Why is viewability a poor guide to site quality?
Part of MFA avoidance
The short answer
Viewability records whether enough of an ad was on screen for long enough, so it measures the slot, not the site. Made-for-advertising sites post good viewability and low invalid traffic, as the ANA and its partners noted, because their layouts are built to keep ads in view. Judge editorial quality separately, then compare viewability among sites that already pass.
Key takeaways
- Viewability describes an opportunity to see an ad; it reads nothing in the article around it.
- The trade groups that defined MFA in 2023 said these sites exhibit high measurability rates, good viewability rates and low levels of invalid traffic.
- A strong publisher can post a modest viewable rate because of long articles, bottom of page slots or screen size.
- Viewability goals work best when they are applied within a vetted inclusion list and read by site and device, never as a ranking of sites.
How can I distinguish viewability from editorial quality?
Ask what each one is about. Viewability is about the ad's position and timing. The Media Rating Council's Viewable Ad Impression Guidelines (opens in a new tab) (version 2.0, August 18, 2015) classify a served impression as viewable based on pre-established criteria such as the percent of ad pixels within the viewable space and the length of time the ad is in the viewable space of the browser, and recognize that an opportunity to see the ad exists with a viewable ad impression. Editorial quality is about the content: who made it, whether it is original and whether a reader would choose it.
Definition
- Editorial quality
- How far a site's content is original, accurate, attributed and made for readers, judged from the articles themselves rather than from the ads that surround them.
Google's Active View (opens in a new tab) rules show how mechanical the test is. A display ad is counted as viewable when at least 50% of its area is visible on the screen for at least 1 second; large display ads of 242,500 pixels or more need 30% of their area for one second; and video needs 50% of its area on screen while playing for 2 seconds. None of these thresholds looks at the page's text.
| Question | Viewability measurement | Editorial review |
|---|---|---|
| What is being judged | One ad impression in one slot | The site's articles, sampled over time |
| Evidence used | Pixels in view and seconds on screen | Bylines, originality, sourcing, dates, purpose |
| When it is known | After the impression is served | Before the site joins a plan |
| What a strong result shows | The ad had a chance to be seen | The page is worth being seen beside |
| What it cannot show | Whether anyone read the page or would return | Whether a given slot is in view |
Why do MFA sites score well on viewability?
The trade groups said so when they defined the category. The September 2023 announcement (opens in a new tab) by the ANA, the 4A's, the WFA and ISBA states that MFA websites exhibit high measurability rates, good viewability rates, low levels of invalid traffic, and usually have brand-safe environments. The same announcement lists the layouts behind those numbers: a high ad-to-content ratio, numerous refreshing banner ads, autoplay video and slide shows forcing visitors to click through multiple pages with multiple ads.
Each of those layouts keeps ad slots on screen. A short page with ads stacked above the fold leaves little room for a slot to sit out of view. A slot that refreshes while the visitor stays put serves new impressions into space that is already visible. A slideshow that loads a new page per click renders a fresh set of ads at the top of each one. Visitors bought from real people are not bots, which fits the low invalid traffic the trade groups described. The result can be a site that looks efficient on a delivery report and offers readers little.
The cluster's pages on ad refresh and slideshow arbitrage sites cover those two layouts in depth. The point here is narrower: strong viewability is what an ad heavy layout can produce, so on its own it cannot separate a good site from a made-for-advertising one.
Can a quality site have low viewability?
Yes, and for reasons that say nothing about its journalism. Google's AdSense page on Active View data (opens in a new tab) gives the basic case: if a user needs to scroll down to see the bottom of the page, the user might not see the ad, yet for a user with a larger screen, the entire page might appear in a viewable space. Long articles, slots placed after the story and readers who leave once they have their answer all lower the viewable rate on sites that serve readers well.
Device matters as well. The MRC's Mobile Viewable Ad Impression Measurement Guidelines (opens in a new tab) (June 28, 2016) tell measurers to segregate mobile ad impressions from desktop ad impressions in their reporting, and say measured rates can differ significantly between desktop and mobile. A blended site figure can hide a strong phone experience behind weak desktop rails, or the reverse.
| Cause | Why the rate falls | What to ask the seller |
|---|---|---|
| Long form articles | Mid article and end of article slots load far below the first screen | Viewable rate by slot position, not for the site as a whole |
| Fewer ads above the fold | The publisher keeps the opening screen for content | Which slots are offered in the deal |
| Readers who leave once answered | Lower slots never come into view | Viewability for the slots the plan will buy |
| Device mix | Desktop and mobile rates differ and blend into one figure | Separate mobile web and desktop figures |
| Low measured rate | Unmeasured impressions shrink the reliable sample | The measured rate beside the viewable rate |
Which quality checks should sit beside viewability?
