Ad Viewability and IAB Standards: A Publisher Guide

Ad viewability tells you whether an ad you placed on a page actually entered the visible part of a visitor's screen; an ad that was served but sat below a fold nobody scrolled to still shows up as an impression in the counter. The sections ahead explain the IAB and MRC viewability definition, why measurement vendors report different rates, and how a publisher can turn the rate in hand into placement decisions.

At the end you will find a worksheet for tracking your ad placements in one table and a regular review flow. Figures appear only as examples; your real values must come from your own measurement.

What Ad Viewability Is

An ad server counts an impression when it delivers the ad as the page loads. Yet that ad may sit at the very bottom of the page, in a section the visitor never reaches. Viewability closes this gap: it tells you whether the ad appeared in the visitor's viewable area and met a defined set of criteria.

The idea has two sides. The technical side measures which pixels were visible and for how long. The commercial side reflects that an advertiser wants to pay for impressions that had a chance to be seen. That is why viewability began as a request from buyers more than from publishers; today it is also read as an indicator of inventory quality.

One distinction up front: viewable does not mean viewed. The ad can be on screen while the visitor is looking elsewhere. The standard measures only the opportunity to see; attention, recall and clicks are separate concepts. We cover click and performance measurement in the guide to measuring banner ad performance; this article looks at the step before that, whether the ad reached the screen at all.

Here is why it matters to a publisher. Two ad slots on the same page can produce the same number of impressions while one sits in the first screen and the other in a lower section most visitors never reach. The impression count treats them as equals; the viewability rate separates them. In price conversations, in placement reviews and in the media kit you send to buyers, that separation becomes decisive.

The IAB and MRC Viewability Definition

The source of the definition is the MRC (Media Rating Council) guideline prepared in collaboration with the IAB. The MRC viewable ad impression measurement guideline (Opens in a new tab) hosted by the IAB (Version 1.0, June 30, 2014) describes a viewable impression as a served impression that was contained in the viewable space of the browser window, on an in-focus tab, based on pre-established criteria. The document adds that a viewable impression carries an "opportunity to see" the ad, which a plain served impression may not.

For display ads the criteria have two parts:

• Pixel requirement: at least 50% of the ad's pixels must be in the viewable space of an in-focus browser tab.
• Time requirement: that pixel requirement must be met for at least one continuous second after the ad renders.

The guideline also fixes the order: the pixel test comes first, and the clock starts only once the pixel threshold is met. For video the time requirement is two continuous seconds at the same pixel requirement.

The large ad exception

According to the document, for display ads of 242,500 pixels or more (equivalent to a 970 x 250 unit) a viewable impression may be counted at 30% of pixels instead of 50%, but only if this practice is fully disclosed to data users. If you run a large masthead unit, ask your vendor which threshold the report uses.

The guideline further says that when a measurer can establish a strong user interaction, the ad may be counted as viewable even without meeting the pixel and time criteria, and those impressions should be reported separately. This guide relies on the 2014 version; ask your measurement vendor which current version of the guideline it is accredited against.

Viewable Impressions Versus Impressions

Impressions and viewable impressions are different counters, and when a report shows both, the gap between them is the first clue to placement quality. An impression records that the ad was served; a viewable impression records that the criteria above were met. The viewable count can never exceed the impression count.

Reports usually mix up three separate ideas:

• Served impressions: the number of ads the ad server delivered.
• Measurable impressions: impressions whose viewability could be measured. In some environments the measurement code cannot run, so viewability is unknown.
• Viewable impressions: measurable impressions that met the criteria.

When you read a viewability rate, find out what the denominator is. Is the rate divided by served impressions or by measurable impressions? The two methods give different results for the same page, because the rate changes depending on whether unmeasurable impressions stay in the denominator. Confirm the method in the vendor's documentation; that way you and the buyer speak the same language.

Imagine a content site (an example scenario) that sees high impressions and a relatively low viewable count in its monthly report. Before hunting for the cause, the team checks the measurable rate and finds it is low. The problem is therefore a measurement coverage issue before it is a viewability issue; a page template or browser environment where the measurement code does not run may be dragging the numbers down. Questioning coverage before changing placements saves time.

Why Measurement Vendors Report Different Rates

If you measure the same page with two tools and see two different rates, that is common and usually not a bug. On its Open Measurement SDK page, IAB Tech Lab notes that without a standard framework, core signals such as on-screen placement are "collected using different methodologies," which makes results hard to compare. The same page says the SDK facilitates third-party verification and can work with multiple verification partners at once.

