Singapore

Google Ads Invalid Traffic Card: How to Read It Without Overreading It

As of November 2026, the newest number on your Google Ads Overview page is an estimate of money Google says it stopped you from being charged, not a refund and not a verdict on lead quality. Advertisers working with an sem agency will see it appear under the Campaigns tab, and the practical question is narrow: should this figure change a decision, or only trigger a check?

Short answer: it triggers a check. Pull the campaign-level breakdown, run the adjusted-cost report before you quote a closed month's CPA, and keep lead-quality judgement on your CRM data. Everything below explains why that order holds and what each figure actually measures.

What the Invalid Traffic Overview Card reports

The card covers three reporting areas: estimated savings, credited cost, and daily and campaign-level reporting. Each answers a different question, and mixing them up is where most of the confusion starts.

Estimated savings is the advertising cost Google's systems avoided charging because they identified and filtered invalid activity before the invoice was raised. The bulk of the filtering happens at this stage, so this is normally the larger of the two money figures.

Credited cost is different. It is the amount returned after the fact, for activity detected in a billing period that had already closed. That figure arrives later, and often in a month different from the one it belongs to.

The third area is the one with the most working value. Clicking through to details opens a daily trend and a campaign table, and the data exports as CSV. Google also points you to the monthly invoices where credits were applied.

A marketer at a desk opens a daily trend chart beside a stack of monthly invoices.

Where to find the card

Open the Campaigns tab and look at the Overview page. The card sits there alongside the usual spend and conversion blocks, so it is easy to treat as another headline metric. Treat it instead as an entry point: the headline number tells you that filtering happened, and the campaign table tells you where.

The card's figures are also mirrored in Google's enterprise search ads platform for Google Ads accounts. If you report through a third-party dashboard that reads the Campaigns table, the card's numbers will not appear there automatically.

Why post-billing credits do not change Cost or CPA

Google Ads locks standard campaign metrics at month-end so the reports match the billing records. When more invalid traffic surfaces after that lock, the credits are recorded, but the Cost and CPA in the Campaigns table stay as they were.

The practical consequence is a mismatch between two views of the same month. Your Cost column shows the gross figure; the card shows what was given back. A CPA computed on gross cost is slightly overstated, and the size of the overstatement depends on how much was credited after the lock. An sem agency singapore team that reports on the raw table will quote that overstated figure to clients.

Google's answer is the Invalid Activity Credit: Search & PMax report, available in Report editor. It carries 10 columns, including Adjusted cost and Adjusted cost / conv, so you can review Search and Performance Max performance with credits factored in. Credits are attributed to the dates the ads originally ran, even when Google issued them later.

To open it: Insights and reports, then Report editor, then Template gallery, then Performance Summary, then Invalid Activity Credit: Search & PMax.

A short check before you quote a closed month

The order matters more than the effort. When a month closes and someone asks for the final CPA:

Export the Campaigns table as it stands. That is the billing view.
Run the Invalid Activity Credit: Search & PMax report for the same date range. Compare Adjusted cost with Cost.
Divide by conversions as reported. If the two CPAs differ by a wide margin, quote the adjusted one and say which report it came from.

This takes a few minutes and removes a recurring argument about which number is "right". Both are right; they answer different questions. The Campaigns table answers what you were billed, and the adjusted report answers what that billing bought after credits.

Two colleagues compare two printed monthly performance reports at a table, a calculator between them.

How to read estimated savings campaign by campaign

The daily and campaign-level breakdown exists because one account-level figure hides the campaigns with the most filtering. A campaign with a large budget can show large estimated savings for no better reason than volume.

Compare ratios, not amounts. Estimated savings divided by spend gives a rough share of filtered activity for each campaign, and that share is what is comparable across campaigns of different sizes. This is the part a good sem agency review spends its time on. A campaign that spends heavily and shows a small share is not necessarily cleaner than a small campaign showing a larger share, but the ratio is the fairer starting point than the dollar figure.

The daily trend answers a second question: whether filtering is steady or spiky. A sudden jump on one date is worth investigating, since it may point to a new traffic source, a new placement set, or a change in targeting that opened a door you did not intend.

Download the CSV before making any of these comparisons. The card's on-screen table is a view, not a record, and screen numbers change as the month fills in.

What the figures cannot tell you

Three limits are worth stating plainly.

First, the card reports activity Google's systems identified as invalid. It does not state how much invalid traffic went undetected. A low figure is not proof of clean traffic.

