Viewability is the industry measurement of whether an advertisement met a minimum threshold of on-screen exposure, expressed as a share of the ad's pixels visible for a minimum continuous duration. For standard display advertising, the threshold set by the Media Rating Council (MRC), the US non-profit body that accredits media measurement, is 50 percent of an ad's pixels visible for at least one continuous second. Video carries a stricter two-second requirement. The metric exists because a served impression, the count an ad server logs the moment it dispatches a response, says nothing about whether the ad ever rendered where a user could look at it. Viewability closes that gap, without claiming anyone actually looked.

How viewability is measured

A viewable impression sits several steps downstream from an ad request. An ad server selects a winning creative and serves it to a browser or app, and the creative must then load and paint, a stage called begin-to-render, formalised in October 2017 guidelines from the IAB, MRC and Mobile Marketing Association. Only after rendering does viewability measurement begin, checking what share of pixels sit inside the visible portion of the browser tab, and for how long. The standard threshold is 50 percent of pixels for one continuous second on display; large units of 242,500 pixels or more, roughly a 970x250 unit, may use a reduced 30 percent threshold if disclosed to buyers. Video requires two continuous seconds at the same pixel share, and connected television counts as viewable under the same rule, since there is no scrollable viewport to measure.

Most measurement runs on the browser's Intersection Observer API, reporting how much of an element overlaps the visible viewport. According to Google Ad Manager documentation, its Active View technology wraps the rendered creative and measures that element rather than its containing iframe, following MRC guidance that the ad itself, not its container, needs to be visible. Three counting buckets follow, and their ratios are what buyers watch: served impressions form the denominator, measurable impressions are those where the technology captured visibility data, and viewable impressions are the subset clearing the threshold. Google's own example: of 800,000 measurable impressions, 350,000 clearing the bar yields a 43.75 percent viewable rate, calculated against the measurable base.

Publishers monitor viewability to defend inventory pricing and diagnose underperforming placements. Buyers, principally trading desks inside demand-side platforms such as Display & Video 360, The Trade Desk or Amazon DSP, set viewability floors as a targeting parameter. Independent verification vendors, chiefly DoubleVerify, Integral Ad Science (IAS) and HUMAN Security, run measurement tags alongside platform-native tools and reconcile the resulting numbers, which rarely match exactly since vendors use different sampling methods.

Origin and evolution

Viewability formalised what publishers had long suspected: a served impression frequently went nowhere near a user's eyes. Early ad-verification companies, including Alenty, are credited with building the first commercial visibility-measurement tools around 2008, before any shared standard existed. In March 2014 the MRC lifted its advisory against transacting on viewable impressions and formally issued Viewable Ad Impression Measurement Guidelines that June, a threshold produced through more than a year of drafting inside the IAB's Emerging Innovations Task Force followed by a 30-day public comment period, requiring 50 percent of an ad's pixels visible for one continuous second to qualify.

Reconciliation projects across accredited vendors soon surfaced discrepancies in reported counts for identical campaigns, since implementation details, not the headline threshold, produced different numbers. The MRC released Viewable Ad Impression Measurement Guidelines 2.0 in August 2015 to address this while leaving the baseline unchanged, and issued a separate mobile specification since device pixel density and app structures behave differently from a desktop browser tab. In-app measurement carried a further problem: every vendor needed its own SDK bundled into an app's code. The IAB Tech Lab's answer was the Open Measurement SDK, announced in 2017 and released for market adoption in April 2018 alongside a companion protocol letting accredited vendors read standardised signals through one integration, reaching more than 2 billion devices by 2018.

Counting methodology kept moving downstream. Ad servers migrated from a served basis to a downloaded basis, a shift Google Ad Manager completed on October 2, 2017, and begin-to-render, requiring the creative to start painting before counting at all, followed over subsequent years; Display & Video 360 did not complete its own shift until September 2025. A viewable impression can never exceed a begin-to-render impression, which can never exceed a served one. Invalid traffic filtration runs in parallel, splitting traffic into General and Sophisticated categories under a standard the MRC first issued in October 2015, and a viewable impression figure is only as trustworthy as the filtration beneath it, since a change to that filtration can shift the base viewability is calculated against without any change in actual fraud levels.

