Media quality remains one of advertising's most elusive concepts, with buyers unable to agree on what it means or how to measure it, creating both challenges and opportunities in the programmatic marketplace.

"Quality is like our industry's 'six-seven,'" said David Simon, chief revenue officer of Verve and president of Verve Marketplace, on a recent episode of "Behind the Numbers." "It genuinely is so subjective that doing the research was interesting because there were directly contradictory data points."

EMARKETER and Verve conducted a recent survey to understand how advertisers evaluate inventory quality, their confidence in measurement frameworks, and where gaps remain. The findings reveal an industry grappling with inconsistent definitions and missed opportunities.

The quality paradox

Media buyers say quality matters, yet struggle to act on it. While some respondents identified reach as a primary quality indicator, others pointed to brand-safe environments, high dwell time, quality audience signals, or exclusive platforms.

"Some people say quality is super important and a lot of people say, 'I can't buy quality,'" Simon said. "If it's so important and you know what it looks like, then why aren't you buying it?"

Without standardized quality definitions, advertisers must develop brand-specific criteria rather than relying on industry labels. This requires understanding what drives outcomes for their specific products and audiences.

Reach dominates, but doesn't deliver

Reach emerged as the most common quality metric, reflecting a legacy approach to media buying that prioritizes scale over precision.

"Brands need to continually reach new eyeballs, need to continually engage new consumers," said our analyst Arielle Feger. "And I think that's just probably the easiest way to know that you are reaching a certain number of people."

However, reach doesn't guarantee effectiveness. It doesn't ensure advertisers reach the right people or the most engaged audiences.

For connected TV specifically, the top evaluation method was audience data quality, not reach, suggesting buyers are beginning to prioritize precision over scale in fragmented channels.

The emphasis on reach reflects measurement convenience rather than business outcomes. Advertisers should define quality through the lens of revenue-positive outcomes rather than defaulting to volume metrics.

The transparency trap

Buyers cite lack of transparency as a barrier to assessing quality, but Simon argues this misses the point.

"If I'm selling you a revenue-positive outcome, how much do you care about the transparency?" Simon said. "Assuming it's a real user who put their credit card in and made a purchase, do you really care at an intimate level like how exactly I made the decision to do that?"

He compared media buying to purchasing a chicken sandwich: Consumers don't ask about ingredient costs if the product delivers value at an acceptable price.

The transparency debate often serves as proxy for fee negotiations rather than quality assessment. On performance-focused campaigns, particularly in gaming and app install, advertisers rarely demand granular transparency because they receive clear conversion signals.

Rather than demanding transparency into every cost component, advertisers should focus on sharing more conversion data with partners. This enables machine learning systems to optimize toward actual business outcomes rather than proxy metrics.

Machine learning makes transparency harder

As advertising increasingly relies on machine learning for decisioning, explaining individual ad placements becomes virtually impossible.

"If you look at some of the most sophisticated models that predict the likelihood of an outcome and are in charge of buying media profitably, they have models that are millions of features long," said Simon. "To identify why that model made an exact decision is virtually impossible. But it doesn't matter because you got the decision that you wanted and the outcome was generated."

He said performance advertisers like Instacart share six conversion steps with buying partners, from app install through second purchase, enabling optimization without requiring placement-level transparency.

Traditional brand advertisers often share less data due to legal constraints or competitive concerns, limiting their partners' ability to optimize effectively.

Brands must share more conversion signals with agencies and platforms to enable effective machine learning optimization. The alternative is continuing to optimize toward imprecise proxy metrics while competitors leverage superior data.

Listen to the full episode.

This was originally featured in the EMARKETER Daily newsletter. For more marketing insights, statistics, and trends, subscribe here.