homepage Welcome to WebmasterWorld Guest from 54.234.2.94
register, free tools, login, search, pro membership, help, library, announcements, recent posts, open posts,
Pubcon Platinum Sponsor 2014
Visit PubCon.com
Home / Forums Index / Google / Google AdSense
Forum Library, Charter, Moderators: incrediBILL & jatar k & martinibuster

Google AdSense Forum

    
If Adsense is Based on Price,
With Least Paying Ads at Bottom, so why is an MFA on Top?
Member02




msg:1340047
 3:45 am on Apr 14, 2006 (gmt 0)

If an adsense ad shows up on your site doesn’t it mean that it beat its competitors to occupy a high space on your site?

I’m confused, what I know is an obvious MFA site, it happens to be at the top of my skyscraper, why is this?

Isn’t adsense based on price, the least paying ads at the bottom, so why is this MFA ad at the top?

Does this mean it’s an MFA site that is paying well? Which I doubt!

 

martinibuster




msg:1340048
 3:57 am on Apr 14, 2006 (gmt 0)

AdSense shows ads based on their ability to pay better. So an ad that they estimate will be clicked twenty times at five cents per click per thousand impressions (one dollar) is worth more than an ad in which the advertiser is bidding ten cents per click but will only be clicked on five times per 1,000 impressions (fifty cents).

AdSense is not really based on price alone.

Member02




msg:1340049
 4:04 am on Apr 14, 2006 (gmt 0)

So in your opinion (or anyone else's), should I filter out this MFA site or not?

It has the URL too in the ad, so I typed the URL in my browser and visited the site. I was shocked at just how much of an MFA it was.

But if like you said, Google has decided to place it at the top of my skyscraper then I should leave it alone?

martinibuster




msg:1340050
 5:34 am on Apr 14, 2006 (gmt 0)

That is a polarizing question. Some people will tell you to block them because they feel it's a quality issue or that they don't pay as much and you'll end making less money. Others will tell you that Google knows best and to let the dog wag the tail and not vice-versa.

Because generally there are naturally occurring wide variances in earnings during the course of a month, as well as month to month, it can be difficult to tell with 100% certainty if blocking MFAs is effective. Although doing tests that last at least a week, if not a month is probably the best.

Also, there can be so many advertisers in the pool, the ads you see on your site are just a fraction of what's being shown across the USA, and the world. You will never see a complete list of who is advertising on your site. You can never know all of them, the information is not available, not even through the preview tool.

Here is some related reading:

An Explanation of Variations in Earnings
A fresh look at the ups and downs of AdSense earnings, a discussion of what we should expect as a normal fluctuation, and how to spot an abnormal deviation from the normal fluctuations in earnings
[webmasterworld.com...]

Blocking Advertisers Waste of Time
seen em round the world
[webmasterworld.com...]

david_uk




msg:1340051
 7:05 am on Apr 14, 2006 (gmt 0)

Remove it - beyond a shadow of a doubt.

"Ability to pay better" is a totally misleading phrase, and I'm surprised the moderator chose to use it. You have to ask yourself who is likely to pay more - a genuine advertiser selling products direct to end consumers, or a site that has nothing to sell but lives off trying to funnel punters through their site by buying cheap clicks and passing you through to expensive clicks from real advertisers. It's a no brainer isn't it?

The target bot takes many factors into account when deciding what ad to place, and Google tries to guess what ads will result in the most money for them (not you). Past performance of ads on the Google network in general is a major part of this, and this is not based in any way on how the ads have performed historically on YOUR site.

I started blocking MFA's last July on the grounds that if I allowed the graph of earnings to dive at the rate it was, in two months time I'd be paying Google for ads on my site. I simply didn't believe that the MFA's I was seeing were likely to represent the best income. They had nothing to sell - how could they? I blocked them, and have not looked back since. The immediate effect was lower ctr, but a MUCH HIGHER epc, so my income started to recover immediately. The long term effect of blocking ads is that the good ads I see on my site pay well, and smart pricing likes that. The last 6 months or so I've had an extremely stable EPC as a result, and earnings are up by a great deal since blocking started.

I recently conducted an experiment whereby I took out my entire block list to see what happened when the target bot was allowed free reign. The experiment was not to see what happened financially, but to see if Google's targeting had improved. What happened was that the advertisers I was used to seeing that do sell products and services one by one dissapeared and were replaced by MFA's. The first two to appear were the ones that started me blocking in the first place.

