Forum Moderators: martinibuster
Anyhow, I decided I had to do some other things for the last 5 days... and slowly but surely my EPC has been climbing up, it's now at least double what it was before and as each day ticks past I seem to get more "large" EPC's.
Perhaps it's just a trend for this week - perhaps it's really because google likes stable sites that don't change /too/ much.
... anyone?
When I make a change I like to leave it in place for a week or so, and I try not to make lots of changes at once as I can't really work out what changes have worked and what ones haven't.
I'm questioning if AdSense drives your EPC down (short term) due to site content changes.
Yes, you should be experimenting with different placements, colours etc, but you need to be able to measure how they have worked (or not as the case might be).
Sometimes big changes in numbers seem to cause a temporary negative dip that recovers. Also, on occasions a positive blip is experienced that is also short lived. That's why I say run with changes for a week or more before making a decision on if they have worked or not.
3 to 5 days seems to be an acceptable time frame for me to gauge whether the changes have made any impact.
EPC is up again today - though I'll just have to wait and see what it gets like in another 20 hours time (ie, I find that my EPC tapers off as the day wears on).
I agree ..
Best to settle down quickly after experimentation ...
If things are stagnant, yes leave it alone, if every thing you try lowers your earnings, yes leave it alone, I am yet to put down the shovel and take off my gloves and it has bean years and years of building and testing.
I would love to get to stagnant, it's just not foreseeable right now.
I realise this might contradict some Google Adsense advice, but the people who build Formula One engines don't necessarily make the best drivers.
-Main page doesn't have adverts (so it won't affect the AdSense stability measure ;) )
-Existing pages are mostly all articles so they will remain the same over time
-New content goes to new pages, no ammending (unless really required)
EPC is still reasonably good today, certainly double what it was over last week.
Yes to the above--couldn't get more stagnant--and I'm definite done twiddling! I do have the feeling I've just been making things worse and worse. As I posted somewhere before, I feel like I broke it! ;-¦
If I mess with those pages after the ads become more appropriate (after a week or two), I usually have to start the crawler learning (read waiting) process all over again. I once swapped those pages out with YPN equivalents and when I "retreated" to the AdSense versions... had to wait almost a month before the ads were back to where I'd left off!
Chapman
I think it used to be the case that new pages started with the maximum smartprice and EPC then declined as the conversion rates were collected for that page. However, I've noticed, recently, that when putting ads on a page (or reintroducing them after taking them off for a while) the EPC seems to start very low then climb.
I suspect there has been an algo change at some point in the last year in the (month-long) page-related element of the smartprice calculation. Periods without ads are now regarded as having zero conversion rather than full conversion. Ie new pages start off assuming a month of zero conversion rate and therefore have to build up a month's worth of good conversion data before getting better EPCs.
Periods without ads are now regarded as having zero conversion rather than full conversion. Ie new pages start off assuming a month of zero conversion rate and therefore have to build up a month's worth of good conversion data before getting better EPCs.
That's an interesting theory and one that would explain some of the unexpected performance failures on a few of my recently "enhanced" pages.
Chapman
That would certainly make sense to deter the "make-money-FAST" crowd, especially if the ramp-up time gives G more chance to catch the crooks before they clock up significant clicks.
For those of us that expect to stick around for a while I guess losing a month upfront to have it back spread out over later months one way or another is a GoodThing(TM).
"Time wounds all heels."
I don't even attempt to assess the popularity/value of a new page until after about a month's worth of clicks anyway, to allow (1) for the page to get indexed and visited (2) to even out more than a couple of weekly cycles, holidays, etc (3) see most of a typical ad-budget cycle.
Rgds
Damon
My EPC has since tripled _and_ my CTR has doubled, no doubt because the adverts are now of /better/ quality.
Definately looks like leaving pages alone (if they've got good content to start with!) works wonders.
I suspect if a page crashes down then it's possible that
- it's become irrelevant
- advertisers have moved to different topics
Could be a few other reasons.
For now I'm going to ride this one out, so far I'm liking it.
My next questions would be
- did the impressions drop off
- did the EPC drop off
- did the CTR drop off
My first response was to check the page (using the ip range that it is mostly viewed under) to see if anything had changed or if some MFA had taken over. I could not pinpoint anything.
I mean, there's alot of backup here for the theory that the "leave the page alone" technique makes for better performance in itself, but I just changed my highest earning page around completely and the ctr was pretty much the same, but the earnings went up triple a few days later... I was kindof thinking that perhaps it was due to not changing the adsense channels on that page for awhile now hence more statistical value for google.(I'm awful for changing my channels around constantly - twiddling)
I also wonder if the actual referer of the traffic, if it's possible has anything to do with the pricing(good/poor neighborhood). (been getting some incredibly well targetted "good" users)
Happens frequently to me. I just leave it.. It comes back to its normal level after 2- 3 days..Max a week..
