| 8:24 pm on Jan 24, 2005 (gmt 0)|
Thanks for the link. Interesting article.
But I question your conclusion: "The price for words has skyrocketed." In the article, they cite "refinance" as a word for which bids have risen from a few pennies to $12. That's the only example given. What has happened to OTHER high-value words? And even if other high-value words have also gone up, what about the more run-of-the-mill words? I don't think they've skyrocketed--perhaps they have, but there's evidence for that in the article.
| 8:38 pm on Jan 24, 2005 (gmt 0)|
|had written a simple software program that relentlessly clicked on his ads, burning up his ad budget and pushing his links off the search sites by lunchtime each day. |
Can anyone hear the sound of endless WWers programming this new "feature" for themselves. What's scary is that this can hurt adwords as much as adsense.
I wonder if one day Internet Protocol will grow to the point of creating a mechanism to ensure no more spoofing IPs? You request a page? Then I'll ping you w/ an ID, send it back to me and I'll serve it. At least once per session. Should slow down the internet browsing experience, but stop spoofing cold.
| 9:12 pm on Jan 24, 2005 (gmt 0)|
Great for the topic to make it to the home page Brett. It is one of the issues for 2005. But I would like a serious investigative reporter to do the job properly. When I saw:
I thought you meant the story was one of the minors doing a program that drained their own budget under a controlled experiment. now that WOULD be a story wouldn't it? set aside a few hundred dollars and see how simple it is to "burn" the money on the phrase "something non-sensical" or a phrase that couldn't possibly burn $100 in a month.
|a competitor—had written a simple software program that relentlessly clicked on his ads |
Doing that WOULD be a story and would be an excellent litmus test especually if done accross the various PPC suppliers for a comparative analysis.
That is the way to do properly report the issue I would have thought.
| 1:04 pm on Jan 25, 2005 (gmt 0)|
I had to read that twice, but now I get it. Bid on an esoteric keyword on all the major PPC sites. Write some code to drain it with varying level of sophistication. IOW, try it all from one IP, try varying IPs, use timing patterns, use patterns on where the ad gets seen, etc. and do an analysis of how well each system identifies the click fraud. Then see if any refund is offered. If none, then complain and see how much refund comes about.
Anyone who does that will get a nice boost of traffic once the report comes out.
| 3:15 pm on Jan 25, 2005 (gmt 0)|
That's it. You explained it so much more eloquently, Clark. Someone sticky me if they run the gauntlet please - I'd love to see the results.
| 6:29 pm on Jan 25, 2005 (gmt 0)|
Must it be pointed out that fraud and theft are still fraud and theft and in my state anything over $100 is considered a felony? Clicking a mouse over and over or running a script to "bump" your competitor may seem innocent enough until you're standing behind bars with your new best friend Bubba.
| 6:57 pm on Jan 25, 2005 (gmt 0)|
|Must it be pointed out that fraud and theft are still fraud and theft and in my state anything over $100 is considered a felony? Clicking a mouse over and over or running a script to "bump" your competitor may seem innocent enough until you're standing behind bars with your new best friend Bubba. |
You bring up a very good point. However, I believe that clicking ads on Google search results is perfectly legal, no matter how many times you are doing it. You are not legally bound to any contract when running a search on Google, and it is their responsibility to judge click fraud.
AdSense is a different beast however. You are profiting from these clicks and violating TOS.
| 7:01 pm on Jan 25, 2005 (gmt 0)|
incredibill, in the test case we're discussing, the idea is to click your own ads to see how well the antifraud measures work.
| 10:00 pm on Jan 25, 2005 (gmt 0)|
I was referring to the original topic post, which was a link to the article about the guy that bumped off his competitor daily by (they think) running a script that exhausted his daily bankroll until his ads went away.
Clicking on a link may not be illegal, unless you're deliberately trying to knock off a competitors ads which the topic was about. There are laws in many states regarding unfair business practices, and that would certainly fall into besides fraud and theft. Sadly, his ads were supposedly attacked by a German competitor and trying to sue someone internationally just isn't worth your time or money.
Speaking of fraud....
Maybe some day I'll take the time and type in the long story of how some NY charlatans cloned my web site and had about 10 copies running when I discovered it, all of them running AdSense. The short story is Google didnt care they were paying thieves that clearly violated their TOS, no matter how many hundreds of pages of proof I had, and I got no help from them.
Luckily when I invoked my rights under the Digital Millennium Copyright Act (DMCA) the ISP hosting them caved and nuked all the sites. They have yet to return many months later.
On the upside, my blood pressure has returned to normal since then.
| 6:18 pm on Jan 30, 2005 (gmt 0)|
It's interesting that they only picked one word. I have had some weird experiences there but found that finding alternative keywords served a better way of mounting an "attack" against someone that is trying to exhaust your budget. I went with common words initially...had a little fraud but not enough to report. I kept very close watch on them and when it happened I pulled the words altogether or dropped them to $0.05. It was probably a PTR or something like that because dropping to a very very low rate proved to do the trick and other competitors dropped off the results by 2-4 pm est. Very weird never really could figure out what was happening.
Ended up finding alternative keywords, I would note NOT through G's or O's recommendations, by asking different people in the company that have frequent but obscure customer contact. Through them I learned that there was an easily mined subset of keywords already in our database...this proved to be huge. The problem was that as I introduced the content into the PPCs the keywords started to show up on the "recommendations" list. Grrrrr. I guess I "created" the traffic? Which is ridiculous...it was already there just not being served and therefore not accurately captured and projected. Since then pricing has moved up considerably as generic affiliates started moving in. They have taken up the top spots but our sell through and my traffic costs have remained around 10% cost for a 2.3% sell through. We have a very high margin, broad product set, one to 5 sales per SKU per year over tens of thousands of individual SKUs.
Moral: Find obscure words from your customers...don't rely on the SE's. Keep your ads priced relatively the same...it hasn't effected our sell through.
Good luck. The only reason I am posting this is to help people find quality content. If you can help them do this then everybody is happy...customers AND vendors.
| 9:48 pm on Jan 30, 2005 (gmt 0)|
|Through them I learned that there was an easily mined subset of keywords already in our database...this proved to be huge. The problem was that as I introduced the content into the PPCs the keywords started to show up on the "recommendations" list. Grrrrr. I guess I "created" the traffic? |
I have 3 reactions to this:
1. As an innovative advertiser finding unique minor keywords...GRRRR. How can they do this?
2. As an adwords advertiser sometimes not having the time/energy to come up with a good keyword list. Bravo, good job Google. Makes my life easier.
3. As an adsense publisher, Bravo! Keep the income coming in.
But at the end I think it's unfair to the hard-working innovative advertiser. Spending a lot of time and effort only to give your work to competitors? Maybe they can be more fair by only adding keywords that are used by x% of other advertisers.
What I would do if this happened a lot is to try and game the system. Adding keywords with traffic I can use but totally irrelevant so the keyword list would get messed up.