Forum Moderators: martinibuster
However, two weeks after I made the changes my revenue CPM numbers dropped significantly. They dropped from one day to the next and everything is now below my old average. What gives? I thought I had done some great work with my optimization and that I was actually going to start doing well with Google. Then it only lasts for a couple weeks and I now have more ads clogging up my site and am making less money for it.
This same thing happened to me last time I did some big optimization last year. The numbers went up and I had my biggest month at the time, then my CPM numbers dropped down just as quick. Does Google do something to regulate your numbers or something? Why is that everytime I make changes I see a big jump and then my CPM drops big time soon afterward?
- As a result of your optimization, CTR is going up, but conversion rate is declining. That's triggering a "smart pricing" discount.
And/or...
- You're using up your allotment of higher-paying ads and are being served more lower-paying ads, leading to an overall decrease in EPC and eCPM.
Did your optimiziation result in you adding more ad blocks?
The OP has already answered this question:
"I now have more ads clogging up my site and am making less money for it"
I rest my case.
You know what's funny - I only started doing this because a Google employee contacted me letting me know of ways I could increase my numbers. He gave me some ideas on new ad blocks (yes, he was the one suggesting I add more ad blocks). I took his advice and made some of my own changes. Thanks to him I'm receiving less revenue now.
I checked all my numbers. Only the CPM rate dropped. And as I said, it dropped dramatically from one day to the next, specially on October 2nd. I can't see any reason except for smart pricing.
Being a publisher and an advertiser I strongly disagree with what they're trying to do with this smart pricing. Adsense is a CPC ad medium, not an affiliate program. Why should I be held accountable for how well an advertiser has designed their site for "conversion rates"? I have no control over their ability to convert a lead into a customer. I did my job by sending them the lead. If it doesn't work for them they should choose new keywords. Bad business in my opinion. Publishers are definitely getting the short end of the stick here. I don't know of any magazines that get refunds if the ads they run don't convert.
As an advertiser I wouldn't expect to be refunded for ads that didn't work. It's my responsibility to come up with better marketing strategies. Google is putting more responsiblity on the side of the publisher to help sell the advertisers' products. And there are too many things that are not in the control of the publisher. And yet, they lose out anyway.
1) You have some cheap payers floating around, nix anything that offers something free for a survey or something too good to be true.
2) You are being CPMed from a text ad standpoint. When you see one ad eating up the whole block with a slight larger than normal text that is usually a CPM ad. The advertiser is targeting you site and usually paying cheap for the views hoping to harvest more clicks than CPC.
3) When you go image only you are opening yourself up to more CPM ads most of the time image ads are CPM ads.
[edited by: Khensu at 2:58 am (utc) on Oct. 11, 2006]
changing ad types from text to image ads
IMO this could be a large part of the problem. Unless you have extremely strong design reasons to allow image ads only or text ads only, you are best off simply allowing both. It means that you have the largest number of advertisers competing for space. I allow both and actually only ever see text ads--what you see will depend on your keywords.
By the way, image ads come in both PPC and CPM variants....
I agree with others that you may have over-optimized. My experience is that MOST of the things I try have no effect or a negative one. The most positive changes I've been able to engineer, and which stuck, were: taking ads off two high-traffic but very low CTR pages, which actually increased earnings; redesigning my home page to incorporate an adlink block, where I previously had had no ads on that page; adding content in areas that I knew interested my visitors and that got good ads.
And it's definitely true that once you reach a certain point, tinkering doesn't do much good. That's when link development and content building become more worthwhile.
Good luck!
ludachris:
You know what's funny - I only started doing this because a Google employee contacted me letting me know of ways I could increase my numbers. He gave me some ideas on new ad blocks (yes, he was the one suggesting I add more ad blocks).
It seems ALL Google Optimization employees always suggest adding MORE Ad blocks! I've also been contacted and a day after I implemented the changes, my income dived :-(
Haven't read anything here that an employee actually suggested removing Ad blocks...
Looks to me like these employees are as clueless as we are how the algorithm works!
Being a publisher and an advertiser I strongly disagree with what they're trying to do with this smart pricing. Adsense is a CPC ad medium, not an affiliate program. Why should I be held accountable for how well an advertiser has designed their site for "conversion rates"? I have no control over their ability to convert a lead into a customer. I did my job by sending them the lead. If it doesn't work for them they should choose new keywords. Bad business in my opinion. Publishers are definitely getting the short end of the stick here. I don't know of any magazines that get refunds if the ads they run don't convert.
This has been discussed endlessly in other threads, so I won't repeat the arguments here, except to say that:
1) Smart pricing has been around for more than two years, so there's no reason to believe that it will go away.
2) A publisher can't blame the advertiser's copywriter if his or her traffic converts at a lower rate than the norm for that advertiser. (Also, a "conversion" doesn't have to be a transaction--it can be an inquiry, a registration, a certain amount of time on the advertiser's site, etc. So if, for example, a high percentage of visitors from Site X spend only two seconds on the advertiser's landing page while visitors from other sites spend an average of 10 seconds, it stands to reason that Site X is a strong candidate for smart pricing.)
