Forum Moderators: martinibuster
Nope No Way.... I have a website to help developers. I may have people clicking on my links but that is what they want right? There was a day I put up a gallery of photos with the ad at the bottom the next 2 days I got a lot more impressions and only a few extra clicks...
Anyone else have this happen?
I emailed them and they have NOT responded yet.
I am FED UP....
Sounds like another CON to get your ads for FREE for them.
I have found adsense and the people behind it to be extremely fair and honest (and nice people).
What are the other best Ads to use....
I have a site that has not been using ads and I just decided to pick them up. I would like to use the best ones... only professional ones.
It seems there is a growing number of AdSense users who are getting the "fraudulent clicks email" without deserving it. I'm in the same boat, I just lost AdSense and while I did click a very few times in the beginning, after the first warning letter I simply never touched them. Unfortunately, Google doesn't agree.
For me the problem was that I simply had no idea how strict Google would be. New AdSense users need to be aware that to survive Google's understandably strict TOS, they mustn't just not click their own ads, they must avoid anything that will at all suggest it. Clicking your ads at all is suicide, and CTRs over %6.0, or even checking your AdSense account from different computers (which is what I did - work and home), is asking for trouble. If someone from another computer that I checked my account on somehow got to my site and clicked an ad, well, I think that's what got me. That's my ignorance for sure, but if Google had educated me a little I'd never have done this.
What's even worse is that after receiving the June check in South Korea quickly, the July check still hasn't arrived. Although I only just lost my account July's check is so late that I'm starting to wonder if my account wasn't actually on the ropes 4 weeks earlier, when the July checks were sent.
You said: "Sounds like another CON to get your ads for FREE for them." No offence, but this moaning approach will plain and simple not help you. Keep working at your content, try other ad networks, and use the experience to motivate you. Before AdSense my site had no revenue, for a little while I had a shot, now no revenue again. But I've learned loads and the money has made me realise what I can really do with my site.
Let's just hope that one of these days we can apply again and get rightfully accepted, and that the AdSense algo sorts itself out.
Best Wishes,
Jeremy
I've had clickthroughs much higher than 6% so thats not the issue (and I got the warning email too)
The issue is you clicking on your ads and maybe the mult ip logging in.
I think you were wrong clicking on the ads and are working from that huge problem but if I were you I would apologize to adsense and ask for another chance.
Maybe you were just testing it initially to see if the ads were being tracked?
If you are given another chance, I would not be careful, I would treat adsense with the respect it deserves.
Put the ads on your site and dont click on them or ask anyone to click on them.
Drive traffic to them and let nature take its course.
even checking your AdSense account from different computers
maybe the mult ip logging in.
No, this does not have any bearing at all. I've logged into my account from numerous IP addresses, including from the Thistle during PubCon, and I am sure there were others logging in from the same IP address during that weekend too. If something would have raised a flag, that definitely would have ;)
Email the AdSense team nicely, offering your logs, and offering to block the IP address from your site that generated the fraudulent clicks.
Did they give you a specific reason why? Or was it the standard form letter you received?
CTRs over %6.0
I simply had no idea how strict Google would be
Their terms/policies/FAQ made it pretty clear how strict they would be. And they are acting on behalf of their AdWords advertisers, and obviously those advertisers don't want to be paying for any webmasters "testing" or "checking" AdSense ads.
So to keep the advertisers happy (who are the ones paying for the $ clicks that AdSense publishers receive) Google has to be strict. Otherwise fewer AdWords advertisers will choose to run their ads on the content sites, Google will get less money from their advertisers, and all AdSense publishers would suffer from the consequences of the few who test-click.
but do they really maintain a log of which IP's are used to log in?
It wouldn't surprise me one bit if they have that information filed away somewhere for the month end account audit and fraud investigations.
They have said before about monitoring IP addresses to exclude owner-clicks, so they would have to keep a log of those IP addresses.
The click through ratios shouldn't be a problem either as long as they are somewhat consistant. 20-30% is not odd, there are many sites/industries that have this.
Most spikes in click ratio can be explained...and Google surely knows this. Say for example I have a site on Widgets, and it gets 100 visitors/day and a 1% ratio. Nothing to write home about huh. Then one day, Leo from The Screen Savers or Call for help (TechTV) mentions the site and traffic shoots up to 100,000 unique visitors for the first day, 5,000 after that for a week. This is legit, and happens everyday...and google knows this...