So there are major faults in the algorithms - they should NOT place MFA's where genuine advertisers are paying for placement. What you earn has no relationship to what the advertiser pays. It may well be that as smart pricing zaps your earnings on clicks, the advertiser (including MFA's) is still paying a hefty whack for the clicks with Google keeping all of the profit.

As Google are likely to profit from MFA's (thanks to smart pricing) they will keep giving them preference over genuine ads thay can't smart price so much. Do not let Google's target bot have free reign - it will cost you!

milanmk




msg:1340052
 7:29 am on Apr 14, 2006 (gmt 0)

I will go with david_uk. Reading much about MFA and their low paying clicks I decided to block around 10 such sites, resulting in around 200 percent increment in CPM and 50 percent increase in CTR. Although it has been just a while that I have implemented this but I am sure it is due to blocking of MFA’s and my total earnings are gradually on rise since then.

Milan

martinibuster




msg:1340053
 7:33 am on Apr 14, 2006 (gmt 0)

You have to ask yourself who is likely to pay more - a genuine advertiser selling products direct to end consumers, or a site that has nothing to sell but lives off trying to funnel punters through their site by buying cheap clicks and passing you through to expensive clicks from real advertisers. It's a no brainer isn't it?

David, I respect your opinion very much, and I'm not saying you are wrong, so correct me if I'm wrong, but by "pay more" it seems you may be confusing Earnings Per Click (EPC) with Earnings. A high EPC is not an indicator of how much you will ultimately earn.

I understand your point that an ad from an actual retailer (and even an affiliate site) may convert and lead to a sale, and thus be worth more as a click according to Smart Pricing, the issue of earnings still has to be considered.

So should we believe that Google will do right by us and show the best revenue producing ad? Or are we being shafted in favor of the advertisers when Google asks us to trust them that the highest revenue producing ads are being shown?

Leaving aside arguments about quality, let us focus on earnings, i.e., how much money goes into the bank and pays the bills.

An ad that pays 5 cents per click and is likely to be clicked twenty times ($1 earnings) will earn you more than a twenty cents per click ad that will be clicked on once or twice ($0.20 -$0.40 earnings.)

Ads only show if they represent the highest earning potential.
Well, that's what Google says, and without seeing the actual data on a site per site basis, you have to ultimately decide whether to believe Google or not. And I think that's the crux of where the decision has to be made with regard to blocking or not blocking.

So what does Google say? ASA pointed out over here re Google Ads appearing on publisher sites:
[webmasterworld.com...]

they should only appear on your site if they represent the highest revenue potential for that particular ad unit at that moment in time.

And over here, in relation to CPM Ads:
[webmasterworld.com...]

They still have to compete against all other ads in your site inventory, including CPC ads. The ads determined to hold the highest revenue potential will appear on your site.

Now let's take a peek at the AdSense Blog and see what Google [adsense.blogspot.com] says about it:

Don't believe the myth about blocking 'low-paying advertisers'
Our auction system automatically selects the best performing ads for each page to help you earn the most possible money.

And right after that they say this:
Don't go overboard with filters
Only filter URLs when absolutely necessary. As mentioned, each URL you add to your filter list may cut into your AdSense revenue potential, so it's important to think carefully before deciding what to block.

I think it's important to lay the facts down and let people make an informed decision before running off to do what others are doing.

[edited by: martinibuster at 8:24 am (utc) on April 14, 2006]

david_uk




msg:1340054
 8:20 am on Apr 14, 2006 (gmt 0)

By "pay more" you are confusing Earnings Per Click (EPC) with Earnings. A high EPC is not an indicator of how much you will ultimately earn.

Nope - I'm not. It's been my experience that blocking MFA's increases BOTH EPC and bottom line $$. My average EPC hovers between the male and female UK retirement ages (in cents) - can an MFA possibly afford to pay that before smartpricing? Real advertisers can, and do.

Yes, my bottom line cash figure is greatly improved. it's a shame we are not allowed to post charts here, as the effects of blocking are dramatic.

I would also like to mention the long term effect on smart pricing on blocking MFA's. OK - it's not a fact I can prove, and I'm not saying it is, but there are few that would disbelieve that having quality ads that pay well on your site is going to have a long term negative effect on smartpricing.