Viewability belongs in a set of checks, each answering a different question, as part of the standing practice described on the MFA avoidance hub. The page level checks come first, because they decide whether a site belongs on the plan at all. Google's own publisher rules include one such check: Google says it does not allow Google-served ads on screens with more ads or other paid promotional material than publisher-content (opens in a new tab).
Checks to run beside a viewability figure
- Sampled articles are original, dated and attributed, read on both phone and desktop.
- Ads take up less of the screen than the publisher's own content on the pages sampled.
- Ad slots do not reload on a timer while the visitor stays on the page.
- Stories are not split into click through galleries that load new ads per page.
- Traffic does not depend on sponsored link units or reward offers.
- The site's ads.txt and the exchanges' sellers.json files agree on who sells it.
- The site has a real audience in the campaign's market, shown by public popularity data.
- Viewability is reported by device and slot, with the measured rate shown.
How should viewability goals be set without rewarding MFA?
A goal applied across the open exchange pulls spend toward whatever inventory clears it with the least effort, and MFA layouts can clear it, as the trade groups' description of good viewability rates on MFA sites suggests. Google's Display & Video 360 help on viewability targeting (opens in a new tab) adds a reach warning: impressions with a high predicted viewability can be a small subset of all inventory that's available, and a high setting will result in fewer matches. Setting the bar higher shrinks the pool and leaves the sites that are easiest to keep in view.
Five rules for a viewability goal
- Apply the goal inside an approved inclusion list. The list decides which sites can run; the goal decides which slots on those sites are worth paying for.
- Set targets by device and format. Use separate figures for mobile web and desktop, following the MRC's segregation guidance.
- Report the measured rate with the viewable rate. A high viewable rate on a small measured sample is weak evidence.
- Treat a site far above its peers as a question. Open its pages and check for refresh, stacking and galleries before praising the number.
- Never rank or add sites by viewable rate. Use viewability to compare slots and prices among sites that already passed the quality checks.
What does a viewability review look like on a real brief?
Fictional example
A tile showroom's display plan in Arizona and New Mexico
A fictional ceramic tile and stone showroom with locations in Phoenix, Tucson and Albuquerque asks its agency for English display beside home design, remodeling and local news content. The client's previous agency optimized toward a high viewable rate goal and reported strong numbers.
The planner breaks the past delivery out by site. Three home decor sites at the top of the viewability table share a short page layout with ads between paragraphs, slots that reload while the page sits open and stories split across galleries. A regional design magazine near the bottom of the table runs long features with slots after each section, and its mobile figure is far stronger than its desktop figure.
The three decor sites go to the exclusion list on their layouts, with the viewability figures kept in the record as context. The magazine stays, bought on its mobile placements and its first two article slots. The new plan sets viewability targets by device inside the approved list.
How does PlacementGraph look at layout and quality before viewability?
Frequently asked questions
Is a viewable rate near 100 percent a warning sign?
It is a reason to look, not a finding. Layouts that stack ads in the first screen or reload slots in place can produce very high rates, and so can a short, well built page with one slot at the top. Open several articles on phone and desktop and check for refresh and galleries before deciding.
Should viewability be written into deal terms?
It can be, as long as the deal's site list was vetted first. A viewability term then governs the slots and price within approved inventory. Name the measurement standard, ask for mobile and desktop to be reported apart, and include the measured rate so the term rests on a reliable sample.
Does a non-viewable impression mean the site is low quality?
No. Under the MRC guidelines a non-viewable impression is one that did not meet the pixel and time criteria, which can happen on a long article where the reader stopped before the slot. It tells you about the slot's position and the reader's scroll, not about the reporting.
Why are viewability figures different on mobile and desktop for one site?
Layouts, screen sizes and measurement methods differ. The MRC's mobile guidelines say measured rates can differ significantly between desktop and mobile, partly because, when they were written in 2016, measurement solutions were more limited on mobile, and ask measurers to report the two apart. Compare each device's figure with peers on the same device.
Sources
- Leading Trade Groups Define “Made For Advertising” Websites (opens in a new tab)ANA, with the 4A's, WFA and ISBA, September 26, 2023
- MRC Viewable Ad Impression Measurement Guidelines, Version 2.0 (opens in a new tab)Media Rating Council, August 18, 2015
- MRC Mobile Viewable Ad Impression Measurement Guidelines, Final Version (opens in a new tab)Media Rating Council, June 28, 2016
- Active View (opens in a new tab)Google Ads Help
- Active View data in reporting (opens in a new tab)Google AdSense Help
- Viewability targeting (opens in a new tab)Google Display & Video 360 Help
- More ads or paid promotional material than publisher-content (opens in a new tab)Google Publisher Policies Help
Researched and written by PlacementGraph Research. Read how we research, source and update these pages. Corrections: diana@placementgraph.com.