Forum Moderators: martinibuster
Another reason I did so was I heard that 728x90 and 300x250 converted better and I wanted to make the design fit these.
Seems to have been a disaster! My CTR is down 50-80% YoY
So what do I do, go back to the old, dated design, or stick with it and try and find the solution.
If I go with the latter, does anyone know the solution?! ;)
Thanks,
Chris
My CTR is down 50-80%
How are your traffic numbers holding yp? Are they off a lot too?
Did you change the location of the ads as well as the size?
How wide is the new layout? Is it flexible?
Are the ads visible without side scrolling?
Did you change the ad borders, colors, background color, etc?
[edited by: martinibuster at 5:28 pm (utc) on Nov. 6, 2009]
[edit reason] Fixed typo. [/edit]
Last year I tried a new header design with my site. Just changed the top of the page, not the meat where the content and ads were. Didn't think it should change my earnings. Well it did. My earnings dropped 30% with the new design. My source code did not even change, as I use a javascript function to display the header... so the source was identical.
Moral of the story... Finding that perfect layout is very hard to do... once you get something that works... don't rock the boat too much... ;-)
I ended up going back to the original design, earnings shot right back up.
In terms of traffic, that's probably better than last year. Impressions are up.
The layout is fixed width, increased from 770px to 990px. It was 2 column, it now has an additional 160px left column.
The ads are in pretty much the same spots. Below the header and above the main content and top of the right column too on some pages.
I think the ad relevance is OK on the whole, and the ads are well above the fold.
I got the impression that any change in ad units is detected and adsense will knock you back to zero on the quality scale before they get a chance to evaluate the changes. Not awfully helpful if you are also trying to evaluate the changes.
Or it could have been randomness. I generally give changes a week and knock them back if they don't suit me. The changes that work for me are nothing like the generally recommended ones.
The layout is fixed width, increased from 770px to 990px. It was 2 column, it now has an additional 160px left column.
Change nothing except if things are so bad you've become desperate.
In the early days changes proved disasterous and took months to recover. I leave things well alone.
Change nothing except if things are so bad you've become desperate.
Ditto. I think we've had this discussion here on the forums before. I had redesigned one of my well-to-do website. Earning and the works hit the pits for a good few months. Then they gradually began to climb up to previous levels and only after a while they actually surpassed the previous averages.
I personally think that AdSense bot raises flags based on a major change in the site's or a page's structure and takes time to "reconcile", for lack of a better word. It really doesn't care if the change has been done for good or bad.
In my case, I kept my redesign and didn't revert because the redesign was primarily done to improve user experience.
I redesigned my site years ago because it had a dated look, and the ctr on my best performing page tanked. Even the ads changed. I decided to take the adsense earnings loss because I felt the site would benefit from the updated look by attracting new advertisers, and it did.
sounds like you've got more stuff to look at above the fold now, even if it is the same stuff. that is probably affecting the CTR a bit. more text, more links, more images above the fold, less chance they'll click on your ad.
This sounds like it could be the core of the issue to me, especially if you added new site navigation links above the fold, or if you made them a lot more obvious.
Combine that with even seemingly modest changes in ad location and the impact could be huge.
Make changes as simple as moving an ad from below a title or nav bar and placing the ads above the title or nav bar, and BANG, your CTR can take a real dive in a fast hurry.
How long ago did you make these changes?
If you have mostly new traffic maybe waiting it out and seeing what happens is easier than if you have mostly return visitors.
Tough call.
How does the saying go: "There is no progress without change, but change isn't necessarily progress."
I think I may try moving the left hand nav column over to the far right of the page. Or maybe even make it javascript show/hide - hidden by default.
I guess the only way forward is experimentation
Change nothing except if things are so bad you've become desperate.
Ditto here.
i noticed that when you take ads off a page or change the size, that this is detected and the system "forgets" about the page history, i.e. it treats the page/ad as if it was brand new. Not good if you have a page that has been working okayish for years...
One of the most interesting threads that I've read on this subject is here: [webmasterworld.com...]
Read all the posts by the Original Poster. She's gone back to the basics, using html and static pages. She's making an incredible amount per day with the changes. While I'm sure many these types of numbers will only be in our dreams, there is some excellent information there.
I've rewrote massive chunks of code on my site while the design remained the same, and watched the earnings halve overnight, never to return to normal, resulting in the loss of thousands of dollars a month.
I would give it a week or two, the earnings may stabilize.