Leaving aside arguments about quality, let us focus on earnings, i.e., how much money goes into the bank and pays the bills.

Believe me, I do! Adsense replaces my wife's job so the most important figure is bottom line cash. having a great epc but few clicks doesn't pay for kids shoes - bottom line cash figure does.

An ad that pays 5 cents per click and is likely to be clicked twenty times ($1 earnings) will earn you more than a twenty cents per click ad that will be clicked on once or twice ($0.20 -$0.40 earnings.)

I'm aware of the official Google line, but I can only believe there is any truth in it when I see evidence that the algorithms do indeed give you maximum earning when allowed freedom to place what they want. That hasn't been my experience. Google's supposed targeting has cost me money. It seems to me that the majority of people who have tried blocking MFA's targeting have said it increases the bottom line figure. OK - some of them report that it's not possible to continually block all of them and give up due to lack of tools from Google, but the result is still the same - blocking increases bottom line $.

Ads only show if they represent the highest earning potential.

They should only appear on your site if they represent the highest revenue potential for that particular ad unit at that moment in time.

They still have to compete against all other ads in your site inventory, including CPC ads. The ads determined to hold the highest revenue potential will appear on your site.

Again, quoting the official Google line. Again, webmasters need to see this actually working in practice to believe it. When you have many people experiencing the exact opposite, it starts to appear to be a false claim.

Now let's take a peek at the AdSense Blog and see what Google says about it:

Don't believe the myth about blocking 'low-paying advertisers'
Our auction system automatically selects the best performing ads for each page to help you earn the most possible money.

Yes, I can read what Google say about adsense, but we webmasters need to see this work in practice. If it did, then MFA's would NOT replace well paying as as they undoubtedly do. Google is not a god-like eminence that we simply have to believe unquestioningly, it's a worldwide money making machine, and any claims it makes need to be backed up with evidence for increasingly skeptical publishers. They aren't doing this.

And right after that they say this:
Don't go overboard with filters
Only filter URLs when absolutely necessary. As mentioned, each URL you add to your filter list may cut into your AdSense revenue potential, so it's important to think carefully before deciding what to block.

So far the only thing I can agree with. If you read my posts, I'm against random blocking based on who you percieve to be a poor payer, against using the preview tool to block ads that probably never will be seen on your site, and against blocking simply because the landing page carries ads itself and so on. I'm also against blocking competitors simply because they are competitors.

I believe that there is a sensible way on blocking, and in past discussions ASA has said that using the competitive ad filter to block MFA's is perfectly legitimate as long as webmasters doing this have some mechanism to measure the changes made by the blocking to be sure they aren't losing money on it.

I think it's important to lay the facts down and let people make an informed decision before running off to do what others are doing.

Yes, it's important that people are aware of Google's official line, but it's also important that people are aware that in many people's experience it doesn't hold water - these are also facts.

foxtunes




msg:1340055
 9:25 am on Apr 14, 2006 (gmt 0)

After scratching my head looking at an increased ctr but ever decreasing epc I decided to block. Most of the MFAs milking me for 3 cent clicks are gone - for the moment. Companies advertising real products are back - income up 35 percent.

The only problem is the 200 filter limit, in competitive sectors that will quickly get filled with bottom feeders.

The main reason these mfas appear above more high paying companies is their abilty to write snappy titles with a high ctr. However these titles are almost always deceptive.

"Incredible deals on cheap widgets here!"
"Your comprehensive guide to cheap widgets!
"Arthur, look no more - deals on the grail here!"
"Smeagol - Click for cheap one ring"

Clicking these invariably leads to a page with two large rectangles side by side, an adlink menu bar. An attempt at real content consisting of a few scraped links, and a leaderboard at the bottom.

21_blue




msg:1340056
 11:00 am on Apr 14, 2006 (gmt 0)

Member02 wrote:
why is this MFA ad at the top?

Welcome to WW, Member02.

The question you ask does not, unfortunately, have a definitive answer, so most responses are going to be based on opinion rather than fact. I'll offer my opinion, and you can then take your choice from the ones offered.