As I said before, all other numbers (like clicks, impressions, etc) have been the same since before the drop. The only number to drop was the CPM. It seems that once I started delivering more clicks, and my CTR went up, I started making more money. Then one day my eCPM dropped like a rock while all other numbers remained the same. There isn't much else I could attribute to this other than smart pricing.
All others,
Thanks for the suggestions and for the feedback. I'll see about making some more changes to find what works best for my site. Maybe other ad programs are worth looking into. Thanks guys/gals.
I really believe Google is screwing publishers this way. I am a firm believer in a publisher only being responsible for delivering the lead, and if the lead is delivered the payment should be made, unless of course you choose to sign up for an affiliate program. Publishers should not be held responsible for the advertisers' ability to close the deal - no matter what they consider being a "conversion". Again, what other advertising medium out there is so advertiser-friendly that the publisher gets penalized in situations like this? Here we are taking value away from our members by implementing advertising links and making them appear to be part of our content and we're at the mercy of the advertisers' sites to make money. Seems a little backwards to me. Seriously. In any case, what voice do we have really? It seems the advertisers have all the influence in this situation. The publishers just take what they're given and are told to be happy with it.
Here we are taking value away from our members by implementing advertising links and making them appear to be part of our content...
If you're doing that, you're reducing the value of your site not only for readers, but also for advertisers--and your site is a poster child for the "Why Smart Pricing is Necessary" movement.
It seems the advertisers have all the influence in this situation.
Well, they are the paying customers.
It's also important to remember that all audiences and all traffic are not the same. To use a hypothetical example from another thread, a request for a Ferrari brochure from THE ROBB REPORT or Forbes.com may represent an actual sales lead, while a request for that same brochure from teen-drivers.com or free-brochures.com is almost certain to be from someone whose fantasy life is nourished by upscale car brochures. Blaming the brochure copywriter if a teen-drivers.com or free-brochures.com lead doesn't convert into a Ferrari sale is either disingenuous or naive.
the cpm rate doesn't tell you anything, it's a worthless number wrt contextual advertising.
1) what happened to the epc.
2) has your google search traffic changed? pages gone supplemental? etc.
3) what do your channels tell you? forum ads vs. image page ads vs. below-the-fold positioning? etc.
if the forum pages are not making money(per your channel stats), i'd get rid of the adsense on those pages, because it could be driving down your average epc... which in turn brings in that low-rent garbage cpm advertising.
there is nothing wrong with multiple ad blocks per se... google puts up a dozen advertisers per page on it's search results.
I had a quick look at what appears to be the OP's site.
Two points:
1) Someone from G suggested putting a slim AdLinks strip where the eye naturally falls just above the main content. My revenue has gone up >4x with this change for my site, and the placement is not, to my eye, massively intrusive. I think that it could work for the OP's site.
2) Less definitely can be more, but not unconditionally/always. My ad-layout algorithm automatically/randomly hunts for the best layout for CTR (I have to tweak for eCPM if they turn out not to be the same). After I put in the slim AdLink unit as in (1) the best CTR shifted to a layout with no towers and no other ads above the fold except (often) a banner above the AdLink strip. Quite strange because until then a left tower layout had been *by far* the best.
So:
a) YMMV.
b) Not all G suggestions work for all sites, so see point (a).
c) The different ad blocks interact with one another, eg in the user's view of page "clutter", so you really have to try combinations of things. If my auto-retest-layouts algorithm had not been running (and it still is running) then I might well not have noticed the new optimal combination for my site; see (a).
d) YMMV.
Rgds
Damon
If you're doing that, you're reducing the value of your site not only for readers, but also for advertisers--and your site is a poster child for the "Why Smart Pricing is Necessary" movement.I think any time you have to implement ANY advertising as a means to pay the costs it's a trade-off and/or a sacrifice in the user experience. I'm implementing as tastefully as I can without making the site look terribly cluttered. So I would say that I am not a poster child for the smart pricing model. The empty sites that are built solely to produce Adsense income are the sites that should be taking the hit, not content sites like mine.
Well, they are the paying customers.It's also important to remember that all audiences and all traffic are not the same. To use a hypothetical example from another thread, a request for a Ferrari brochure from THE ROBB REPORT or Forbes.com may represent an actual sales lead, while a request for that same brochure from teen-drivers.com or free-brochures.com is almost certain to be from someone whose fantasy life is nourished by upscale car brochures. Blaming the brochure copywriter if a teen-drivers.com or free-brochures.com lead doesn't convert into a Ferrari sale is either disingenuous or naive.
Again, all those who replied with recommendations, thank you.
deterioration caused this thread to be chopped and locked
[edited by: jatar_k at 6:35 pm (utc) on Oct. 12, 2006]
