Google Ads click fraud is killing any confidence I had in this platform

[i4mt3hwin]

Dunno – I always combat this by steering the algo with micro conv. I always see bot clicks vanish after a week or two when I do that.

You say one of the rare times – figure out how to make those non bots leads and next thing you know you’ll be getting all people. Some of those people will book and then eventually drop the micro conv to secondary. 

[Fit-Classic-9295]

This is one of my biggest hesitations. I don’t have a budget for bots 😂

[TrumpisaRussianCuck]

Unfortunately if you’re not using a smart bidding strategy you have these issues and just have to continually layer on more and more exclusions.

[kubrador]

google’s invalid click rate is like asking a restaurant how clean their kitchen is – shocking nobody reports it at a solid 95% when they’re the ones grading themselves. the real play is getting aggressive with account structure and conversion tracking so you can actually see what’s converting, then just accept google’s gonna tax you for the privilege of finding out.

[BlueGridMedia]

Problem is that when you run manual CPC with tight keywords on a new campaign you are telling Google to give you cheap clicks, doesn’t matter who (It works for some campaigns, but I think its just outdated now). So Google sends those cheap clicks your way, even on search only, theres still a ton of garbage traffic. The algo won’t protect you until it has conversion signals, so Google’s solution is “use automation so we can filter those out.” It sucks, tell me. Unfortunately, that’s just reality.

First make sure that you have conversion tracking for calls, forms, call extensions, and some micro like time on site, scroll 50%, etc. If you don’t have 3-5 signals you are basically flying blind. Good news is that bots don’t trigger these things, humans do.

Second thing, I’m beginning to think that in 2026 starting out with manual cpc or max clicks is a bit outdated. I’m trying now to use maximize conversions with a high CPA, even with no data. I do this because Google immediately activates traffic quality filters, fraud models, user history, etc.

I tested a new campaign with maximize conversions, then switched to manual cpc, and guess what? I only got trash with manual cpc and maximize clicks. So I’m sticking with maximize conversions which sent me a couple leads very early on, and those were qualified leads, I’m doing this even for new campaigns. Less junk comes my way. You are basically buying access to filtering, it sucks, but it works, so I would suggest trying this out.

[Free-Way-9220]

Contact google and you can enjoy being gaslight about it

[ManagedNerds]

I block the most egregious ad fraud with cloudflare and then also implement conversion tracking that pushes over conversion value. From there, optimize around conversions and go.

[namalleh]

If that’s the case and you’re tracking conversions – and you honestly want to reduce this

I have a solution I’m working on I can give you early access to

[Electrical-Youth6817]

Try that in garage door nieche this past month over 12k spent 660 per conversion it’s absolutely ridiculous

[stovetopmuse]

You are not crazy, this is a super common pattern lately, especially on hyper local lead gen. I see the same thing where the campaign looks clean on paper but session replays are just dead clicks with no intent signals. The hardest part is that even when you tighten geo, keywords, and placements, the junk still leaks through and Google’s invalid click numbers never line up with what you actually observe. At this point I treat reported CPC and conversion rate as secondary and spend more time looking at behavior level data to decide if a campaign is even worth scaling. Curious if you saw any difference when layering in stricter hour scheduling or aggressive exclusions, or if it was noisy no matter what.

[QuantumWolf99]

Click fraud paranoia usually masks targeting problems… for all the roofing clients I have worked with, analyzed thousands of Clarity sessions and “bot behavior” you’re seeing is actually mobile users on 4G where page load takes 8 seconds so they bounce before anything renders… industry data shows average invalid click rate sits at 11.5% across Google Ads with home services seeing legitimate rates between 8-15%… if you’re claiming 90% invalid traffic your conversion tracking is broken or you’re bidding on informational keywords like “roof repair cost” attracting researchers not buyers.

I’ve never seen legitimate roofing campaigns at scale fail to convert unless landing page has no phone number above fold or you’re targeting 50-mile radius… Google’s auto-exclusion is actually aggressive because refunding advertisers directly costs them revenue, they filtered $84B in ad fraud in 2023 alone according to latest stats.

[BottingWorks]

What’s the daily budget and country?

How many keywords are you using and what’s the match type?

What does the organic traffic convert at, % wise?

[Electrical-Arm-3869]

Yeah, I would also say manual CPC is killing your performance. We’re basically telling Google, “I don’t want the good clicks, I just want the cheap ones.” What has worked for me continuously despite all the AI? Focusing on exact match keywords. Just identifying the most important 30-50 keywords from my niche or the niche of my clients. And that has worked in just delivering good and steady results. Clicks are expensive, especially for some niches. That’s just how it is.

[ppcwithyrv]

Click fraud is a way of life now with PPC. Focus on conversions—-> even mid level conversions like landing page view or time spent on siite

[ernosem]

Unpopular opinion:
– Most of the ‘click fraud’ is due to bad settings, 90% of the time when someone is complaining about click fraud it’s not actually click fraud, most of it could have been avoided with proper setup, eg: excluding the search partner network or adding negative keywords.

– Yes, there is click fraud with Google, I have seen it in my own eyes multiple times, however it’s still lower compared to alternatives, like Facebook or Adroll etc.

Back to Roofing:
Roofing is a hard market, we have several roofing clients in the US and it’s not just about the settings in Google Ads. This is not the time when people jump on in to replace the roof, so you need to be consistent for another 2-3 months.

Also the offer is very important on the landing page, the reviews and general branding of the company.

We have seen 3X difference in CPA between a 17 years and a 1 year old brand (window replacement market, roughly the same are), so if your client is a new brand that also can be an ‘issue’, just there is not enough trust. (Before you ask, the 3X difference was measured for Non-Brand terms only not for blended account performance)

I hope this helped.