The MFA is at the top, probably, because it has a very high CTR compared to normal ads. My theory is - and this is only a theory - that there is sometimes a chain of events on a site, similar to the spread of SARS or bird flu in a country if unchecked:

  • the MFA manages to enter your site through a poor performing page (this is because conventional ads do so badly on that page, that they are smartpriced down to a level that MFAs can compete with).
  • once there, your site-wide smartprice starts to decline even further, because MFA ads probably have inordinately low conversion rates
  • the adsense algo then sees the eCPM of other slightly-less bad pages declining; so, as the MFAs have higher eCPM (due to high CTR), it tries the MFAs on those other pages (that were previously performing better), which hits your site-wide smartprice further
  • eventually, MFAs spread from the lowest performing pages upwards; they become the dominant ads because of the combined impact of smartprice and the MFA's high CTR sucking poor performing pages into their reach, a bit like a black hole gobbling up the universe (apologies if I'm using too many metaphors!)

To deal with this, imho you have two choices:

  1. Block the MFA ads.
  2. Get the CTR of conventional ads up so that they displace the MFAs and drive them off your site

Blocking ads will work, but they are difficult to find (as incrediBill's thread that martinibuster suggested will explain), it is something that you have to keep doing, and you'll quickly run up against the limit of the block list.

To get the CTR of conventional ads up, you need to:

  • block ads, but only in the short term
  • identify poor performing pages (based on a depressed EPC relative to Adwords rates or EPC of other pages on your site in that niche)
  • temporarily remove ads from those poor performing pages altogether (ie close the back door by which MFAs enter the site).
  • optimise your other pages, eg: by cutting down on the number of ads and repositioning above the fold, to make sure that conventional, high-paying ads get a better CTR (this puts their eCPM above the bid rates that allow MFAs onto the site).
  • when you are MFA free, you can experiment with reintroducing ads to some pages on a limited basis, though if you are getting good income it may be safer to leave them ad-free and just put links to better-performing pages.

The reason for reducing the number of ads is to avoid the situation where MFAs can creep onto the page in the bottom slots - you want to have less ad units than there are well-paying ads.

Some other stuff I've written that may help explain how I've come to this theory is a thread on Poison Words [webmasterworld.com] and the recent one on EPC based strategy [webmasterworld.com]

david_uk




msg:1340057
 11:50 am on Apr 14, 2006 (gmt 0)

I haven't thought of likening containing MFA's do controlling a disease, but I like the analogy. I think Anne or Annej recently likened removing MFA's to picking off fleas! :)

I think you raise some very valid points here. Maybe one of the reasons I don't find blocking difficult is that I have a minimum amount of ad blocks anyway. I've long since been removing under-performing units because they - well - under-perform! The purpose of this was to maintain epc, and reduce ad blindness to hopefully increase clicks. But if it's had the effect of cutting down on the chances of being "infected" by the plague of MFA's by the back door as well then that's great!

If maintaining a high epc by keeping down the number of poorly performing ad blocks has the effect of raising the click price above the limit MFA's are able to pay, then this strategy has the potential to be more effective than blocking is.

I appreciate that like a lot of discussions here it's speculation, but it's certainly worth trying. I'd like to see some experimentation on this, but have to say that it would be pretty well impossible to actually prove the theory due to the large number of other factors that affect adsense. But we do know that removing blocks can cause income to increase, so whatever the effect on MFA's it's good advice.

[edited by: david_uk at 12:10 pm (utc) on April 14, 2006]

milanmk




msg:1340058
 11:51 am on Apr 14, 2006 (gmt 0)

An ad that pays 5 cents per click and is likely to be clicked twenty times ($1 earnings) will earn you more than a twenty cents per click ad that will be clicked on once or twice ($0.20 -$0.40 earnings.)

Any basis behind this assumption? Why it is not possible that an ad giving twenty cents per click will be clicked on once or twice or thrice or twenty times? If at all it does have just the half CTR then $0.20 x 10 = $2, almost double earnings!

My question being how come anyone says that MFA will be having high CTR and other conventional ads will be having low CTR? Nobody knows what exactly is going on in a user’s mind when they click on Google Ads; do they know that this click will earn the webmaster of that site a penny or whether they are smart enough to see the title and url for themselves and choose whether to click or not? If quality is to be talked about then don’t you think the user who clicked on the MFA ad will be having “bad image” about your site as they didn’t got what they were looking for? What about returning users who will be irked by those banners and scrappers as last time he visited your site he had clicked on one of those MFA ads and have created a bad impression about those ad units on your site? I presume that if anyone is coming to your site finding some information or tools or products then they surely have some basic knowledge of internet and are not that dumb that they will unreasonably click on ir/relevant contextual ads been showed on your pages.

