Is Click Fraud Ruining PPC Ads?

[Peter (IMC)]

]

With reports that over 50% of PPC ad revenue comes from fraudulent clicks, it appears that the Google/Overture cash cow may be experiencing some growing pains.

When I read that I could not believe it. 50%? So I read the article and it said this:

According to research released by Web analytics provider Clicklab, fraudulent clicks can account for more than 50 percent of all advertising fees attributable to certain categories.

That´s not 50%,.. that´s 50 % of certain categories. Thus the majority of categories are fine. Just a few categories that are really terrorized by click fraud.

Please do not represent the reality in a way that would scare many people away for the wrong reasons.

This sounds more like less than 0.1 % of all clicks are fraudulent, and even that is probably a high estimate.

However, some categories may be extremely competitive which would force some advertisers into less than honorable practices.

The idea that clickfraud would be generated by owners of adsense accounts doesn't seem very logical to me. It would simply cost too much and fraud is much easier to detect. When you do arrange for people to click on your adsense ads, you would have to make sure that all kinds of behavior patterns would be consistently random/natural. The chance for Google to pick up these kind of fraudulent clicks is easy to detect.

But, of course there are those that are able to fool the detection systems. However, If click fraud would be such a huge problem as it is made to believed, then hardly anybody would be making any money from PPC.

Lets see what you can do if you suspect click fraude:

1) Lower your Max bid! That position 1 isn't half as effective as position 3 or 4.

2) Increase CTR

3) Learn which ads work best. Never stop learning. Always try to create even better ads.

If you can manage to improve your ad performance, your positions will go up again, which means you can lower your max bid even more. Your fraudulent competitor will see his costs rise and rise while your costs go down. His fraudulent clicks will then become very helpful to you.

In the end PPC is not about how much you pay per click, but about how much you make per click.

[fundraiser]

I've been using ah-ha.com (Enhance Interactive) for over a year. Similar to experiences I've read here - I noticed an 80% abandon rate after the first click during time periods when I advertised. When I turned the ads off - it when back under 50%. EVERY TIME I reloaded funds or turned the keywords on I got a ton of hits that went no where. To test- I tried 6:00 am EST starts and got over 40 clicks immediately on the site. There were engines that were clicking once a day early in the day while funds were available and not again till the next day- I assume to stay under the radar of enhances fraud spotters.

I notified them of my suspicions and they dd nothing. I pursued and got a manager who asked for data. I provided 2 weeks worth of intensive backup which included clicks from engines where I could find my site listed under ANY keyword. I heard nothing back.

I dropped PPC altogether and my although my traffic is half of what it was- my business is better than ever since I took the funds and placed ads where I know customers will be.

I am still fuming at the fact that I was asked to produce evidence and spent such time in good faith doing so for no reason.


I agree with the author and believe the estimate may be low.

[pencom]

We are a small company and were using Ad words until we realized that our competitors were simply clicking our ads until they were exhausted. We contacted Google, provided the web log and they never got back to us.

We can't afford a huge advertising budget and thought Ad Words would work. It was a complete waste of money and very frustrating that Google turns a blind eye to click fraud.

Mike

[awto]

Hi James, I'm sorry to hear about your misfortune, we are in the very same boat, I also had the very same experience, but with smaler amounts, as I configured a much lower daily limit to see what was happening, and just like you I tried with Kanoodle, only $50.00 and within one hour it was gone, and just like you no one browsed the site so I tried again with the sam e amount and same story one hour all gone, I tried with Looksmart and it was better, not so many un justified clicks 60% of the clicks did have visitor to the site and, some business also.
I also tried with google, and I will say about 40% of the clicks did browse two or more pages on the site, but the 60% no record of even looking at the site other than the index page, we are a cash handling equipment manufacture located in China, even we are not Chinese but from Europe, and we also have our own hosting company, in fact we are IPC licenced in China, and therefore have access to much more precise tracking of IP's. You are right it looks to me also a script click, as the search keyword used are all the same.
Also I see lots of adverts saying " get paid to browse the websites " meaning that some search engines would pay some % for people to just click on the PPC adverts ?

I fully understand your furstration I'm the same.

