes. Today it represents 18 to 20 percent of Fetch's total media budget.
The more consequential evidence came from an unplanned test (04:49). An attribution bug forced Fetch to pause all CTV activity for several weeks, producing an involuntary incrementality test. CAC softened across non-CTV campaigns during the pause. Resurrection rates dropped.
Both metrics recovered when CTV came back online, confirming that the channel had been supporting re-engagement across the full funnel, not only driving installs at the top.
Measuring CTV Against the Full Range of Downstream Conversion Events
Building on Fetch's experience, Sean Galligan (07:43) addressed the measurement gap that prevents most teams from reading CTV accurately. CTV generates downstream events well beyond installs: purchases, subscriptions, store visits, and website sessions. Teams measuring CTV exclusively against install volume are capturing a fraction of what the channel moves.
The deeper issue, per Galligan (09:34), is single-methodology measurement. Last-touch attribution, MMM, and incrementality testing each answer different questions; no single approach answers all of them simultaneously. Running all three together is what Galligan describes as a measurement operating system, and it represents the minimum standard for attributing CTV performance with any degree of accuracy.
Cross-Screen Attribution: Why CTV Contribution Disappears from Most Dashboards
Galligan's explanation of how Kochava approaches cross-screen attribution (11:12) clarifies a structural problem in how most teams measure the channel. The consumer journey initiates on the large screen and completes on mobile, often days later and outside the original household.
Connecting that impression to that install requires device graph resolution and household-level matching. Without it, CTV's contribution is credited to whatever channel closed the path, which is typically search or social.
Kochava's multi-touch attribution data quantifies this directly:
In app install journeys where both CTV and search appear, CTV precedes search 96% of the time. Where CTV and social appear together, CTV leads 94% of the time. CTV is the origin point of user intent in the majority of cross-channel journeys; default attribution configurations assign it the role of assist.
Fixing that gap requires more than a methodology change. The YouAppi, Jampp & Kochava 2026 CTV Growth Guide covers the specific attribution configuration errors that cause CTV's contribution to go unaccounted for, along with the cross-screen measurement frameworks to address them.
Supply-Side Signals: What Performance-Ready CTV Inventory Actually Requires
Kunal Nagpal (13:41) addressed why TV's performance potential was consistently recognized but rarely realized in practice. The scale and attention quality of the screen were never in question. What was missing was inventory that could support accurate measurement. Content owners operated under a closed-ecosystem model for years, a structure that made proper attribution architecturally impossible.
The shift of the past two years reflects a change in that model: content owners and OEMs have begun exposing the signals that performance measurement requires, including behavioral data, device identifiers, and the postback structures that allow DSPs and MMPs to attribute cross-screen conversions properly.
Nagpal's working definition of performance-ready inventory encompasses three requirements: the inventory passes signals for measurement, it supports proper attribution, and it feeds data back into campaign optimization. Inventory that satisfies only two of those three conditions does not produce incomplete results that are visibly incomplete; it produces an incomplete picture that reads as clean.
"CTV alone is not going to solve all your issues. Despite the many reasons to be attracted to it as a channel, it can't be the only channel."
— Kunal Nagpal, Chief Business Officer, InMobi
What the Panel Establishes About CTV's Role in Performance Budgets
The through-line across all three segments is structural: the gap between CTV's share of consumer attention and its share of performance budgets is not a channel problem. It is a measurement and infrastructure problem, and both are now solvable.
First-party campaign data confirms the performance case
Fetch's trajectory from zero to 18–20% of total media budget in under a year, at efficiency and retention parity with established channels, represents the kind of evidence that moves budget decisions. The forced incrementality test strengthened that case further: the effects of CTV extended into re-engagement across the full funnel, visible only when the channel was removed.
Attribution methodology determines what CTV gets credit for
The measurement operating system Galligan describes, combining last-touch attribution, MMM, and incrementality testing, is the framework that makes CTV's contribution legible. Reach and frequency metrics were designed for a different medium. Applied to CTV, they measure the wrong things and systematically undercount outcomes that the channel is generating.
Supply quality is a prerequisite, not a variable
Nagpal's three-condition definition of performance-ready inventory establishes the supply standard that makes accurate measurement possible in the first place. CTV's performance case depends on inventory that passes signals, supports attribution, and feeds optimization loops. Two out of three is not a partial solution; it is a measurement gap that looks like a clean result.
Three things the panel made clear:
- CTV is no longer a test channel. Fetch's trajectory from zero to 18–20% of total media budget, confirmed by an unplanned incrementality test, establishes a performance case built on campaign data, not projections.
- Measurement methodology determines what CTV gets credit for. Last-touch alone undercounts the channel. The combination of last-touch attribution, MMM, and incrementality testing is the minimum standard for reading CTV performance accurately.
- CTV amplifies mobile retargeting; it does not replace it. The outcomes are strongest when both channels operate within the same cross-screen system, with shared audience segmentation, attribution, and optimization.
For the attribution configuration and budget sequencing frameworks that operationalize this approach, the YouAppi, Jammp & Kochava 2026 CTV Growth Guide covers the three measurement errors most likely to cause CTV's contribution to disappear from your dashboards. Download it here to understand more on the topic.