Forum Moderators: martinibuster
I run a website with around 40,000 impressions a day. Daily uniques average around 5,000 to 10,000. We thought Adsense would be a nice way to monetize traffic. So, we joined about four months ago. We've made some "good" money, I suppose.
However, we believe that we are not getting paid what our traffic is "worth". How do we arrive at that conclusion? Well, here's how:
Say for three days in a row we receive "X" in revenue from Adsense. During that three day span our revenue per click is, "Y". Our effective cpm on the pages we have Adsense at is "Z". And, finally, lets say our conversion of impressions to clicks is "10%".
Okay, then the next two days everything tanks. Our revenue drops by 50%. Our per click drops by 50%. Our CPM drops 50%, and the click conversions drop by 50%.
What bothers me is that the things Google controls (what they pay us) drops but the the click-throughs drop as well. The click throughs are something that is a product of OUR TRAFFIC.
Next, it seems just a tad coincidental that our click-throughs drop, AND what we get paid per click drops, too. We know, also, that the keywords we target are BIG revenue generators because we know what folks pay for them.
It just frustrates the daylights out of us. We WANT to send Google traffic. We work our tails off to do that, and there is just no rhyme or reason to the way they operate.
Finally, the system of "customer service" (notice my use of quotation marks) is, well, abysmal. What business model can you think of that says, "we won't tell you what you'll make. we won't tell you how we decide what ads to serve. we won't tell you what anyone pays per click. we won't give you a customer service number to call. we won't let you talk with anyone about this... etc"?
Look, if ANYONE has some advice on what we can do to make Adsense work, please help us out. I think that Google has a great product with Adsense but I just can't figure out how to get some basic (and I mean BASIC) questions answered.
My fear is that to Google, our "x" per day is, literally, meaningless. They figure that for every guy like me that really tracks this stuff and wants answers, there are 1000 more who don't question and don't follow it closely enough to know the difference between a good or bad deal.
Any thoughts are greatly (and, I mean, GREATLY) appreciated.
So someone stating a six-figure income would mean $100,000 or more.
But doesn't Google track conversions "objectively" after a while? I mean using the methods supplied by AdWords?So, even if one starts at rock bottom EPC, shouldn't it go to a "fair value" after a while?
Can someone provide any feedback on that?
Has conversion tracking code been widely adopted?
I mean, whether a Adsense publisher starts at $5/click or at $0.03/click, shouldn't it eventually converge to some "fair" EPC? (dependant on supply/demand ofcourse, if there are no advertisers for your topic...)
I mean, whether a Adsense publisher starts at $5/click or at $0.03/click, shouldn't it eventually converge to some "fair" EPC? (dependant on supply/demand ofcourse, if there are no advertisers for your topic...)
An AdSense publisher doesn't start at a certain rate per click; a publisher starts at a whole collection of rates per click, with those rates being determined by:
1) Advertisers' bids for the keywords that trigger ads on your site;
2) The discounts that Google is applying to the advertisers' bids for those ads on different types of content pages.
In other words, if you have 100 different ads on your site one day, each of those ads may have a different price--and the next day, you could have a different mix of 100 ads, with different prices for each.
BTW, advertisers' bids have nothing to do with your site individually. (To paraphrase Gertrude Stein, "a bid is a bid is a bid.") Advertiser discounts could be determined at the individual site or page level, if Google uses conversion tracking when calculating the discount rate for any given page and ad. (Google isn't telling exactly how it computes advertiser discounts, so who knows?)
I don't want to be associated with this kind of advertisers so I blocked it already but I am concerned that more are coming.
On our HK travel guides we had all 4 ads offering flowers to HK. Most of those domains we blocked already but new flower companies are still showing up.
Most of those advertisers have co.uk domains. I wish I could block in one shot all .co.uk domains as many of those advertisers pass on the traffic to UK search engines at which they get paid higher I guess.
Also I notice on the German pages we have english advertisers and on the English pages we have German and Spanish advertisers.
Both Amazon and my affiliate providers (hotel reservatioins) have for the first time outperformed Adsense in three straight days and I start removing Adsense on 146,000 pages.
I will still keep it on our Japanese pages as EPC for one click on our Japanese content is so much higher compared to the English content pages which I feel is getting spammed by low bids coming from the UK.