Second, for lead-generation accounts, invalid traffic reporting does not explain why fake or poor-quality leads keep arriving. Google's press liaison, Ginny Marvin, made the point that bad leads can come from legitimate clicks or submissions that never originated from an ad at all. An account can show little invalid traffic and still struggle with lead quality.

Third, the dashboard stops at the click. Platform-native reports end at their own data boundary, so what happens on the landing page, in the form, and in the CRM afterwards never enters the platform's numbers. A savings figure cannot explain a leak that happens three steps downstream.

A screen shows a click trail fading out, while a nearby inbox logs new enquiries.

Reconciling the card with CRM data

The only way to settle whether filtered traffic is the cause of a lead-quality problem is to join the two data sets. Pull the daily invalid-traffic export, join it to CRM records by date and by campaign, and look at what the leads from high-filtering days did afterwards.

If the leads from high-filtering days were substantively worse, filtering is a partial explanation. If they were the same as leads from clean days, traffic quality is not the problem, and effort belongs on the landing page or the follow-up. This is a small piece of analysis, but it is the only one that answers the question the card raises.

Two hygiene points make that join usable, and both belong in the scope of any sem agency audit before a single conclusion is drawn. Keep the CRM's date field consistent with the account's time zone, since a day boundary mismatch silently duplicates or drops records. And reconcile clicked traffic against sessions actually recorded on the site; when the two do not match, the card is describing clicks that never produced a visit.

Where Topkee sees the card fitting

An invalid traffic card reports on filtered interactions, and that is the only thing it reports on. Campaign-level reporting of this kind becomes actionable only when it is tied to downstream inquiry and deal stages, because a platform-side figure cannot see past its own report. Topkee's Campaign Intelligence Report puts paid and organic traffic on one attribution ruler and follows a six-layer funnel from impression through landing-page interaction, inquiry and qualified lead to quote and signed deal, with a fixed ROAS denominator measured as signed-deal value divided by ad spend.

The complement on the advertiser side is naming the waste the platform does not itemise. Topkee's Spend Shield service points at waste from blank search terms and fatigued creatives cell by cell, drawing its signals from the client's own CDP, pixel and CRM data rather than platform dashboards alone. Delivery is a numbered, traceable action list rather than a set of recommendations, and its status stays visible in the client account. Access runs on a short read-only authorization, revocable at any time, with the report delivered in a few business days.

That matters for how a team in Singapore handles this card. A read-only grant lets an outside reviewer look at the account's traffic-quality data without any ability to change ads, budgets or payment settings. The waste audit ships with every report at no extra charge, and it is not an add-on. What the card reports and what an audit reports are different measurements, held side by side rather than converted into one another.

Recommended next step

Do the reconciliation once before you draw any conclusion. Export the daily invalid-traffic data, run the Invalid Activity Credit: Search & PMax report for the most recent closed month, join both against your CRM by date, and write down the gap. If you are reading this as a business running search in the region, the same discipline applies whether you manage the account in-house or through a sem agency singapore team: the report definitions matter more than the headline figure.

For the reporting vocabulary behind these numbers, see the Google Ads Channel Reports 2027 brief; for how filtered clicks and scaled content abuse sit together in search supply, the SEM brief on scaled content abuse covers the supply side.

FAQ

Does the credited cost figure reduce what I owe for this month?

No. Credits apply to a billing period that has already closed, so they appear as a return against that period rather than a deduction on the current invoice. The card and the monthly invoice are the two places to confirm which period a credit belongs to.

Why is the Cost column unchanged when the card shows credits?

Google Ads locks standard campaign metrics at month-end to match billing records. Credits issued after that lock do not rewrite the locked figures, which is why the Invalid Activity Credit: Search & PMax report in Report editor exists.

Is a high estimated savings figure a sign of poor traffic quality?

Not on its own. A campaign with more spend generates more traffic, so its estimated savings are naturally higher. Compare estimated savings against spend campaign by campaign rather than comparing dollar amounts between campaigns.

Can the card tell me why my leads are poor quality?

No. It reports invalid interactions Google identified, and it does not say how much went undetected. Bad leads can come from legitimate clicks or submissions that never originated from an ad, so the comparison has to be made against CRM data.

Do I need admin access to review this with an agency?

No. A view-only grant on the Google Ads account is enough for someone outside your team to read the card and the reports, and it can be withdrawn whenever you choose.