Typical reasons for a gap include:

• Measurement method: whether the tool relies on browser APIs or on in-page calculation changes the result.
• Coverage: ads running inside iframes are measurable to different degrees, and tools do not all handle them the same way.
• Denominator: the rate definition difference described above.
• Thresholds and exceptions: whether the 30% large-unit threshold or an interaction rule is applied.
• Sampling and timing: polling interval and reporting delay.

So instead of asking which tool is right, ask which one you will decide with and which one the buyer accepts. A practical rule for publishers is to pick one reference tool and compare periods using only that tool. If you switch tools, record the date of the switch; otherwise a jump in the rate may look like a placement improvement.

Placement Decisions and Viewability

Viewability is most directly tied to placement, and placement is the variable a publisher controls most. The MRC document itself reminds readers that placement and viewable status are separate attributes: being at the top of a page does not make a slot viewable, because the "fold" moves with browser settings, resolution and window size.

When deciding on a placement, ask these questions in order:

1. Does the visitor see this slot without scrolling, or does it need scrolling?
2. Does it sit in the part of the page people actually read, or in gaps they skim past?
3. Does the visitor linger near the slot, meaning is the content dense there?
4. Does the layout shift while the slot loads?
5. Where does the same slot land on mobile and how much room does it take?

Slots that sit beside read content at natural pause points tend to stay on screen longer, while slots pushed to the end of the page produce low viewability because many visitors never get there. Still, do not turn this into a fixed rule that the top slot is always more viewable; your own measurement exists to test that assumption.

Placement also has a user experience boundary. Moving slots in front of the content may lift the rate in the short term and still drive readers away. We discuss that balance in detail in the guide to ad placements and user experience; keep that limit in view whenever you make a viewability decision.

On mobile, placement is a separate question: a narrow screen has one column, and sidebar slots either drop into the content or disappear. Track a slot's mobile counterpart as its own unit rather than deriving it from the desktop rate, and confirm current mobile guidance with your measurement vendor, since the MRC document says its guideline was designed primarily for desktop browser ads.

Lazy Loading and Sticky Units

Lazy loading, which defers an off-screen ad until the visitor approaches it, can help page speed and cuts both ways for viewability. If the ad is not loaded until the visitor gets close, no resources go to slots that would never be reached, and the viewable rate is not artificially depressed, since those slots were never counted as impressions either. But if the loading threshold is set too late, the ad appears after the visitor has already passed the slot, and the time requirement cannot be met.

When using lazy loading, check the following:

• Does loading begin before the visitor reaches the slot? The response time of the ad request must be factored in.
• Is space reserved for the slot? If not, content shifts when the ad loads.
• Is the impression counted when the ad actually renders, or when the request is sent?

Sticky units stay fixed on screen while the visitor scrolls. This lengthens time on screen and makes the time requirement easy to meet. Yet the same behavior tires the visitor: sticky units that shrink the reading area, cover content or hide the close button can push people to leave or install a blocker. The risk is higher on mobile, where a sticky unit can take up much of the visible surface.

The Effect of Page Speed and Scroll Behavior

Viewability is shaped not only by design but also by how the page loads and how visitors behave. Because the time requirement starts after the ad renders, an ad that loads late may still be loading when the visitor leaves the page and cannot count as viewable. A slow page weakens both the visitor experience and viewability.

Scroll behavior is a separate variable: a short news summary keeps most visitors in the first screen, while a long guide invites scrolling, so the same unit yields very different rates. Read viewability by page type, not as one site average.

On the speed side, watch for these points:

• Ad scripts should not delay the page's main content.
• Reserve space for the slot; layout shift harms both the experience and the measurement.
• If many ad requests fire at once, responses arrive late and a slot can stay empty.

We cover the balance between ads and page performance in the guide on ads and Core Web Vitals. The criteria there describe page experience and the criteria here describe the chance to see an ad; the two often improve together but neither replaces the other.

Viewability Expectations from the Buyer Side

For an advertiser, viewability is evidence that budget is not being wasted. A buyer may want to pay for viewable impressions or buy only inventory above a given viewability level. This request signals that the impression count a publisher reports is not considered enough.

From the buyer's side, viewability usually turns into questions like these:

• With which method and tool is this slot measured?
• Does the report use served or measurable impressions as its base?
• Are mobile and desktop rates reported separately?
• Which threshold is used for large units, and is it disclosed?

Answering these questions in advance builds trust. It also explains where viewability fits in pricing: it does not change the pricing model, but it becomes an argument in negotiation. We compare which pricing model (per thousand impressions, per click or flat fee) suits which situation in the guide to ad pricing models. How viewability affects price varies from buyer to buyer; there is no formula that says a given rate brings a given price increase.