Europeforvisitor - I visited yesterday your website and notice that you too have many of those kind of advertisers with the .co.uk domain. If I were you, block them.
Many of those sites are shoppingmalls and they pass on those traffic to another affiliate system that handles online reservations and I guess they walk away with 5% per transactions. (Not bad for them for a couple of cents per visitor)
We will become from next Monday an Adwords client as I smell big opportunties. Maybe after all that's the way to go.
Anyone who knows how little an Amazon associate makes will say that it's a sad day when that affiliate program outstrips AdSense.
--
I just feel the qualities of adverisers dropped and I am ended up sometimes with the last trash available.
We have 600,000 pages of content and I am ony able to block 200 ads.
2) The discounts that Google is applying to the advertisers' bids for those ads on different types of content pages.
europeforvisitors,
thank you for your feedback. Yes, "covergence" of the % discount was what I meant to say, in fact I now know that AdWords advertisers can bid seperately for content vs search.
So if an advertiser has a $1 bid for a keyword for content sites, Google would then apply a variable discount, based on its feedback on how well a referral from each site converts.
With the availability of Adsense tracking scripts (just installed one), it's now possible to have valuable insight how it works for your particular sector.
Btw I think that what we're seeing today in terms of # of Adwords advertisers is just the tip of the iceberg. Juding from what I've seen sofar there are still very few advertisers. You'll find plentry of ads for free hosting or software etc products, but very few ads in other sectors.
I believe the # of advertisers is going to increase exponentially, as more businesses realise it could be THEIR OWN AD sitting there on the site they're viewing. The little "ads by Google" link on top of every Adsense block will take care of that.
And I think that conventional mass media (tv, printed press etc) are going to have a VERY HARD time in the next few years, trying to compete with personalised advertising delivered to millions of users. For all demographic groups where the internet is used.
There are some problematic issues in Adsense that I find it hard to believe that Google hasn't fixed yet:
1) Ability for the publsher to tell Adsense what a page is all about, so to avoid un-targeted ads. Ability to use negative keywords.
2) Get rid of the HUGE image in the Adwords conversion tracking script, so that more Adwords advertisers use it. It's distracting, it has no place there. Google DOESN'T NEED more exposure, ANYONE who wants to join Adwords can click the "ads by Google". So "fair" % discount on content CPC bid is given after a while, rather than Google trying to GUESS which pages convert better. If a publisher doesn't use "conversion tracking" then % discount in CPC should be minimal.
3) Better tech support. I'm thinking about e.g. non-latin charset ads which appear garbled most of the time on content sites. I still see this phenomenon very often and it's VERY annoying and makes Adsense ads overall look very unprofessional when it happens.
So if an advertiser has a $1 bid for a keyword for content sites, Google would then apply a variable discount, based on its feedback on how well a referral from each site converts.
Conversion may be one factor in calculating the advertiser's discount, but I wonder if it's tracked by page or site, or by the type of content (forums vs. editorial pages vs. DomainPark, etc.). From what Google told publishers in an e-mail at the beginning of April, tracking by type of content seems most likely--at least for now. I don't don't think site- or page-level tracking and adjustment will be practical untel more advertisers use conversion tracking--and in any case, the sheer number of publishers in the AdSense network would create enormous overhead if Google had to track conversion data for thousands of sites and calculate advertiser discounts in real time. (Right now, Google has enough trouble just serving relevant ads and keeping publisher reports upto date!)
Btw I think that what we're seeing today in terms of # of Adwords advertisers is just the tip of the iceberg. Juding from what I've seen sofar there are still very few advertisers. You'll find plentry of ads for free hosting or software etc products, but very few ads in other sectors.
I think it depends on the sector. There are many, many advertisers in the travel sector, for example, and they range from major airlines and travel agencies like Orbitz and Expedia to bed-and-breakfasts and vacation cottages. Average CPC and EPC are pretty decent, too (nothing like the pennies per click that publishers in some sectors are complaining about).
I believe the # of advertisers is going to increase exponentially, as more businesses realise it could be THEIR OWN AD sitting there on the site they're viewing. The little "ads by Google" link on top of every Adsense block will take care of that.
