| 3:31 pm on Mar 6, 2014 (gmt 0)|
What gets me is they keep making BAD tests and moves. Almost every change they have made for years just drives numbers down even more. I make 1/15th of what I used to on the same ads and traffic on many sections. You wonder why mess with something that works or at least go back to something that performed much better.
| 7:36 pm on Mar 6, 2014 (gmt 0)|
adamxcl, keep in mind they don't care at all about any individual publisher, you or I. They care about the change as a whole, if it makes more money. Their main responsibility is to their shareholders, period. You might even consider they care more about how much money they make overall than, how the ad actually looks...
| 2:51 pm on Mar 7, 2014 (gmt 0)|
It is frustrating that the more AdSense tests the more everyone loses out. Publishers numbers are down and AdSense numbers are down. AdSense is desperate to try and reverse their downward trend, but they are making the earnings worse for themselves and publishers. Not to mention, their unsuccessful testing has been going on for long enough to know which changes are working - which are not. They cannot be basing these tests on actual data. They need someone from Adwords to go over to AdSense and explain how to do testing.
| 4:09 pm on Mar 7, 2014 (gmt 0)|
|It is frustrating that the more AdSense tests the more everyone loses out. |
"Everyone" is not having the same experience with Adsense right now and "Everyone" does not see what Google is doing as bad. Some actually get the fact that this process may just plain be necessary for the survival of the program. Just as there are winners and losers in Bull Markets and Bear Markets. So it is with Adsense. Some are finding success. Others are having problems.
| 5:42 pm on Mar 7, 2014 (gmt 0)|
Webcentric - when googles adsense earnings are down as much as they are. Numbers they are communicating - it means there are more losers than winners. No one is saying the testing isn't helping some people. But again, googles numbers prove their testing isn't helping the majority of publishers nor themselves. That's my point.
| 6:00 pm on Mar 7, 2014 (gmt 0)|
@avalon37 -- Fair enough and can't argue against your general point. My question is whether or not the experimentation is causing the general downturn or it's a reaction to the general downturn? I think it's the later which makes it a "necessary" thing and that would be my explanation for why it's being done. It's necessary for the survival of the program. I don't see it as a cause, I see it as an effect.
| 7:02 pm on Mar 7, 2014 (gmt 0)|
They've been testing for years. There's only so much you can do with an ad. You would think by now they'd know which ad style is the most effective.
Seems to me like it's nothing more than job security for some people.
| 9:11 pm on Mar 7, 2014 (gmt 0)|
My point was that the lousy testing started years ago when things were very good. At the peak, maybe 2007, they changed things and a dip. Then change and a dip and it just went on and on. I myself have never seen a change and then a tick upward.
| 9:20 pm on Mar 7, 2014 (gmt 0)|
That's what Google does... they keep trying to fix things that aren't broken. In fact, they keep screwing things up.
| 9:23 pm on Mar 7, 2014 (gmt 0)|
One thing over time that sticks in my mind is the way they've been moving towards getting more and more ads into a single ad space. That's one of the "experiments" that seems to have been going on for ever. Panels are just a newer permutation of that experiment it seems. The variations they've tried have been messing with my sense of balance for years so I'll ad my sense of annoyance to that point. Cramming more ads in has to fall under the "greedy" category and who knows, maybe that continues to be a huge mistake. Some variations worked fairly well (2 ads side by side in a leaderboard was pretty decent) but some have been a complete mess (six ads in a leaderboard for example). Maybe the chicken did come before the egg but we may also at the same time be in a place where continued "experimentation" is the only option left.
| This 40 message thread spans 2 pages: < < 40 ( 1  ) |