Singapore

Google Ads Invalid Traffic Card: What the Credit Misses

As of November 2026, the Invalid Traffic card on the Google Ads Overview page shows two figures: estimated savings and credited cost. Both concern money Google says you should not have been charged for invalid activity, and both appear in your account now. The question worth asking before you forward the screenshot to your finance lead is narrower. If this card shows only the money Google noticed and returned, who is looking at the budget that went to real clicks that never had a decent chance of converting? An sem agency should be able to answer that with a report, not a reassurance.

Key takeaways

Google announced the Invalid Traffic Overview Card in November 2026 to show advertisers the amount saved and the credits issued for invalid activity.
Google says it has reduced advertiser exposure to invalid leads by an average of 30% so far in 2026.
Estimated savings and credited cost measure different moments: filtering before billing and crediting after it.
An investigation request covers activity within the last 60 days, and it needs your 10-digit customer ID.
Invalid conversions can be removed later, and a campaign that bought mostly invalid traffic will still report its own metrics.

What the Invalid Traffic card actually shows

Search Engine Roundtable reported the rollout on 2026-11-12, quoting Keerat Sharma of Google on a card that gives advertisers front-end transparency into how campaigns are protected. Advertisers who want that transparency explained line by line usually start with an sem google ads partner. Google's own announcement describes the purpose in similar terms: for the first time, businesses can see both the amount saved and the credits Google provides automatically for invalid activity.

The two numbers sit against a rule Google has held for years. It does not charge for invalid traffic, and its filters work before the advertiser is billed. When invalid activity is found later, Google issues a credit where appropriate and possible. The card is a window onto that process, and the help documentation explains what each half measures.

Estimated savings is the estimated cost saved by automated protections from invalid traffic, which include automated filters and proactive manual reviews. Credited cost is the amount credited back to you for past invalid activity that was detected after the billing period ended. On the card, View details opens a page with daily graphs, an all-campaign breakdown and links to past invoices. Google notes the card is rolling out gradually, so your account may not have it yet.

The distinction matters because the two figures answer different questions. Estimated savings tells you what the filters caught before the invoice. Credited cost tells you what was caught late and returned as an adjustment. A high first number and a low second one is the expected shape when filters work, and it is not a sign that anything is wrong.

A marketer studies a monitor where filtered clicks and a late credit appear side by side above an invoice.

Why the two numbers differ from your invalid clicks column

Advertisers who add the Invalid clicks column to their campaign data table often find it far larger than any credit on the invoice. Google's help page explains the gap directly, and the reason is worth knowing before anyone treats the difference as an accounting error.

The Invalid clicks column shows traffic Google filtered before you were billed. Those clicks never reached your invoice. Credits apply to invalid activity detected after the billing cycle closed, plus policy violations that affect your placements. To see the count of credited clicks, you use the Credited clicks column in the Invalid Activity Credit Report in Report Editor.

There is a second reason the numbers refuse to line up. Google freezes campaign metrics at the end of the month so they match your invoice, which means a later credit appears on the invoice but does not rewrite the campaign data. The Invalid Activity Credit Report is the view where those credits are applied, showing adjusted clicks, cost and conversion rate.

A rising invalid clicks figure is not a warning about your account. Google reads it the opposite way: the filters caught and removed more invalid interactions from your metrics and your bill. The number is a record of protection, and it moves with how much invalid traffic exists at a given time.

How to ask for an investigation within the 60-day window

Google's help page sets a hard limit that surprises people. If you suspect invalid traffic that went unfiltered, you can submit an investigation request for activity within the last 60 days. Anything older is outside the process.

Google lists what the request needs. The customer ID is your 10-digit Google Ads account number. The form also asks for the date range, the campaign and ad group names where the spike appeared, and the suspicious keywords. Add web server logs for the activity, with the IP addresses and devices behind it, plus the GCLIDs that let Google map a visit back to an ad interaction. For invalid leads, a spreadsheet of GCLIDs with your assessment of each lead helps.

The request goes through the Click Quality Form. An sem agency that keeps GCLIDs and server logs on file can put this pack together quickly. Google's specialists review the data manually and reply by email when the investigation ends. If they find additional invalid traffic the automated filters missed, they notify you of any credits.

One boundary is stated plainly. An investigation cannot confirm whether specific clicks were marked valid or invalid, because that would expose how detection works. A third-party tool that flags duplicate IP addresses or bot activity may be pointing at traffic Google already removed from your bill, which is part of why such reports never match the platform.

An analyst assembles server logs, a GCLID spreadsheet and a calendar with a highlighted 60-day window.

The part the card leaves out, after the click

The card and the investigation both stop at the ad interaction. Google's help page is explicit about where its systems lose sight: because you capture leads on your site and in your CRM, Google cannot track or control what a visitor does after clicking. Invalid traffic filters run after an ad serves, so they cannot block traffic from reaching your site.

That leaves a second kind of waste in the funnel, and the card has no figure for it. A click can be perfectly valid by every detection standard and still never produce a lead, because the landing page asked for too much, loaded slowly on a phone, or addressed the wrong intent. None of that is invalid traffic, and none of it shows in estimated savings or credited cost.

Google's own remedy list points in this direction while staying inside its boundaries. It suggests choosing Leads as the campaign objective, securing forms with reCAPTCHA and server-side validation, considering a double opt-in, and connecting first-party data through the Google tag and enhanced conversions. All of these are advertiser-side duties. The platform filters what it can see; the rest needs data from your own stack.