Why viewability matters commercially

Viewability became the baseline quality signal both sides of a transaction could point to when neither trusted the other's raw impression count. For buyers, it converted an invisible risk, paying full price for an unseen ad, into a measurable, biddable parameter: DSPs let buyers set minimum viewability thresholds pre-bid. For publishers, it created both an incentive and a constraint, since placements that consistently score low, commonly those at the very top of a long page, where users scroll away before the clock elapses, or at the bottom, where users rarely scroll far enough to reach, become harder to sell at full rates.

That commercial weight is why viewability became contested territory. DoubleVerify has reported North America's Authentic Viewable Rate at 71 percent, up 3 percent year over year, with display viewability at 73 percent and video at 79 percent, self-reported figures worth noting whenever a verification vendor publishes its own aggregate rates. Independent accreditation keeps such numbers honest: HUMAN Security's viewability product earned MRC accreditation in April 2026, following a similar accreditation DoubleVerify received for connected TV viewability in April 2024. IAS earned a comparable accreditation in November 2025 for third-party measurement on Amazon DSP, granted only after an independent audit paid for by the vendor seeking accreditation.

Limitations and disputes

The most persistent criticism is structural: the standard measures opportunity to see, not the act of seeing. An ad auto-playing muted in a corner of a screen, technically 50 percent visible for the required second, counts identically to an ad a reader deliberately paused to read. The MRC's own guidelines acknowledge the metric establishes exposure opportunity, not confirmed attention, yet viewable rate is routinely bought against as proof of exposure the methodology does not fully support.

That gap produced a documented commercial dispute. In summer 2024, Kroger Precision Marketing analysed DoubleVerify's Authentic Attention product and, according to trade coverage, could not find an uplift in clicks or sales attributable to placements classified as high-attention, while the same analysis found a positive correlation between conventional viewability and sales. Kroger senior vice-president Cara Pratt characterised attention, in comments reported by industry press, as an unproven metric relative to viewability, and the question of which metric better predicts sales outcomes remains unsettled.

The MRC and IAB responded to the wider standardisation gap by releasing Attention Measurement Guidelines covering data signals, visual tracking, physiological observation and panel-based methodologies. PPC Land's own coverage dates the release to November 2025, though the exact day varies across articles and should be rechecked against the original announcement. The guidelines caution against treating attention as a direct outcome measure, since providers still compute scores through incompatible proprietary methods, meaning a score of 60 in one vendor's system does not correspond to a score of 60 in another.

A more technical limitation concerns measurement gaps: cross-domain iframes can prevent measurement tools from capturing position data, producing impressions logged as not measurable. A separate criticism, raised chiefly by independent ad-fraud researchers, holds that the 50 percent, one-second threshold is an arbitrary compromise reached after years of committee discussion, and that chasing 100 percent viewability can reward fraudulent inventory, since fabricated traffic can trivially show full viewability while real publisher pages, constrained by genuine layouts, rarely can.

Disambiguation

Begin-to-render is an earlier, looser gate than viewability, requiring only that a creative start painting or enter the document object model, with no requirement on how much is visible or for how long; it can be served entirely off-screen and still qualify.

Attention measurement is a later, stricter and less standardised concept. Where viewability is a binary pass or fail against a fixed threshold, attention metrics attempt to score engagement intensity using signals such as dwell time or eye-tracking, computed through incompatible proprietary methods.

Invalid traffic (IVT) filtration is a separate gate applied before impressions are counted at all, addressing whether traffic is human rather than whether it was visible. An impression can be valid yet still fail to qualify as viewable.

Click-through rate describes what a user did after an ad appeared, not whether it met a visibility standard beforehand.

Recent developments

Google implemented changes to multiple MRC-accredited metrics across Display & Video 360 on July 14, 2025, communicated in advance through its Help Center documentation, and independent accreditation activity has continued since, with IAS and HUMAN Security both earning fresh MRC accreditations for viewability and invalid-traffic products through late 2025 and into 2026.

