?

Log in

No account? Create an account

Previous Entry | Next Entry

I started up a thread titled "PPC Advertising, Click Fraud, and Its Effect on Search Engines" on the TELECOM Digest recently, in the hopes of trying to find some sources of technical discussion on the issue. One of the responses I got was somewhat disconcerting, although expected:


   me: I have described many scenarios such as this where there is no
       reliable way to differentiate them from clickstreams where the users
       do not find what they are looking for at the advertiser's site, or
       decide (non-fraudulently) not to buy, or are just window shopping.

   Yup that's the point. It's *impossible* to do so. Which answers your
   question about "why aren't people working on better fraud-detection?"
   Those who have seriously looked at the problem recognize that
   detecting such actions 'in progress' chews up exorbitant amounts of
   resources, and costs more than the fraud does. The 'simple' stuff --
   e.g. paying only for 'unique' clicks over time -- kills off all but
   the sophisticated fraudster. The sophisticated fraudster, on the
   other hand, is effectively _impossible_ to so much as slow down.



I was hoping this might turn into some kind of business opportunity for me, but that might not happen. There are some companies that do that, and some of the bigger search engines do that in-house, but either they're not hiring, or I don't have the right background for the openings. Because there are so many companies, I'm not sure if the market will support any more. (I suspect many of these companies will eventually have to consolidate.)

There doesn't seem to be much interest in the open-source community either, at least based on looking around the Apache, web stats, and intrusion detection communities. OTOH, there is a sizable and growing open-source movement to fight email spam. Perhaps that's because there are far fewer people who are affected by click fraud than email spam. Although I imagine there's a fair amount of geeks who support their web sites through programs like AdSense, but perhaps they're not targets of click fraud (yet), or they don't notice it enough to matter. I've heard of a few people whose AdSense accounts have been shut down because their ads on their sites received an inordinate amount of clicks, but perhaps it hasn't affected the people who run these sites enough to try to do something about it.

Also, as a result of engaging in the debates over what is/isn't click fraud, and exploring possible solutions, my bp has gone up again and I'm not sleeping well. :( I wish I could figure out how to make a living while maintaining good health ...

Comments

( 4 comments — Leave a comment )
firecat
Apr. 21st, 2005 06:01 pm (UTC)
Just because one person doesn't have the imagination to see that click fraud detection technology might be made cost-effective doesn't mean that it can't be made cost-effective.

If there isn't a lot of it out there, then maybe there's an opportunity, if you're interested in forging it.

(Granted, given your comment on good health, doing the promotional/building work might not be what you want, and I can definitely understand that.)
gregbo
Apr. 21st, 2005 10:22 pm (UTC)
"Just because one person doesn't have the imagination to see that click fraud detection technology might be made cost-effective doesn't mean that it can't be made cost-effective."

True, but I am concerned about the lack of activity among the technical community on the subject. I don't remember the early history of fighting spam, but I remember that there were a lot of people involved, especially in the open-source community. I suppose it helped that people who were highly respected in the open-source community like Paul Vixie were involved.
gconnor
Apr. 21st, 2005 08:18 pm (UTC)
Also, as a result of engaging in the debates over what is/isn't click fraud, and exploring possible solutions, my bp has gone up again and I'm not sleeping well. :( I wish I could figure out how to make a living while maintaining good health ...

I suppose you have thought a bit about what you like doing and whether there are any jobs out there that make you think "Now there's something I could do all day and not get tired of." I'm not sure if I've seen you write about it, but since you wrote about problem solving and more recently about click-tracking, I jumped to the conclusion that you were passionate about both.

In my case, I'm passionate about fighting spam (email spam), but I'm probably not qualified for any jobs that would have me doing that all day. So, I have the next best thing, a Unix sysadmin job that lets me work on spam sometimes, but not all the time. This will do for now... Either I will find a job that's a closer fit to my passion, or I'll become passionate about something else :)

Good luck.
gregbo
Apr. 22nd, 2005 03:37 am (UTC)
It's not so much a question of "would I get tired of doing it all day?" as "what is the most likely work that I will get at this juncture?" I would prefer to get a job doing network software development, but that's not likely. I don't have recent experience, there is lots of competition, and that sector of the computer industry is still sluggish. In comparison, search (especially paid search) is going great guns, and money is being made, not just by search engines but by people who advertise on them. To the extent that money would be lost due to click fraud, there seems to be more interest in click fraud prevention from a business standpoint. I have far more current and recent experience with the problem and possible solutions than I have with anything in computer networking, unfortunately.

If I could get a web analytics job where I am part of a peer group of engineers, where there is an upward career path, and full support from all levels of management for getting the job done, I could accept that, even though it would not be my preference. Mostly, I don't want to wind up again in the situation I am now. I want to be competitive with my peers in the computer industry, as judged by potential hiring managers, other software engineers, etc.



( 4 comments — Leave a comment )