-Julian
I doubt it is so much that suspicious clicks come from a single IP that triggers alarms for most webmasters, but a concentration of clicks from IP's associated with a particular ISP or particular geographic region. Sure, if you want to make it easy for them, click your own ads from your own computer - perhaps even with the Google Toolbar turned on - but I have no doubt that they can detect less clumsy attempts at fraud.
There is always a risk with any effort to limit fraud that innocent people will be warned, or even cut off, along with the guilty. I am confident that Google is aware of that problem, and will continue to refine its system to minimize any negative consequences for innocent webmasters.
It is my impression that the statistics Google provides probably are relatively accurate for CTR over any given 24 hour period, I am not convinced that they are accurate as presented throughout the day. A sudden spike may be the result of any number of things, including a slow server, a technical problem at Google's end, or even the approval of clicks which were electronically flagged as suspect.
I guess now that I have been there ONE full month... and they would have to PAY me they have decided against paying me....
I guess one more CON goes down in my BOOK. I will tell my huge list of Developers about the CON so they don't fall for it too.
Have to wonder if some of the "PSA days" as we refer to them around here are control factors implimented by google.
Care to elaborate? I ask since PSA clicks cannot be tracked by Google unless the target sites share their web server logs with Google so I don't follow how Google could determine anything by displaying them.
Besides it's been a slow month on webmasterworld for conspiracy theories:) Think pubcon took alot out of people:).
Why wouldn't Google be able to monitor PSA clicks in the very same way it monitors other clicks? The links, after all, are on their servers, not ours. Just like with regular ads.
I don't think so. There's no tracking code. Just look at the Properties of the link.
links, after all, are on their servers, not ours. Just like with regular ads.
I believe you are mistaken. Paid ad links have a target URL on the host pagead2.googlesyndication.com which redirects to the destination site. Non-paying ads (PSAs) have a target URL of the destination site directly.
Have to wonder if some of the "PSA days" as we refer to them around here are control factors implimented by google.
Figure a site has 100 clicks per day in a 90/10 revenue/psa ad mix. If Google changes the mix to 60/40 and the site still has 100 clicks per day, that's a pretty good indication that something's awry.
On a separate note, it also seems that Google may 'worsen' targeting to temporarily punish Webmasters suspected of improper practices. I just got a note from Google insisting that I fix some stuff on my site (change some wording as it relates to AdSense), and perhaps not coincidentally, I've noticed that my site has shown predominantly PSA ads over the last week whereas I had had few PSAs before.
If there were control days implimented into the algo by google, then they know which days should be lower revenue.
I understand. But what would Google hope to learn by doing this? Are you suggesting there is some publisher behavior they could detect by doing this and if so, what?
If they typically serve 10% PSA's and on any certain day served up 30% PSA's to a site, would expect decreased revenue.
And except for negatively affecting publisher revenue, Google revenue and advertiser ROI what would this accomplish?
Take a site of mine for example. The site has 1,000 pages, all serving AdSense ads, the CPC and CTR of each page vary daily (just guessing based on my knowledge of Adwords and some testing I've done) as do the impressions on each page (which I can measure). 15% of the pages (not page views, associated page views are lower) always show PSAs because of poor content-targeting and negative trigger words and on average an additional 10% of pages (again, not page views, associated page views are higher) show PSAs due to lack of supply of content-targeted advertising.
Let's say Google runs a series of daily tests (days chosen randomly) over a long period of time by which they force a higher percentage of pages (and as a result page views) to display PSAs. I think we've already agreed that it's unlikely Google's PSA partner sites share web server log data with Google so they can't measure impressions on the PSAs. What could Google learn from these tests?
I'm not asking to be a pain, I'm asking because I still don't understand what you're suggesting and I'd like to understand. If anyone else understands feel free to give me a virtual slap to the head and enlighten me.
Longterm tracking of stats shows some variances that cannot be explained by end of month decreases, or other factors.
Variances in revenue or some other metric? Are you confident that you can account for "other factors"? I have more to say about this, but frankly my main interest is understanding what you think Google would learn from these tests.
It is not unusual to see a "high day" following this occurence. Is it a tracking issue that creates this with revenue being adjusted the next day?...or are ads adjusted to make it up?
