Google today began rolling out a card in Google Ads that puts two sums in front of advertisers: the amount the company says it saved them by filtering invalid traffic before billing, and the credits it issued for invalid activity caught after an invoice had gone out. The announcement came in a three-paragraph post on Google's Ads & Commerce Blog by Keerat Sharma, VP & General Manager, Ads Privacy and Safety, titled "We're introducing a new transparency tool for invalid traffic in Google Ads."

The post names the feature the Invalid Traffic Overview Card. It also carries the only new statistic Google offered today, and that number is not about the card at all. It concerns leads.

Most of the operational detail arrived separately, in a three-post thread from Google's official Ads Liaison account on X. The thread says the card is "Rolling out globally starting this week," that its savings figure covers "your selected date range," and that a View details link opens daily and campaign-level data that can be downloaded.

In Short

Google is adding a card to the Google Ads Overview page that shows how much it says advertisers saved through invalid traffic filtering, next to the credits issued for invalid activity found after billing. That matters to every advertiser paying per click or per interaction, because until now this information sat in a click-count column, in billing adjustments and in a separate report. You get the figures by date range, by day and by campaign, with a global rollout starting this week, but Google has not said how the savings estimate is calculated or how much invalid traffic its systems see overall.

What Google said

The post opens with a statement of long-standing practice rather than a change in policy. "We've long built defenses directly into our ad systems to block invalid traffic, helping businesses get the most out of their budgets and spend with confidence," Sharma wrote. "Today, we're rolling out the Invalid Traffic Overview Card in Google Ads to give advertisers front-end transparency into how we protect their campaigns."

Billing is the core of the second paragraph. "We don't charge for invalid traffic, and our multi-layered defenses have always worked in the background to catch it before an advertiser is billed," according to the post. Activity that slips past those defenses is handled afterwards: "We issue a credit when it's detected later, where appropriate and possible."

That qualifier, "where appropriate and possible," does real work. The post does not define when a credit would be inappropriate or impossible, and nothing in it suggests the card will show activity that was flagged but not credited.

Then comes the claim that gives the feature its framing. "Now, for the first time, businesses can see both the amount saved and the credits we automatically provide for invalid activity in Google Ads," Sharma wrote.

The third paragraph moves from the card to enforcement. "As bad actors become more sophisticated, we're continually strengthening our approach to keep policy violators off our platforms," according to the post, which links the phrase "strengthening our approach" to Google's August 2025 post on using AI against invalid ad traffic. The final sentence reads: "So far in 2026, we've reduced advertiser exposure to invalid leads by an average of 30% through continuous system upgrades."

That is the whole blog announcement. It contains no total for invalid traffic filtered or credited, no share of clicks found to be invalid, no list of markets, no rollout schedule and no word on which campaign types are covered. The Ads Liaison thread answers the questions on markets and timing. The others remain open.

Where the card sits and what it shows

"Look for the card under the Campaigns tab on the Overview page in your account," according to the Ads Liaison thread. Google's help page on invalid traffic gives the same location and adds that the card is rolling out gradually and may not yet be available in every account. The "today" in Sharma's post therefore marks the start of a rollout, not universal availability.

The thread opens with Google's own case for the feature. "Google's defenses against invalid traffic (IVT) run quietly in the background to protect your budget," the first post reads. "But until now, there hasn't been a single place to understand the full impact of those protections on your campaigns." The new card, it continues, changes that by "showing both pre-invoice savings and post-invoice credits in a single, downloadable view." That phrasing, a single place rather than a first sight of the data, is narrower than the blog's "for the first time," and closer to what the record supports.

An image posted with the thread shows the card itself. It carries the heading Invalid traffic, a subheading reading "Your invalid traffic protection and credits," and an explanation: "You aren't charged for invalid traffic, like bots or accidental clicks. Automated protection filters it out before you're billed, or it's credited back to you if ongoing analysis detects it later, where appropriate and possible." Below sit two figures, Est. savings, captioned "Automated protection," and Credited cost, captioned "Past invalid activity," followed by a View details link. The amounts in the image, $13,531.08 and $1,653.90, are sample values; Google did not say whether they come from a real account, and they say nothing about typical results.

What the two figures count

The second post in the thread defines both numbers. Estimated savings is "What Google's automated filters and proactive reviews saved during your selected date range by catching things like bots, click farms, and accidental clicks before billing." The post adds: "This is when our defenses catch the vast majority of IVT, keeping it off your invoice entirely." Credited cost is "Credits automatically applied to your account for invalid activity detected after a monthly billing cycle has closed."

Two details in those definitions matter. First, accidental clicks sit in the same bucket as bots and click farms. The savings figure is therefore not a measure of fraud; it combines deliberate abuse with ordinary misclicks, and the card does not separate them. Second, "the vast majority" is the only indication of how invalid traffic splits between the pre-billing and post-billing stages. Google gave no percentage.

The third element is the breakdown. "Click 'View details' to analyze daily trends and campaign-level data, or download the reports to share with stakeholders," the thread says.