Transparency builds buyer trust too. IAB Tech Lab describes the mission of the ads.txt (Opens in a new tab) project as increasing transparency in the programmatic advertising ecosystem, a way for publishers to publicly name the companies authorized to sell their inventory. Our ads.txt guide covers the setup; a tidy file next to your viewability report tells buyers about quality and transparency together.

Interpreting the Viewability Rate and Setting Targets

Once you have a rate, the first need is to know what to compare it with. There is no universal "good rate," and quoting an industry average without a verified source misleads. The most reliable comparison is your own earlier periods and the other slots on the same site.

When setting targets, follow these principles:

• Per-slot targets: set a separate target for each slot rather than one site-wide figure; a slot may have a structural limit.
• Separate by device: target mobile and desktop separately.
• Direction targets: begin with a direction such as "better than the previous period" instead of a fixed number.
• Traffic volume: the rate of a slot with very few impressions is volatile; wait until enough volume accumulates before interpreting it.

Raising the rate should not be the goal on its own. If you remove a slot, the rate of the remaining slots rises while total revenue may fall. So track viewability together with revenue: when the rate and revenue per slot are read together, you can tell a truly productive improvement from a change that merely polishes the number.

Imagine a guide site (an example scenario) that removes a slot at the end of the page and sees the rate climb on the remaining slots. When revenue per slot and total revenue are examined together, the total has dropped. Instead of removing the slot, the team decides to move it to a natural pause point in the middle of the content and measures the outcome the following period. The decision rests not on the number but on reading the numbers together.

A Regular Viewability Review Flow

Viewability is not something you set once and forget. Template changes, new ad units, browser updates and changes in the measurement tool all shift rates over time. A short monthly flow that runs in the same order lets you catch that drift early.

The suggested flow has five steps:

1. Measurement health: check whether the measurable rate is stable. A sudden drop usually points to broken measurement code.
2. Slot ranking: rank slots by viewability rate and revenue and flag the lowest two or three.
3. Cause search: for flagged slots, look at placement, load timing, layout shift and the device breakdown.
4. One change: pick a single change per slot and record its date.
5. Result check: next period, compare the same slot with the same criteria.

During the review, evaluate direct sales slots and programmatic slots separately. In direct deals the buyer may have requested a specific placement, while in programmatic the slot serves many buyers. How the two approaches differ is explained in the article on direct ad buying versus programmatic.

Sample Template: Ad Placement Viewability Worksheet

This worksheet gathers the monthly review into one table. The values are example values that only demonstrate the method of filling it in; replace them with your own measurements.

Imagine a team that, while completing the worksheet, sees the end-of-page slot trailing on both viewability and revenue. Scroll depth data confirms that most visitors never reach that point; the team moves the slot to a natural pause in the middle of the content and compares the result next period. The value of the worksheet is that every slot appears in one row with an owner and a next step.

Frequently Asked Questions

What does ad viewability mean?

Ad viewability means an ad entered the visitor's visible screen area and met defined criteria. A served ad that never reached the screen is not counted as viewable. The measure describes the opportunity to see the ad; it does not mean attention, recall or a click.

What is the IAB viewability standard?

Under the MRC guideline hosted by the IAB, a display ad needs at least half of its pixels in an in-focus tab for at least one continuous second. Video needs two seconds. Large units may use a 30% threshold if it is disclosed. Confirm the current version with your measurement vendor.

How do you improve ad viewability?

Place slots where visitors actually read, reduce layout shift and slow loading, and move slots that visitors rarely reach. Avoid making several changes at once and measure each change separately. Track viewability together with revenue, not as a rate alone.

Are above-the-fold ads always more viewable?

Usually, but this is not a rule. The fold depends on the visitor's browser settings, resolution and window size. If visitors scroll fast or the ad loads late, a top slot can still show a low rate. Check your own site's measurement before assuming.

Does viewability affect ad pricing?

It can, but there is no fixed rule. Some buyers pay more for viewable impressions or buy only inventory above a viewability level. The size of the effect depends on the buyer, the contract and the pricing model; no particular rate can be tied to a particular price.

Which tool measures viewability?

Ad servers and third-party verification tools measure it. Different tools can report different rates because of method and coverage, so choose one reference tool and compare periods with it. The Open Measurement SDK helps several verification partners measure within a shared framework.

Conclusion and Next Step

When you read a viewability rate, first clarify the definition and the measurement method, then look slot by slot, and only then go into placement, loading and scroll behavior, testing each change on its own. A rate can be interpreted only when the denominator, the threshold and the tool are known; going into detail without those three leads to wrong decisions.

Before you buy a slot, put this guide's questions about position and measurement method to it; then pick one on the ad spaces page. You choose a space, complete cart and payment, and the seller's approval and a delivery notice follow, with a dispute process available if needed.