I'm sure you're right. I also believe AdWords/AdSense will be used by more advertisers when it becomes easier for businesses to find agencies or media-buying services that can handle their PPC advertising chores.
And I think that conventional mass media (tv, printed press etc) are going to have a VERY HARD time in the next few years, trying to compete with personalised advertising delivered to millions of users. For all demographic groups where the internet is used.
Sales promotion, direct marketing, loyalty marketing, etc. have already taken a big chunk out of traditional advertising budgets. Networks like Overture and AdWords/AdSense are just following in that tradition. Still, personalized direct-response ads aren't the most efficient medium for every product or service, and mass-market advertising is still going to be more coft-effective if you're selling beer or laundry detergent or cars.
There are some problematic issues in Adsense that I find it hard to believe that Google hasn't fixed yet
Google is a young company staffed by young, bright people who may have a tendency to bite off more than they can chew. That's been fairly apparent on the search side; why should be the contextual advertising side of things be any different?
1) Ability for the publsher to tell Adsense what a page is all about, so to avoid un-targeted ads. Ability to use negative keywords.
Google obviously trusts its algorithms more than it trusts publishers. Such trust is often misplaced, to judge from the Iowa and Wyoming ads that I keep seeing on a page about Venice, Italy. :-)
So "fair" % discount on content CPC bid is given after a while, rather than Google trying to GUESS which pages convert better. If a publisher doesn't use "conversion tracking" then % discount in CPC should be minimal.
You mean "if an advertiser doesn't use conversion tracking," don't you? (I assume that was a typing mistake; if not, I've got to look at my AdSense control panel more carefully!)
In any case, I don't think Google is necessarily wrong in assigning different values to different types of content. If conversion-tracking statistics indicate that, on average, clicks from forums convert at X% of the rate of clicks from text pages, is it reasonable for Google to charge advertisers the same price for each? To put it another way, if the price per click doesn't reflect the value to the advertiser, the advertiser isn't going to use content ads.
3) Better tech support. I'm thinking about e.g. non-latin charset ads which appear garbled most of the time on content sites. I still see this phenomenon very often and it's VERY annoying and makes Adsense ads overall look very unprofessional when it happens.
True, but it's a lot cheaper to have a key that's programmed with a macro that reads: "We've looked at your pages, and we're unable to replicate the problem." :-)
You mean "if an advertiser doesn't use conversion tracking," don't you? (I assume that was a typing mistake; if not, I've got to look at my AdSense control panel more carefully!)
Yes, ofcourse I meant "advertiser", not "publisher" (content-site owner).
In any case, I don't think Google is necessarily wrong in assigning different values to different types of content. If conversion-tracking statistics indicate that, on average, clicks from forums convert at X% of the rate of clicks from text pages, is it reasonable for Google to charge advertisers the same price for each?
Yes, imo different EPC per content site is the only way to go, as long as Google is very liberal in letting webmasters join Adsense (which is fine imo). The problem is how to differentiate among content sites OBJECTIVELY. PageRank means nothing in that case.
Personally I don't mind a pay-per-conversion thing, whereas conversion would be e.g. PagesViewed > 3 or duration of visit > X min etc.
I don't even think that DomainPark pages should be part of "content sites" network. This is "accidental visits" stuff.
I guess click "value" in the case of forums, can vary a lot, depending on how targeted the forum is. In most cases I wouldn't use Adsense on a forum page though...
To put it another way, if the price per click doesn't reflect the value to the advertiser, the advertiser isn't going to use content ads.
Agreed, and I can understand the reasons. Since Adwords doesn't allow the advertiser to decide WHICH content sites to run his ads on (and I doubt it will ever will, or some big advertisers could do direct deals), many will just refrain from ALL content sites.
Since Adwords doesn't allow the advertiser to decide WHICH content sites to run his ads on (and I doubt it will ever will, or some big advertisers could do direct deals), many will just refrain from ALL content sites.
Yes, and understandably so.
I'm not so sure that letting advertisers control where their ads appear would result in big advertisers trying to do direct deals, though. I think this concern is overblown for two reasons:
First, big advertisers can already do direct deals with big sites, and it isn't cost-effective for them to do direct deals with smaller sites.
Second (and far more important), AdSense ads are targeted by page, not just by site.