In our stats we see these type of "drop out" days that we have been trying to account for. Haven't been able to come up with any logical explanation that fits the numbers we have been seeing. Seeing this thread about "penalized sites" got us thinking again, about this as a possible explanation.
We have no "proof" of this just trying to explain some unexpected patterns we have been seeing. We combine this with some of the responses from google that have been posted by others with psa's showing and came to this as a possible conclusion.
I think this would be one of many anti-fraud checks that google could be using.
I'm pretty sure this is the case.
Figure a site has 100 clicks per day in a 90/10 revenue/psa ad mix. If Google changes the mix to 60/40 and the site still has 100 clicks per day, that's a pretty good indication that something's awry.
Ah. If that's what Visi meant, I understand. So we're talking about tests to detect publishers who use prohibited methods to either hit a per day revenue target or per day click target.
However to come to the conclusion that the publisher is generating clicks through prohibited methods, they'd need to ensure that the results are statistically significant. Since Google was founded by PhDs and many at Google have PhDs I assure you that they don't willy-nilly draw conclusions and take action based on data that isn't statistically significant.
Ignoring all of the factors that affect clicks on a site with many pages (some of which they could account for), at a minimum they'd need to do these tests randomly for
a number of days over a long period of time.
But why bother? First of all this type of testing would only be effective if the publisher didn't notice the PSA percentage spiked (which wouldn't be hard to detect if a publisher wanted to) since the publisher could take adjust his/her behavior accordingly.
Secondly, it would be easier for Google to detect this behavior via other statistical analyses and they could do so without reducing their revenue, publisher revenue and advertiser campaign sizes. Google has lots of data (or can derive it from data they have) about each impression/click available that we don't - its cost, position in the ad, page, date/time, IP, country, click history for that IP, click trends for that IP, CTR for the ad on other publisher sites, CTR for similar publisher sites in general, impression patterns for a page/site, click trends for a page/site, etc.
That's my 2 cents.
Think adamguy post summarizes my position. By simply running a high psa day on your site...a random occurence...if revenue or click rate does not drop with it, indicates something abnormal on that site.
There's the rub - the revenue change doesn't necessarily indicate anything unless the test accounts for other factors and/or is done over a statistically long enough number of measurements.
In our stats we see these type of "drop out" days that we have been trying to account for. Haven't been able to come up with any logical explanation that fits the numbers we have been seeing. Seeing this thread about "penalized sites" got us thinking again, about this as a possible explanation.
I'll share some findings and you can take them with a grain of salt. I have one site that had nearly all of hundreds of pages showing PSAs for an extended period of time due to what in my opinion (and many others') is a problem with how Google's algorithms determine "negative content". In my case the site is about sex crimes so even though there are plenty of relevant ads (albeit with low CPCs) nearly all of the pages contain several to numerous trigger words and therefore show PSAs.
As a result I eliminated AdSense from all pages except a single high traffic page for more than a month and I've seen zero indication of any PSAs being displayed on the page. I did this because I didn't see the point of showing PSAs 100% of the time on 99.9% of the site's pages, but the effect is that Google's reports and my associated statistics are for a single high-traffic page with a very stable group of advertisements (small number of companies advertising, small number of total ads, very little CPC variance). Also, the ad color, ad type, ad position on the page and content of the entire change has remained unchanged.
Yet there is a fair amount of variance for daily clicks, CTR and revenue. I could share the normalized averages, standard deviations and any other statistical measures anyone is interested in, that is if it seems that's not a violation of the TOS. What I mean is instead of saying my CTR on average for the month was 3.0% with a standard deviation of 1.0% I'd say the normalized CTR was 1 with a standard deviation of 0.33.
I think this would be one of many anti-fraud checks that google could be using.
It could be, I just don't think it's likely given reasons I gave here and in an earlier reply. Of course that's just my opinion and I'm just a fairly anonymous guy typing this message.
[edited by: richmondsteve at 12:47 am (utc) on Oct. 1, 2003]
by the way guys i m seeing a increase in clicks rates around from 1.5 to 2.8% is it okay.
Why don't they just filter out any bad eggs.
Oh well I am beating a dead horse ... I just think it sure is convienent to stop my account a day before they are supposed to pay.
Oh well, I will move on to a legimate company.
