How to Read Your Microsoft Ads Invalid Click Report

Your Microsoft Ads invalid click report is the closest thing Microsoft gives you to a fraud readout, and most advertisers never open it. The numbers are there. They just live under a different name than you would expect, in columns you have to add yourself, and they only tell part of the story.

This guide covers where to find the report, what each column means, how to read the numbers, and what to do when they do not match what your own site is seeing.

Microsoft Calls It Low-Quality, Not Invalid

Microsoft sorts every ad click into one of three groups: standard-quality clicks, low-quality clicks and invalid clicks. Standard-quality clicks are the ones it believes came from potential customers. Low-quality and invalid clicks are the ones it associates with spiders, robots, questionable sources and test servers.

The rule that matters for your bill is simple. According to Microsoft, you are only billed for standard-quality clicks. If a click is billed and later found to be low-quality or invalid, the bill is adjusted, and suspected invalid clicks from robots, automated tools or other fraudulent means are credited automatically.

So when you go looking for invalid click numbers, look for the word low-quality. That is the label on the report columns.

Where to Find the Report

The low-quality metrics are not on by default. You add them to a performance report:

• Open Reports and start a performance report, such as an account or campaign report.
• Open the column chooser for the report.
• Add Low-quality clicks and Low-quality click rate (%), plus any of the related columns below.
• Pick a date range long enough to show a trend. A single week tells you very little.
• Save the report and schedule it, so it lands in your inbox without you remembering to run it.

What Each Column Means

• Low-quality clicks. Clicks that Microsoft judged unlikely to have commercial intent. These were filtered and are not what you pay for.
• Low-quality click rate (%). Low-quality clicks as a share of all clicks. This is the number to watch over time.
• Low-quality impressions. Impressions that came from low-quality searches.
• Low-quality conversions. Conversions recorded against low-quality clicks. If this is not close to zero, look at how your conversions are tracked.
• Low-quality general clicks. Clicks caught by general methods, such as deny lists and activity-based detection.
• Low-quality sophisticated clicks. Invalid clicks that used more sophisticated means to look valid.

The general and sophisticated split is useful. General clicks are the easy catches, the known bots and bad lists. A rising share of sophisticated clicks means someone is putting real effort into looking human, and that is the traffic most likely to slip past filtering entirely.

How to Read the Numbers

Watch the rate, not the count

A higher low-quality click count on a busier week means nothing on its own. The rate is what shows a change in the mix. A campaign that sits at a steady rate for months and then jumps is telling you something new is hitting it.

A high rate is not a disaster by itself

A high low-quality rate means Microsoft is catching a lot, and those clicks are not on your bill. The worry is not the clicks it filtered. It is the ones it did not, which never appear in this report at all.

Compare campaigns and networks

If one campaign carries most of the low-quality clicks, check its network setting and its placements. Run the Website URL (publisher) report for the same period. When junk concentrates on a handful of partner sites, excluding those sites often does more than anything else.

What the Report Cannot Tell You

The report counts what Microsoft caught. It does not list which clicks they were, what IP addresses they came from, or which of your billed clicks it missed. And because you are only billed for standard-quality clicks, anything in the billed column is by definition traffic Microsoft believed was fine.

That is the gap. The clicks costing you money are the billed ones that still never become customers. To see those, you need your own view from the landing page side.

Check Billed Clicks Against Your Own Site

The simplest test takes ten minutes:

• Take billed clicks for one campaign and one day from Microsoft.
• Count visits that landed on your site with an msclkid for the same campaign and day.
• Expect some gap, since people close pages before they load. A small gap is normal.
• A large, persistent gap, or visits that arrive and do nothing, is worth a closer look.

If the visits that do arrive show the same few addresses again and again, stay only a second or two, or never run JavaScript, you have found traffic that Microsoft billed and your site rejects. That is the traffic to exclude.

When to Contact Microsoft

If you believe you were charged for invalid clicks, Microsoft asks you to contact its support team. Go in with specifics: the campaign, the dates, the billed click count, what your site recorded for the same clicks, and any repeat addresses you found. A request built on your own records gets further than one built on a hunch.

Credits are Microsoft's decision under its own policy. Plan as if the money is gone and focus on stopping the source from coming back.

Turning the Report Into Action

• Schedule the low-quality columns weekly and watch the rate.
• Check the publisher report whenever the rate moves, and exclude partner sites that send junk.
• Compare billed clicks with landings on your own site.
• Exclude repeat bad addresses within Microsoft's 100-entry limit, ranked by damage. The workflow is in our guide to Microsoft Ads click fraud.

Block Click Fraud does the site-side half for you. It checks every Microsoft Ads click that lands on your pages, marks it clean, suspect or fraud with the reason, and builds the paste-ready exclusion list. For how the two platforms differ, see Microsoft Ads vs Google Ads click fraud.