All this is solely dependent on individual sites and their user base. The only practical thing one can do is to block those MFA’s (taking tips form 21_blue’s post) for some definite period and see for themselves whether it has been for good or bad.

Any suggestions/corrections are more than welcomed about this post.

Milan

martinibuster




msg:1340059
 1:43 pm on Apr 14, 2006 (gmt 0)

My question being how come anyone says that MFA will be having high CTR and other conventional ads will be having low CTR? Nobody knows what exactly is going on in a user’s mind when they click on Google Ads;

Geez.Read the post again.

Here, maybe this will help:

Google's algorithm is determining what ads that are likely to earn more will be shown

The passage you quoted (An ad that pays 5 cents per click and is likely to be clicked twenty times) is a hypothetical EXAMPLE.

If you're seizing on a hypothetical example and quibbling over random numbers I pulled from the air for the sake of example, I think maybe this discussion might be over your head. ;)

21_blue




msg:1340060
 2:42 pm on Apr 14, 2006 (gmt 0)

martinibuster, I think milanmk just misread the post; being from India, English may be a 2nd or 3rd language for milanmk. I sometimes misread posts, and English is my 1st and only language! However, milanmk asks a valid question:
My question being how come anyone says that MFA will be having high CTR and other conventional ads will be having low CTR?

I don't think anyone is saying it will always happen, Milanmk. Rather, the point is that sometimes it happens.

The reason MFAs have sometimes have a high CTR is because they use wording that is attractive but sometimes dishonest or misleading - eg: "learn how to be a millionaire without having to do any work". The reason conventional ads sometimes have a low CTR is that they often have unattractive wording, or the placement on the page by the publisher is poor, or there are simply too many adverts on the page, or the ads are below the fold.

The solution, as illustrated by your own example, is to improve the CTR of conventional ads. As a publisher, you can do this by:

  • changing the placement on the page (eg: to a hotspot)
  • reducing the number of adverts on the page, or
  • making sure all the ads are above the fold
Then the CTR of conventional ads will improve and price the MFAs off the page.
milanmk




msg:1340061
 2:48 pm on Apr 14, 2006 (gmt 0)

I apologize for seizing on a hypothetical example and quibbling over random numbers and maybe this discussion might be over my head…but my earnings are continuously on rise after I banned just 10 MFA sites without doing any other SEO changes.

I just know that customer is a king and I do almost everything to make them feel good about my site and maintain a loyal relationship. Because of them my site is in progression and if I have to change anything on my site to maintain that then I will surely do it right away, including banning those MFA sites.

If anyone is more interested in just gaining dollars from those ads then definitely I am out of it. I have always learnt from WebmasterWorld that always think about your users first and then your earnings.

Edit: 21_blue, I did all the things you mentioned to improve the CTR of conventional ads but my site was completely been ruled by MFA’s and finally I had to block them all.

Milan

[edited by: milanmk at 2:59 pm (utc) on April 14, 2006]

david_uk




msg:1340062
 2:56 pm on Apr 14, 2006 (gmt 0)

My question being how come anyone says that MFA will be having high CTR and other conventional ads will be having low CTR?

I wouldn't like to say that it is a universal pattern, but I do have to say that when I started to block MFA's, ctr did fall to about a third of the previous level. Regarding the ads on my site, I think the reason they had a good ctr was purely ad copy. "You have to read this before you buy a widget!", "Free information on green widgets" and so on are much more enticing than some of the ads that are (quite frankly) dull. I think genuine advertisers are often really poor at writing good, sexy ad copy unfortunately.

As an advertiser, I've tried both boring ads, and sexy ads to see which worked best, and the sexy ones had a much better ctr.

To add yet another perspective to the mix, when I recently stopped blocking to see what happened, overall ctr went down on this occasion. Therefore I'm undecided on if there is a strong link between MFA's and patterns in ctr.