What the thread does not explain is valuation. The blog speaks of "the amount saved"; the card says Est. savings. Neither the blog, the thread nor the help page says how a filtered click is priced: at the cost it would have carried in the auction, at an average cost per click, or by some other method.

What advertisers could already see

Sharma's "for the first time" deserves a closer look, because much of the underlying data has been visible in Google Ads for some time, in scattered places.

The Invalid clicks and Invalid interactions columns in the campaign table count traffic filtered before billing, according to Google's help page. Those columns report volumes, not money.

Credits for activity detected after billing already appear as billing adjustments. According to the help page, they show up under the Billing menu, in the Summary view, inside an Adjustments dropdown in the Month card, labelled Invalid activity. They also appear on invoices and in account transactions.

Then there is the Invalid Activity Credit Report. PPC Land reported on June 1, 2026 that Google had documented the report for the first time. It sits in Report Editor under the Campaigns menu, in the Template gallery, as "Invalid Activity Credit Report: Search & PMax." The report adds three credit metrics (credited clicks, credited interactions and credited amount) and seven adjusted metrics, from adjusted cost to adjusted cost per conversion. It covers Search and Performance Max campaigns only; Display, Video and Demand Gen are outside its scope.

Advertisers can also ask Google to investigate specific clicks. The Click Quality Form accepts requests covering the previous 60 days, according to the help page.

So what is actually new? Two things, on the evidence available. The card turns pre-billing filtering into a money figure rather than a click count, and it puts that figure next to credited cost on the Overview page, the account summary view, rather than in billing or Report Editor. Credits themselves were already reported. A single place for both is the change.

Two stages of detection

The card reflects a two-stage system Google has described for years. In the first stage, invalid activity identified before an invoice is finalised is simply not charged. In the second, activity identified later generates a credit, which Google says it issues automatically in the next billing cycle "where appropriate and possible."

The second stage creates an accounting wrinkle. According to the help page, campaign metrics do not adjust for past activity credits. A click near the close of a billing cycle is attributed to the month in which it happened, while its credit lands in the following month, according to the credit report documentation. Campaign-level cost in the standard tables therefore reflects charges before credits. The Invalid Activity Credit Report exists to produce net figures, but only for Search and Performance Max.

The Ads Liaison thread ties Estimated savings to the selected date range. It does not say whether Credited cost within that range is counted by the date of the original click or the date the credit was issued, a distinction that matters when a credit lands a month after the click it refers to.

The 30% figure on invalid leads

The most specific number in today's post is the 30% average reduction in "advertiser exposure to invalid leads" in 2026. It raises more questions than it answers.

Google did not define an invalid lead in the post, did not give a baseline period, and did not say whether the 30% is an average across advertisers, across campaigns or across leads. "Exposure" is also undefined. It could mean fewer invalid form submissions reaching advertisers, fewer invalid leads counted as conversions, or something else. The post attributes the change to "continuous system upgrades" without naming any of them, and it does not connect the figure to the new card. The Ads Liaison thread does not repeat the number. It points instead to the latest issue of Google's Ads Decoded newsletter on LinkedIn, titled "New invalid traffic tools & lead quality tips," for material on reducing invalid leads in a CRM.

Leads have become a sore point for Google in 2026. In September, Google published a help page titled "Prevent invalid leads." PPC Land reported on September 20, 2026 that the page conceded Google's filters do not stop invalid leads on their own and that most traffic checks happen after an ad serves. The page recommended three technical safeguards for lead forms on advertiser sites: reCAPTCHA, double opt-in through a confirmation email, and server-side validation of email domains and phone formats before data reaches a CRM. The same article noted that lead goals such as qualified lead and converted lead were said to trigger invalid traffic protections, without either Google document explaining what those protections did.

The help page on invalid traffic repeats the limit plainly: Google's filters cannot block traffic arriving directly on an advertiser's own site. Google does say it removes conversions tied to invalid traffic from reporting.

Why does that matter? Because fake form fills do not only waste the cost of a click. When a bot completes a form and that submission is counted as a conversion, it feeds bidding systems. Smart Bidding learns from conversion data, so a polluted signal can steer future spend toward the same sources. In a Lunio survey of 131 senior marketers, 51.1% named automated bidding optimising toward non-human "ghost converters" as the top agentic AI risk, a perception measure rather than an incidence rate.

A 30% average reduction, if it holds across the advertiser base, would be meaningful. Without a baseline, though, it cannot be checked against anything an advertiser sees in an account.

What vendors say the rates look like

Google's card shows money. It does not show a rate, and Google has not published one for Google Ads as a whole. The figures that circulate in the market come from detection vendors, who measure different things in different ways and sell products built on the problem they measure.

Lunio's 2026 Global Invalid Traffic Report, released in January 2026, classified 8.51% of 2.7 billion paid clicks analysed between August 1, 2024 and July 31, 2025 as invalid. Its platform table put Google Ads at 7.57% and Google Search at 5.21%, below TikTok at 24.2% and LinkedIn at 19.88%. Those are Lunio's classifications of the clicks it analysed, not an audited count.

