Google to Visualize the "Annoyance" of Ads - Why Cheap Ad Slots Actually Cost More

When you start reading an article, ads rise from the bottom of the screen. Even if you scroll, videos follow you, and ads line up between the text you finally found.

At such times, readers feel a vague dissatisfaction that "this site has too many ads." However, this experience may not be fully reflected in the reports of the companies placing the ads. The numbers of how many times it was displayed and at what cost do not reveal what was happening around the ads.

Google is trying to shed light on this less visible part. On September 15, 2026, four experimental ad metrics were added to the "Chrome User Experience Report," commonly known as CrUX, which aggregates actual user experiences on Chrome.

An article published by the Australian advertising industry media B&T on October 19 welcomes this move while urging important caution. Improving the advertising environment and making the ads themselves effective are not the same thing.

It's not about measuring "annoyance scores," but the quantity and load of ads

Describing the new metrics as a "system to measure the annoyance of ads" is intuitively understandable. However, Google is not reading users' emotions or scoring discomfort.

What is measured are four elements surrounding the ads.

The point is that it's not just a system that looks at the total number of ad slots.

For example, even if there are multiple ads in a long article, if only one is visible on the reader's screen at a time, the experience differs from a page where multiple ads are always displayed. Also, the experience cannot be compared by numbers alone when there are many small ads versus one large ad occupying the screen.

By adding processing and communication burdens to the number and occupancy rate of ads, it becomes easier to investigate "what leads to difficulty in reading or heaviness."

However, this alone does not reveal the appropriateness of the ad content, the discomfort of seeing the same ad repeatedly, or the favorability towards the brand. The new metrics visualize part of the advertising experience.

There is a distance between "appeared on screen" and "communicated to people"

Leon Stepien, SEO manager at Intender, who contributed to the B&T article, questions the meaning of "cheapness" in advertising costs.

In advertising operations, CPM, which indicates the cost to display 1,000 times, is often used. If more can be delivered with the same budget, it seems as if efficiency has improved.

However, what if there are a lot of other ads around the ad, and the reader is focused on finding the text? Even if the ad is delivered, it may not be recognized.

In general display ad viewability standards, it is basic that more than 50% of the ad area is displayed continuously for more than one second. This is a standard for "having had the opportunity to be seen," not proof of "recognized the content" or "was interested."

Stepien gives an example where even if the CPM is halved, if the proportion of attention-grabbing displays is one-third, the cost per noticed display becomes 1.5 times.

Converted to hypothetical numbers in Japanese yen, it looks like this:

This is a hypothetical explanation of the mechanism and does not mean that CrUX can measure "number of noticed times."

Even so, the difference between buying cheap displays and obtaining effective contact points cheaply becomes clear. Comparing only advertising costs overlooks something.

Expectations for transparency on social media—however, mainly reactions from industry insiders

In publicly available LinkedIn posts about the new metrics, there is an expectation for easier comparison of advertising environments.

Adelaide, which handles ad attention measurement, posted that it will be easier for advertisers to understand the advertising environments they purchase, and for media to demonstrate the quality of the advertising experiences they have built. The company has also revealed its cooperation in this initiative, which should be read as a welcome opinion from companies involved in the development.

Practical implementation has also begun. Dave Smart evaluates the new metrics dealing with ad count, density, CPU, and network traffic as useful and reported adding them to his free history visualization tool.

On the other hand, Muhammad Abuzar's post warns against treating the new metrics as a new scoring system for search rankings. The idea is to separate and confirm the amount of ads, screen occupancy, processing load, etc., along with usage and revenue.

The posts introduced here are public posts from industry insiders about the new metrics, not reader comments on the B&T article. Nor is it a survey showing the pros and cons of the general user population.

What can be read from the confirmed posts is an interest in increasing materials for comparing advertising experiences, rather than a simple demand to "reduce ads."

Assuming "the fewer ads, the better" leads to other failures

Pages with many ads are worth investigating. However, it is premature to uniformly exclude media with high ad density from the target of submissions.

For advertisers, there is a difference in meaning between media where people gather to read highly specialized articles enthusiastically and media that looks neat but hardly attracts target customers. The number of ads alone does not reveal that difference.

Google also explains that the current metrics are not part of Core Web Vitals and do not have recommended target values or thresholds. There are no judgments such as "good," "needs improvement," or "poor."

Therefore, there is no official rule that automatically disqualifies if ad density exceeds a certain percentage. It cannot be concluded that search rankings will drop solely based on the publication of the new metrics.

What is needed is to look at the numbers, form hypotheses, and verify them with actual results.