For advertisers, targeting by page means the ability to reach niche audiences on many different sites. Google finds relevant pages, aggregates them, and lets the advertiser buy leads from all those targeted pages with one purchase.
For publishers, targeting by page means the ability to earn higher effective CPMs than they'd receive if they were selling untargeted ads. It also means they can monetize pages that don't generate enough traffic to interest advertisers.
In other words, Google's technology makes AdSense attractive to both the advertiser and the publisher even if they know each other's identity.
Assuming that EFV is somewhere in Europe (a leap, I admit) he would not see ads that advertisers have specifically targeted to U.S. viewers. The number of examples we could come up with along this line is nearly infinite.
On one of my sites I regularly get complaints from readers steamed by ads that I have never seen. (These tend to be ads for the very products that are pilloried on the page in question, not anything off-color or objectionable).
It's not just geo-targeting, of course, but also the luck of the draw. Ads appear as often as their budget and average click-through rate indicates they should.
About two days after joining AdSense last June I gave up trying to keep track of which ads appeared where, having come to the conclusion that the small slice of the universe visible to me was not sufficient to draw any conclusions.
Further, despite the view often expressed here, highly-targeted sites may be at a disadvantage as industries tend to wax and wane because of patterns unique to them. My largest AdSense site covers topics from automobiles to washing machines (no zoos, sorry). Contrary to what I often read here (that general-interest sites don't have a high CTR), my CTR is quite healthy compared to others hinted at here.
Covering lots of niche topics (which is, after all, what a general-interest site does) is an advantage in that it tends to even out the peaks and valleys -- when curling irons are cool, refrigerators are hot, etc.
I've used the extra revenue from AdSense to beef up my editorial product, increase promotion and add server capacity to handle the increased traffic that's resulted. To me, AdSense is the original "black box" -- I don't know what goes on inside the box and as long as it works reasonably well, I don't mess with it.
Assuming that EFV is somewhere in Europe (a leap, I admit) he would not see ads that advertisers have specifically targeted to U.S. viewers. The number of examples we could come up with along this line is nearly infinite.
Yes, that's a leap, and in any case it's irrelevant because my site has an international audience. Also, an ad for "Denver car broker" or "Iowa getaways" on a page about European travel is off-topic no matter where it's being viewed. I can't speak for others whose sites are showing mistargeted ads, but my site has plenty of relevant "theme" ads that can be used in cases where Google can't find a suitable ad. (Besides, aren't PSAs supposed to be displayed when a suitable ad isn't available?)
Another example: if you are one of 10,000 Blue Widget Trimmer sites, there may not be enough big-bucks advertisers whose budgets will support serving ads to every single visitor to each Blue Widget site. Most AdWords advertisers set fairly modest daily budgets. AdSense deduces how many ads it can serve to hit but not exceed the target. The rest of the impressions get something else. Sometimes that "something else" is close to on-target but not always.
I think we all see occasional examples, like the Denver car broker, that are obvious errors but we don't necessarily see all the spot-on selections.
My only point is that the ads you see are not necessarily representative of the total inventory being served to your visitors.
My only point is that the ads you see are not necessarily representative of the total inventory being served to your visitors.
I think we all know that. But the mistargeted ads are certainly representative of Google's mistakes, and of a problem that needs to be solved.
On his Oceania page, I was expecting something from my part of the world, but no, the ads are well targetted.
My conclusion is that most ads (that I have seen) are on target and that some will occasionally drop on the wrong pages, or the advertiser has deliberately targeted the wrong page with an off-topic keyword or country selection.
I saw some ads in Japanese (guessed from the URL) on his Florence page, but it was a hotel in Florence.
My conclusion is that most ads (that I have seen) are on target and that some will occasionally drop on the wrong pages, or the advertiser has deliberately targeted the wrong page with an off-topic keyword or country selection.
Occasional mistargeting is one thing, but the ads on some pages are consistently mistargeted. And it isn't just because one or two advertisers have deliberately targeted the wrong page or keyword: If the mistargeted ad is blocked, another mistargeted ad takes its place. In such cases, it would be helpful if the publisher could provide "helper" keywords to override the mistargeting. This may not be an issue for publishers who have forums, news sites, etc. where content is constantly changing, but it is a problem for those of us with "evergreen" pages that draw a reasonable amount of traffic.