For a marketing manager, that is the line to draw when briefing an sem google ads team for the next conversation. The Invalid Traffic card answers "how much did Google catch". It does not answer "how much of the spend that passed every filter ended in a conversation with a buyer". The second question needs your CRM, your landing page analytics and your ad account side by side, on the same date range.

A worked comparison makes the gap concrete. Google's help page notes that invalid conversions can be removed if they originate from invalid traffic, or if they remain invalid for longer than 13 months, which may show as a small drop in conversions over time. A report built only on the platform's conversion count will therefore shift under the reader's feet, without any explanation of which layer the loss came from.

What you want to know	Where the answer sits	Who can give it
How much did Google filter before billing	Estimated savings on the Invalid Traffic card	Google Ads
How much came back as a credit	Credited cost, and the Invalid Activity Credit Report	Google Ads
How many clicks Google removed	Invalid clicks column in campaign reports	Google Ads
Where the spend went after a valid click	Landing page and CRM data, on the same date range	You or your agency
Whether a lead was a real buyer	CRM stages, follow-up records	You or your agency

The bottom two rows are the ones this card does not fill, and they carry the budget that a filtered-click count never touches.

What Topkee measures that the card cannot

Topkee's position on this is specific, and it starts by admitting the boundary. The knowledge base records no Topkee product that reports, filters or reconciles Google Ads invalid traffic. That layer belongs to Google Ads, and any agency claiming a front-end invalid-traffic dashboard is describing a Google report with a new wrapper.

What Topkee runs instead is Spend Shield, an ad spend waste audit that names wasted budget cell by cell. It looks for blank or irrelevant search terms, fatigued creatives, and the creative and audience settings that quietly drain a campaign. The signals come from the client's own CDP, pixel and CRM data rather than from a platform's self-reported dashboard.

The audit ships inside every report at no extra charge, and the client-facing document is the Campaign Intelligence Report. It is issued every period and puts paid and organic traffic on one attribution ruler, with a waste view and a numbered action list, which is the kind of document you can hold an sem agency to. Access is read-only, so the service cannot change ads, budgets or payment settings, and the advertiser owns the report and the data.

The reason for this design is that waste survives because nobody names it. Google's card names what its filters caught. It does not name the search terms that never had a chance, the creative that stopped working months ago, or the audience setting that sends mobile visitors to a page built for desktop. The report put a number on that layer for one account and ranked the actions by priority. Read that as one account in one month, not as a benchmark for yours.

Two colleagues compare a search ads dashboard, landing page behaviour and CRM records across three laptops.

Which companies this advice does not suit

Most of this article assumes you run paid search with a budget large enough to make an audit worth the effort. Three kinds of business should read it differently.

A business running a very small Google Ads budget does not need a separate waste audit. The Invalid Traffic card and the Invalid clicks column are enough to catch the money that leaks before billing, and adding a reporting layer costs proportionally more than it recovers.

A business that has not connected its CRM to its ad data cannot use the second half of this article. Spend Shield and its report both draw on first-party customer data, so a firm with no CRM, no pixel and no form tracking has nothing for the audit to name.

A business whose leads close offline over many months, such as a property developer or a specialist clinic, needs a longer measurement window than a monthly report. The card and the investigation are still useful to it, but a report judged on a single period will look worse than the business actually is.

Recommended next action

Open the Overview page in Google Ads and find the Invalid Traffic card, then note both figures for the same date range you use in your own reporting. Write the estimated savings and the credited cost into the next monthly review, labelled by the month they belong to. If the card is not in your account yet, it is rolling out gradually, so check again in a few weeks rather than assume it does not apply to you.

Before that meeting, read the Google Ads channel reports brief, which explains how campaign reports group spend so you can align the card with your own figures. The Google Ads API v26.2 note then covers what an account can now query at customer level. If you want the post-click layer included, the Spend Shield audit is where third-party judgment adds to what Google reports.

FAQ

Do I need to do anything when the Invalid Traffic card appears?

No action is required. The card shows what automated protections already saved and what was already credited. If both figures look low, that usually means the filters caught the traffic before it reached your invoice, which is the intended outcome.

Why is my invalid clicks number so much larger than my credit?

The two measure different moments. The Invalid clicks column counts traffic Google removed before you were billed, so it never appears on an invoice. Credits cover invalid activity found after the billing cycle closed. Google states these numbers are not directly related.

Can I ask Google to investigate clicks from last year?

Only activity within the last 60 days can be submitted for an investigation, and the request needs the 10-digit customer ID, a date range, campaign and ad group names, and server logs with IP addresses and GCLIDs. The form for it is the Click Quality Form.

Does the card tell me my leads were fake?

Not directly. Estimated savings covers the ad interaction, and Google cannot track what happens on your site or in your CRM. Invalid lead forms are possible even when the click was valid, which is why Google suggests reCAPTCHA, server-side validation and double opt-in on your own forms.

Why did my conversions drop with no campaign change?

Some invalid conversions are removed if they come from invalid traffic or remain invalid for longer than 13 months, which can show as a small drop in the conversion count over time. Check the Invalid Activity Credit Report and your CRM before treating the drop as a performance problem.

What should I ask an agency about this card?

Ask which layer they report on, and request one report that shows the spend that passed every filter and still produced no lead. If the answer stops at the platform's card, the post-click loss in your funnel is not being measured by anyone.