Earlier this year, I wrote in AdExchanger about the 20% to 30% of programmatic video spend that gets lost in the gap between what inventory labels promise and what the actual viewing experience delivers.
My core argument was simple. Video buying systems have become exceptionally good at scaling impressions, but far less effective at distinguishing between fundamentally different viewing experiences because they still rely heavily on labels and metadata rather than the execution-level signals that reveal how video actually behaves in-market.
The release of CIMM’s new “Quality Matters” white paper suggests that this conversation is moving into the mainstream.
The paper represents the first serious attempt to give buyers, publishers, platforms and measurement companies a shared language for something the market has been grappling with for years. Not all video impressions create the same opportunity for impact. More importantly, it provides a framework for treating media quality as a measurable and actionable input into buying, pricing and optimization decisions rather than something evaluated after the fact.
To better understand what that means specifically for video advertising, we spoke with one of the paper’s co-authors, Erez Levin, founder of Emet Advisory and a longtime architect of quality-based frameworks and solutions at Google, about the principles he believes matter most.
Three principles that should reshape how video is bought
EREZ LEVIN: If I had to boil the “Quality Matters” paper down to one line, it would be: Stop treating fundamentally different advertising opportunities as if they have the same value.
The industry has relied on averages, proxies and broad classifications to buy and sell media for years now. That’s helped scale, but it’s created a market where very different experiences are often grouped together, priced similarly and optimized toward the same outcomes.
We don’t intend to create one universal quality score. We’re presenting an industry framework for recognizing and valuing those differences more consistently.
Three ideas matter most.
- Stop treating channels as monoliths.
Within every channel, quality ranges enormously. Within online video, there is a wide gap between a sound-on, full-screen exposure inside real editorial video content and an autoplay-muted unit on a sticky outstream player. Within CTV, there can be a wide gap between a full-screen smart TV exposure and the same impression delivered onto a phone browser.
Yet buyers still overpay for inventory that looks premium on paper but delivers a weaker experience. Self-declared classifications and surface-level metadata cannot fully capture those differences. The signals that matter are increasingly placement level and observable: player behavior, audibility, screen environment, content adjacency and other characteristics that shape the actual viewing experience.
- Quality should influence pricing, not just filtering.
The market still treats quality as a pass/fail filter: viewable or not viewable, brand-safe or not, eligible or ineligible. This binary logic was sufficient when variance within a channel was narrow, but that’s not the case anymore.
Quality is multidimensional and lives on a range, including player behavior, screen environment, contextual receptiveness, attention conditions, transparency, content adjacency and publisher signals.
Buyers should be defining their own quality thresholds based on the campaign objective, then operate across that range. A brand-building campaign weighs attention and content quality heavily. A response campaign may weigh time of day and receptiveness more. Same signals, different weights. The goal is to stop paying premium prices for inventory that current systems fail to differentiate properly.
- Measure short-term and long-term value together.
At any given moment, most consumers are not actively in-market for what a brand sells. That’s why video’s value extends beyond immediate outcomes. It helps build memory structures, mental availability and brand equity that influence future purchasing decisions and drive long-term profits.
The problem is that we tend to measure short-term effects far better than long-term ones. It’s a lot easier to see clicks and conversions than the value created when someone remembers a brand months later.
That brings us back to why media quality matters. If video plays a critical role in both performance and brand-building, then buyers need to think not just about how many impressions they buy, but the quality of the experiences those impressions create.
What needs to happen next
The challenge now is operationalizing these ideas. As Chris Milano, global VP of supply at Assembly, notes:
“Buyers need a better way to validate the viewing experience they are paying for. The next step is to make observable media quality signals consistently available and enforceable, so pricing and investment decisions reflect the actual opportunity for attention and impact. It is encouraging to see the CIMM paper give the industry a shared framework for moving in that direction.”
That’s exactly what comes next. The industry doesn’t need to wait for a new measurement standard to start making quality-based decisions. The data already exists. What’s missing is the discipline to use it.
For buyers, that starts with three practical steps:
- Audit. Take a recent CTV or OLV campaign and look at how spend and CPMs distribute across placement type, daypart, device and content. Where is your money landing relative to where you actually believe value lives?
- Treat quality as a continuous variable, not a filter. Even a simple step like weighting bids by daypart and player type breaks the pass/fail model and starts aligning cost with value.
- Put quality requirements into RFPs and deal terms. Specify the placement attributes that matter and validate them through reporting or third-party measurement.
The broader goal is straightforward: Stop overspending on and overpaying for low-quality supply and start funding the publishers and environments that actually deliver attention and incremental effectiveness.
The CIMM paper gives the industry a shared language for that conversation. Now it’s on buyers to translate that language into how budgets actually move.
Programmatic video spent the last decade mastering scale. The next phase will be defined by how effectively the industry learns to value quality.
For more articles featuring Elliot MacNay, click here.