21_blue




msg:1340063
 3:49 pm on Apr 14, 2006 (gmt 0)

milanmk wrote:
Edit: 21_blue, I did all the things you mentioned to improve the CTR of conventional ads but my site was completely been ruled by MFA’s and finally I had to block them all.

Just to clarify, if you already have MFAs on your site then, even using my EPC-based strategy, the first step is to block them (temporarily). After you have blocked the current MFAs, you should do the things I suggested to stop more MFAs invading your site in the future. Purveyors of MFAs keep coming up with new URLs so a blocking-only strategy means you'll be constantly chasing them.

So, what I'm sayin is: block, then improve/optimise the ads, then you can remove the blocks a couple of months later. Don't just block.

milanmk




msg:1340064
 4:03 pm on Apr 14, 2006 (gmt 0)

Got your point and thanks for clarifying it. I have already removed ads from poor performing pages and kept only one ad unit above the fold.

Milan

rbacal




msg:1340065
 4:18 pm on Apr 14, 2006 (gmt 0)

Quick comment I haven't seen mentioned (good thread!)

Two other considerations: What is the effect of MFA ads on the probability that a "real" ad -- something that pays higher, will get clicked? If the MFA ad is clicked because it's at the top, you lose the visitor, and therefore the chance he or she will click a more lucrative ad (I know they can use the back button, but it's just an example).

Second is the issue of what you are willing to lose a visitor for. a cent, two cents, 50 cents...etc. This gets important if you have other revenue generation strategies on your site. I don't want to lose a visitor (who may purchase something, or click something of value later in their page browsing) for pennies.

So, it's not just about "visible" income. 100 clicks at 1 cent are NOT the same, business-wise as 20 clicks at 5 cents, and so on. The visible revenue is identical, but that's just a superficial view (at least for some sites).

21_blue




msg:1340066
 4:40 pm on Apr 14, 2006 (gmt 0)

milanmk, can you let us know in a couple of weeks, and after Easter fluctuations, how your earnings go (up, down, same) after making these changes.

milanmk




msg:1340067
 4:52 pm on Apr 14, 2006 (gmt 0)

Definitely. Even I am quite hopeful to see the cumulative results from couple of weeks of experimental blocking and do the needful changes according to your theory.

Milan

david_uk




msg:1340068
 4:54 pm on Apr 14, 2006 (gmt 0)

Rbacal - I've been using a tracker on and off for a while to see what ads are clicked. it's not on at the moment because it takes time to load, and doesn't record what links are clicked on adlinks pages.

I have noticed that when people click genuine ads, the chances of them coming back are high. The visitors who had the biggest count of pages loaded were the ones that clicked on ads. When I ran my recent test of not blocking, one of the patterns I noticed that overall there was no appreciable difference between those who clicked ads and those who didn't for page load count. The figure was reduced by those who clicked on MFA's not coming back.

So I agree that you have to consider if it's worth losing a visitor for a 1c click.

birdstuff




msg:1340069
 7:10 pm on Apr 14, 2006 (gmt 0)

I don't waste my valuable time blocking advertisers. I prefer spending it on tasks that actually end up making me money for the long term.

david_uk




msg:1340070
 8:58 pm on Apr 14, 2006 (gmt 0)

So ten minutes a week to double your income isn't worth your time?

fischermx




msg:1340071
 6:33 pm on Apr 18, 2006 (gmt 0)

Blocking MFA FAQ.

Fact:
Blocking MFA does increase overall income. Period.

Why should I block them? They pay less.

Why sometimes they are on top of real business? Because many times they have higher CTR due a good creative, which Google prize with higher position or lower CPC.

If I block one, will that space be used by a lower paying ad? No, many times you'll get higher paying clicks that usually don't show due low CTR.

Should I block them? Yes.

What will happen to my account? You may see lower CTR, but you'll get higher EPC, CPM and overall income.

Really, should I block them? Yes.

Global Options:
 top home search open messages active posts  
 

Home / Forums Index / Google / Google AdSense
rss feed

All trademarks and copyrights held by respective owners. Member comments are owned by the poster.
Home ¦ Free Tools ¦ Terms of Service ¦ Privacy Policy ¦ Report Problem ¦ About ¦ Library ¦ Newsletter
WebmasterWorld is a Developer Shed Community owned by Jim Boykin.
© Webmaster World 1996-2014 all rights reserved