Did purchases or inquiries increase after changing the advertising environment? If the purpose was brand awareness, was the brand remembered? If the ad content and target audience have changed significantly at that time, it becomes unclear what led to the improvement.

Simply replacing "buy the cheapest slot" with "buy the slot with the fewest ads" does not improve the quality of judgment.

In Japan, check the actual smartphone screen and data coverage

For Japanese advertisers and media operators, the current metrics can serve as material to review readability on smartphones.

For example, when an ad fixed at the bottom of the screen overlaps with an ad in the middle of an article, how much of the text remains readable? A layout that was not bothersome on a computer might create a sense of pressure on a smartphone. By comparing numbers with the actual screen, specific improvement points can be found.

However, CrUX does not cover all users.

According to Google's explanation, the target platforms include Chrome on Windows, macOS, and Android, but not iOS version Chrome, Android's in-app WebView, or other Chromium-based browsers. Even for sites targeting Japan, one should not assume that this data alone covers experiences on iPhones or in-app browsers.

Furthermore, there are conditions for the publication of ad metrics. It is necessary to have at least one certified seller listed in "ads.txt," which declares the sellers who can sell ad slots.

Just because data does not appear does not mean "there are few ads" or "there is no ad burden."

The published values are also aggregates of actual usage over a certain period. It is practical to use display verification on actual devices for immediate confirmation after changes and to track changes with aggregated data later.

Media should look at "loss from reducing ads" and subsequent actions

For sites operated by advertising revenue, deciding to reduce ads is not easy. Creating articles, conducting interviews, and system operations incur costs. Reducing display slots may also reduce immediate income.

Therefore, there is no need to limit improvements to just reducing the number of ads.

For example, reviewing high-load ads, adjusting layouts that cover the text, and limiting the number of ads visible simultaneously on smartphones are possible methods.

In addition to per-page ad revenue, check if readers continued reading the article, read other articles, or returned again. Making it easier to read might increase continuous usage, but it does not always do so. It needs to be verified with one's own readers and ad configuration.

Advertisers also have reason to check the advertising environment of major submission sites, rather than just increasing the budget because the unit price is low. It is easier to judge by first comparing with part of the budget and seeing if results that match the purpose are obtained, rather than excluding based solely on metrics.

Even if ad slots are organized, if the content the audience wants is omitted, it won't reach them

Even if the advertising environment is improved, if the ad content does not match the audience's interests, it will not lead to effectiveness.

For example, someone just starting to look for online courses might want to know if "beginners can keep up," "can it be balanced with work," or "what can be learned." Being urged to "apply by today" at that stage does not fill in the information needed for judgment.

On the other hand, for someone who already understands the course content and is considering applying, information on start dates and fees would be helpful.

The meaning of the same ad changes depending on what stage the audience is in. This cannot be determined from the number of ads or network traffic.

The value brought by the new metrics lies in being able to add the question "in what environment was it displayed" when considering how to use advertising expenses.

Being displayed, being noticed, and being useful are each different. Now that the burden of ads can be measured, the next question is whether the ad has a reason to use the reader's time.

Source

1. B&T "Google Is Starting To Measure How Annoying Advertising Is"
   A discussion by Leon Stepien on the difference between advertising environments and effects, and the relationship between CPM and attention.
   https://www.bandt.com.au/google-is-starting-to-measure-how-annoying-advertising-is/

2. Google Chrome Official "New ad metrics in Chrome User Experience Report"
   Announcement on September 15, 2026. Confirmation of four experimental metrics, differences from Core Web Vitals, and the absence of recommended thresholds.
   https://developer.chrome.com/blog/crux-ad-metricsdeveloper.chrome.com

3. Google Chrome Official "CrUX ad metrics overview"
   Overview of each metric, aggregation period, target and non-target platforms, and confirmation of no evaluation categories.
   https://developer.chrome.com/docs/ads/metricsdeveloper.chrome.com

4. Google Chrome Official "Ad Count" "Ad Density"
   Definitions of ad count and screen occupancy rate, explanation of measurement methods based on the viewport during browsing.
   https://developer.chrome.com/docs/ads/metrics/count
   https://developer.chrome.com/docs/ads/metrics/densitydeveloper.chrome.com

5. Google Chrome Official "CrUX ad metrics measurement methodology"
   Explanation of conditions for public targets by ads.txt, measurement range, and aggregation method.
   https://developer.chrome.com/docs/ads/methodologydeveloper.chrome.com

6. Adelaide's LinkedIn Post
   A post expressing expectations for increased transparency and evaluation of media's advertising experience, and declaring cooperation in the current initiative.
   https://www.linkedin.com/posts/adelaidemetrics_new-ad-metrics-in-chrome-user-experience-activity-7505981585524027392-IKlULinkedIn##