Regards
Otto

[artview]

There is no question as to fraud. The question is how much. Especially when people do not even try to hide their intent. Look at this as an example:

http://www.scriptlance.com/projects/1095743411.shtml

There was another request there months ago for a programmer to create an 'auto clicker'. Now people submitting on the outside told him/her it was illegal, but what of the private messages going back and forth? We can only guess about what was said.

Fraud is there without a doubt. Is it 50% of all clicks? No I don't believe it is that high as fraudsters would go more after higher paying terms. Ones they have found that they can profit from. But is it .1% as mentioned in another post above? Again I say NO. Nowhere near that low. In my opinion it is somewhere in the 10% to 25% range. But that % drops drastically if you were to remove high paying, high profile terms.

An example with us on Google was we dropped the affiliate sites and only had our ads showing on Google itself. Our charges dropped by half but we still averaged the same number of sales. Imagine that!!

Will we continue using PPC. Yes, they bring us results and as long as the cost per sale is reasonable they'll get our business. But only some of them. There are some, including one mentioned in a prior post that will never see a nickel of my money. There are others that are so small you can deposit $25. and go back a year later to see what funds are left. No fraud there, but no traffic either.

I am curious to see how this new engine from Mike Glaspie is going to work. I pre-registered in it with hopes that will save me some $$'s later when it's open for business. Only problem is the most info I have found on it was here: http://www.affiliate2success.com/Zabang.html
And the hypeness sends out some warning signs. But I am one for the underdog and if they can create a comparible product to Google or Overture I'd give them a try.

PPC is here to stay in my view. But it needs constant monitoring to ensure profitability.

[Chrisl1234]

When I saw the claim pop up in my inbox that up to 50% of revenue is fraudulent, I was also spurred to respond - as I just simply didn't believe it either. But I see it's been addressed already.

But while I'm here....

Like any other kind of business, the possibility of fraud is always present. The only real workable strategy one can employ is to keep a constant watch and jump on it when you see it.

Our Adwords campaigns all have trends - when I see a spike, I investigate the source IP using my tracking links. Apart from a few minor incidents, I can't say I've really experienced too much of a problem (which is why I was rather astounded at the 50% claim)

But conversions is another subject entirely. Ours plummeted on Adwords when I first opened campaigns up to syndicated content - traffic is not as targeted and it shows in conversion rates.

For me, to stay on syndicated content and make it work - the sales process had to be two-step. Distribute our free course and follow-up with an email campaign.

I suspect many people confuse PPC failure with inefficent marketing of their clickthroughs - but that's a subject for another day.

[ppayne]

Thou hast caused me to sell my Google stock. ^_^

[roscoagogo]

I run a life insurance web site providing online quotes in the UK, obviously this is a very competitive market place with a number of companies developing very successful businesses off the back of PPC. However, i also know from my experience of working within the financial industry that it is a common tactic to click your competitors site to abuse their profit margin. Surely it would be possible to hinder this tactic by tracking the company IP during the registration and adding money process, and then ignore any clicks coming from those IP's?

I understand that this would not remove the issue but i'm sure it would drastically reduce the problem.

[activeco]

This vendor is the type that when the highest bid is say .35 cents, he will put in a bid of like $2.00, and if you try to get in at second place with say a $1.00, he will drop to .99 and then you will get a flurry of clicks.

If you are 100% sure about that, than next time they go with $2.00, you just start playing their game.

[MarkMaunder]

"Fraudsters," as Eroshenko calls them, create automated programs that repeatedly go to Google or Yahoo! and click on the same sponsored link. These programs will click through to the landing pages, triggering the advertiser or affiliate fee, abandon the session, and then repeat.

If the above were feasible, I'm sure many unscrupulous AdSense advertisers would be boosting their revenue with scripts.

Google has sophisticated fraud detection algorithms that go beyond just tracking too many clicks originating from a single IP address. And IP is what would catch the offender above.

What I find interesting is that ClickLabs was able to detect fraudulent clicks for their study, and the assumption is that Google is not able to do the same. Either ClickLabs has discovered that the team of PhD's at Google are a bunch of useless academics, or this study is a troll.

Mark.

[/quote]