The attention measurement question has generated commercial activity without resolving the underlying standardisation dispute. Adelaide, Lumen Research and IAS have each expanded attention products into new DSP integrations through 2026, and Google began surfacing Adelaide's Attention Unit scores as a bidding signal inside Display & Video 360's custom bidding tools from mid-July 2026. Whether attention will eventually carry the same accreditation weight viewability now carries, or remain a looser layer on top of the accredited floor, is unsettled.

Timeline

  • 2008: Alenty and other early ad-verification vendors introduce commercial ad-visibility measurement tools ahead of any shared industry standard.
  • March 2014: The Media Rating Council lifts its advisory against transacting on viewable display impressions.
  • June 30, 2014: The MRC and IAB formally issue Viewable Ad Impression Measurement Guidelines version 1.0, setting the 50 percent pixel, one-second display and two-second video thresholds.
  • August 18, 2015: The MRC releases Viewable Ad Impression Measurement Guidelines 2.0, addressing discrepancies found through vendor reconciliation projects while keeping baseline thresholds unchanged.
  • October 27, 2015: The MRC issues its first Invalid Traffic Detection and Filtration Guidelines Addendum, splitting invalid traffic into General and Sophisticated categories.
  • 2017: The IAB Tech Lab announces the Open Measurement SDK to standardise in-app viewability measurement across vendors.
  • June 2017: The OMID API reaches version 1.2, documenting the shared ad-session and verification event lifecycle.
  • October 2, 2017: Google Ad Manager completes its transition from served to downloaded impression counting.
  • April 10, 2018: The IAB Tech Lab formally releases the Open Measurement SDK for market adoption alongside a compliance programme.
  • 2018: The Open Measurement SDK reaches more than 2 billion devices, according to IAB Tech Lab reporting.
  • April 2024: DoubleVerify earns MRC accreditation for connected TV viewability.
  • Summer 2024: Kroger Precision Marketing's internal analysis of DoubleVerify's Authentic Attention product finds no uplift in clicks or sales, while finding a positive correlation with conventional viewability.
  • July 14, 2025: Google implements changes to MRC-accredited metrics in Display & Video 360, materially increasing reported invalid begin-to-render impression counts.
  • September 2025: Display & Video 360 completes its shift from count-on-download to begin-to-render counting for display.
  • November 2025: The MRC and IAB release Attention Measurement Guidelines, standardising vocabulary without mandating a single measurement methodology.
  • November 6, 2025: Campaign Manager 360 completes its shift from count-on-download to begin-to-render counting.
  • November 13, 2025: Integral Ad Science earns MRC accreditation for third-party measurement on Amazon DSP.
  • April 27, 2026: HUMAN Security earns MRC accreditation for its viewability measurement product.
  • Mid-July 2026: Adelaide's Attention Unit becomes available as a bidding signal inside Display & Video 360 custom bidding.

Summary

Who: The Media Rating Council, working with the Interactive Advertising Bureau, sets and accredits the viewability standard; independent verification vendors including DoubleVerify, Integral Ad Science and HUMAN Security measure against it; and platforms including Google Ad Manager, Display & Video 360 and Amazon DSP implement it inside their own reporting stacks.

What: A measurement standard requiring a minimum share of an advertisement's pixels, 50 percent for standard display and video, 30 percent for large display formats, to remain visible in a browser's in-focus viewport for a minimum continuous duration, one second for display and two seconds for video, before an impression counts as viewable.

When: The MRC lifted its advisory against transacting on viewable impressions in March 2014 and formally issued the first Viewable Ad Impression Measurement Guidelines that June, following commercial ad-visibility tools that predate any shared standard back to 2008.

Where: Viewability measurement operates across desktop and mobile web browsers, in-app mobile environments through the IAB Tech Lab's Open Measurement SDK, and connected television, wherever an ad server, DSP or independent verification vendor can capture position and exposure data for a rendered creative.

Why: A served impression alone cannot show whether an advertisement ever appeared where a person could look at it; viewability gives buyers and sellers a shared, independently accredited floor for what counts as a genuine opportunity to be seen, even as the industry continues disputing whether that floor says enough about whether anyone actually paid attention.