Forum Moderators: martinibuster
Every few months I seem to get to the point where my epc slopes off for no apparent reason. Same amount of traffic, same amount of clicks etc. It seems that smart pricing has devalued me slowly, and progressively.
What I do is to remove all the low and moderate paying ad blocks for a couple of weeks, watch the epc rise based on minimal ad blocks with good ads and then put 'em back again. The earnings in the period of fewer blocks tends to be slightly lower than normal, but more than compensated for when I put the ads back again.
I've done this a few times now and I've just completed this cycle, and am enjoying good earnings again. What really hacks me off is that the channels I remove are adlinks units at the bottom of some pages, and are usually fairly good earners despite having a considerably lower ctr than the main ad blocks. Daft situation or what?
why do you bother putting them back at all?
As regards removing ads completely, I don't think that's going to increase the epc over what it does when you remove the low and moderate earners. Thereforore it seems to me the optimum way of balancing this is to leave the high earners running for a while without the moderates and lows.
My second reaction is that even if this IS the true explanation for what you're seeing, it's doubtful you can game the system to your advantage. You're either better off leaving the ads in place or (more likely) removing them permanently.
then the whole system is a piece of crap
Who said the system is perfect?
On face value Google is trying to maximize benefits for advertisers, thus its own profits, publishers are trying to maximize their earnings.
All three right now are obviously doing pretty well, we are still here right?
No one but Google know for sure how that imperfect system works but that will not stop us from guessing & testing ways to raise the bottom line.
Along the way come in publishers forging clicks, advertisers producing MFAs and Google in their own right changing things around..
Not perfect but it works.
it's doubtful you can game the system to your advantage
I wouldn't call it gaming the system, more like riding the wave, and we know it is a wave right?
...adlinks units at the bottom of some pages...
I usually find that putting the Adlinks units across the very top of my pages (before the H1 tags, before any content, and before any other Adsense blocks) works best.
Have you ever tried this? Seems to help my pages to "target" the ads for the other blocks.
What I do is to remove all the low and moderate paying ad blocks for a couple of weeks, watch the epc rise based on minimal ad blocks with good ads and then put 'em back again.
Works well for me too. However in my case I find that it's better to have fewer ads to start off with .
[edited by: Scurramunga at 10:55 pm (utc) on May 13, 2006]
Basically my site is top of the serps for the niche, is well respected and gets a lot of quality traffic. Now I would have thought that Google would have thought it was prime real estate for them. However, they know best and will insist that the "Best 4 sites on the net" are the best ads to show in place of advertisers that want to sell their services to my visitors. Am I gaming the system by excluding them? Are Google gaming the system by taking off good payers to show the MFA's they seem to love so much?
My view is that by exercising editorial control over the ads that show on the site, I earn Google more. They should thank me for making them richer, not reward me with more MFA's :)
As to the morals of the trick (if indeed it is that), I don't view it as artificially inflating the epc but temporarily correcting one of the many and varied flaws in the smart pricing algorithm in order to realise the true value of the site.
But he's trying to manipulate his smart pricing rating so I don't know what else to call it.
(with regards to removing or adding adblocks)
I like to simply call it 'supplier interaction'.
That is:
Advertisers determine the level of demand by setting the price that they are prepared to pay (bids, budgets etc) Smartpricing also has a hand in price.
whilst
Suppliers (publishers) respond in accordance with the price that they are willing to accept by either increasing or decreasing adblock space.
[edited by: Scurramunga at 12:30 am (utc) on May 14, 2006]
I like this.... :P
Are you removing the code itself as well, or just disabling the code and leaving it on the page?
I'm wondering how the Adsense bot responds (or doesn't) to the presence of the code, or the lack of code, versus if the code is just disabled it will always be present, just not necessarily showing ads, if you get my drift. Anybody have any experience with this? Obviously the bot hits the pages at a pretty slow rate these days.
There is a way with Javascript to dynamically disable the code without removing it (the code remains completely unchanged). Using this technique I was thinking of randomly switching between Adsense and Yahoo then comparing performance over a significant period. Also it would make it easy to control the number of ads on a given page.
Observation:
I've been in Adsense since 2003. Early on I noticed if I changed my ads around periodically, like swapping a wide skyscraper for a narrow one, earnings would increase. It seemed I'd actually have to generate new code and then swap it on the site, and boom in a day or two increased earnings for a while. I believe this was in the early days of Smart Pricing. Seeing this I went to the trouble of automating the code swap out on a daily and weekly basis, but over time the benefit seemed to go away. I haven't tried it recently. I used to think of this as resetting Smart Pricing and at the time it did seem to work, but it also seemed I had to go through the step of generating new Adsense code to make it work! I believe many of my visitors are new to the site, so there isn't so much ad blindness generating this effect.
the other possibility is to look at how the traffic stacks up on a day of the week basis, some days will always be better than others... as near as i can tell, adsense epc is best on the days with the highest traffic, and for the last three months, that usually meant beating the **** out of my epc on the days with lower traffic, in order to make up for the good days.
that pattern has not been so obvious this month... but as others out here have suggested, swapping between adsense and ypn on a daily, site-wide basis could be a really good way to break up any patterns that adsense is using to game your website with.
it's also a lot more drastic than simply pulling a few ads for a few days, which makes it a more attractive option :-)