Sector reports from the same vendor point in a similar direction for Google's newer campaign types. Lunio found that invalid traffic on retail search campaigns using AI Max rose from 2.46% in the fourth quarter of 2025 to 5.28% in the second quarter of 2026, while standard search fell from 3.72% to 3.07% over the same period. Google Shopping went from 4.16% to 7.51%. In banking, Google Search carried an average invalid traffic rate of 5.51%, across 86 million clicks analysed from July 2025 to March 2026. In education, competitor-targeting campaigns showed a click-weighted rate of 9.3%, against 4.7% for generic campaigns.

Not every critic thinks click-based measurement goes far enough. Augustine Fou, who sells the FouAnalytics verification product, argued in October 2026 that click-based estimates such as Lunio's undercount fraud on CPM-bought ads, where impressions rather than clicks are the billing unit.

None of these figures can be laid against Google's card. A vendor's rate is the share of traffic its own model flags; Google's Estimated savings is the value of traffic Google's systems flag, expressed in currency, with no denominator. The two may describe overlapping traffic, or they may not. Google's announcements do not attempt a reconciliation, and nothing in the blog post or the thread suggests the card will show a percentage.

The same Lunio survey found that only 5.3% of the 131 marketers surveyed ran a dedicated invalid traffic platform, and 39.7% said the belief that platforms already handle prevention well enough was the main barrier to adopting one. A Google-branded savings figure on the Overview page speaks directly to that belief. It reports Google's own judgement of which traffic was invalid, from the company that also sells the clicks.

Google's record on invalid traffic disclosure

Today's card follows a series of disclosures from Google's ad safety organisation, several of them under Sharma's name.

In August 2025, Google said that new detection applications built with large language models had cut invalid traffic from deceptive or disruptive ad serving by 40%. The figure, from a pilot, related to mobile ads that registered impressions without being visible and to disruptive pop-ups. Google's Ad Traffic Quality team worked with Google Research and Google DeepMind on the effort. That is the post to which today's announcement links.

In November 2025, Sharma said Google had reduced incorrect advertiser account suspensions by more than 80%, with appeals resolved 70% faster. In April 2026, Google's 2025 Ads Safety Report said Gemini-based systems helped block or remove more than 8.3 billion ads, while advertiser account suspensions fell to 24.9 million from 39.2 million in 2024.

On October 20, 2026, Sharma announced that Google would require financial advertisers in the United States and Canada to complete Financial Services Verification in early 2027. That post appears among the related stories under today's announcement.

The sources of invalid traffic have also been in the news. Google banned Videostrong from building certified Android devices in November 2026 after its Threat Intelligence Group tied the company to the XCPE backdoor, which placed more than 1.3 million set-top boxes into a botnet. Ad fraud modules on those devices included one that fired impression beacons for ads that were never shown.

Why it matters for advertisers and agencies

The practical effect of the card depends on details Google has not published. A savings figure with no published valuation method is a statement of Google's own accounting, presented in Google's interface. It is not an independent measure of invalid traffic, and Google does not claim that it is.

Still, the change has weight. The money Google's filters keep off an invoice was previously visible only as a click count, in a column that must be added to the campaign table by hand. Putting a currency figure for it on the Overview page, next to credits, with daily and campaign-level detail and a download option, makes both part of the routine view of an account and of the reports agencies send to clients.

It also gives advertisers a number to question. If the Estimated savings figure looks small next to a vendor's click fraud estimate, or large next to an account's total spend, the gap becomes visible in a way it was not before. The card does not explain such gaps, and because accidental clicks are included, a large savings figure does not by itself indicate a large fraud problem. Google has not said whether the figures will be available through the Google Ads API, or whether the card covers Display, Video and Demand Gen, where the credit report does not reach.

How quickly the rollout completes, and whether Google publishes the method behind the estimate, will determine whether the card becomes a reference point or simply another number on the dashboard.

Timeline

Summary

Who: Google, through Keerat Sharma, VP & General Manager, Ads Privacy and Safety, on the Ads & Commerce Blog, with further detail from Google's Ads Liaison account on X. The change affects every Google Ads advertiser whose account receives the card, and agencies reporting spend to clients.

What: Google began rolling out the Invalid Traffic Overview Card, which shows the estimated amount saved through invalid traffic filtering before billing, including bots, click farms and accidental clicks, and the cost credited for invalid activity detected after a monthly billing cycle closes, with daily and campaign-level data that can be downloaded. Google also said advertiser exposure to invalid leads fell an average of 30% in 2026, without giving a baseline or method.

When: Today, with the rollout continuing globally from this week, according to the Ads Liaison thread. Google's help page says the card may not yet appear in every account.

Where: Globally, on the Overview page under the Campaigns tab in Google Ads, labelled Invalid traffic. Google named no campaign-type limits.

Why: Google framed the card as "front-end transparency into how we protect their campaigns." It arrives after a year in which vendors published rising invalid traffic rates for some Google campaign types, and Google itself acknowledged that its filters do not stop invalid leads on their own. The card reports Google's own classification of invalid traffic and does not include a rate, a split between fraud and accidental clicks, or a published method for the savings estimate.