homepage Welcome to WebmasterWorld Guest from 54.166.65.9
register, free tools, login, search, pro membership, help, library, announcements, recent posts, open posts,
Become a Pro Member
Visit PubCon.com
Home / Forums Index / Google / Google AdSense
Forum Library, Charter, Moderators: incrediBILL & jatar k & martinibuster

Google AdSense Forum

This 131 message thread spans 5 pages: 131 ( [1] 2 3 4 5 > >     
Anatomy of an EPC Collapse
Building a List of Causes
martinibuster




msg:3271231
 6:14 am on Mar 5, 2007 (gmt 0)

This discussion is NOT about Smart Pricing
There are many reasons for an earnings collapse, and it's important to list them. If you cannot diagnose a problem then you cannot cure it. Smart Pricing is not the only cause, so let's make a list of what causes earnings to dive.

  • Too much inventory
    As more people enter a space, the EPC is going to drop. MFAs in some cases may be a symptom of a niche with inventory that's under strain. MFAs may not be the problem, just the symptom of the underlying problem.

    If there are so many publishers that they're cannibalizing inventory by advertising on each others site, then that niche may likely be suffering from too much inventory.

    Speaking from experience, this can affect an older site, too. The bottom falls out, it's not smart pricing.

  • Wrong idea of what constitutes a normal range.
    Too many newbs complain about having been smart priced because their earnings collapsed after their first couple of months. The base amount they're comparing is not large enough if it's only one or three months. If five months out of six you're earning six cents per click, six cents is normal. Twelve cents per click for a month is a newb advertiser that didn't know his content network was turned on.

    And this can affect older sites too. As inventory expands we're going to see lower and lower benchmarks for what represents an average EPC. Let's face it, the content network plays second best to the search network, and it may always earn significantly less than what advertisers are paying per click for search.

  • Newb advertisers
    Newbs can cause prices to go up and stay up until they run out of money. If no one is stepping in to keep up the pace then your EPC is going to collapse.

  • Savvy advertisers
    Everybody suspects that a contextual click is generally not as good as a search click. Whether that is true or not is besides the point, that's the general perception. As advertisers become smarter they're going to adjust their prices accordingly.

    I suspect that all it takes is one or two bidders dropping their bids to cause a collapse and let the air out of a high EPC gravy train. This is not fact, and I admit it. However, this conclusion comes from anecdotal evidence from watching how the same advertisers dominate certain channels makes me aware of how much my EPC is dependent on these advertisers propping up the cost of bidding on my site.

Okay smarty pants... put on those propeller hats. :)
Let's hear what some of you think may be non-smart pricing reasons why EPC can collapse. I'm certain I haven't listed all the non-smartpricing reasons. What else, apart from Smart Pricing, do you think will cause EPC to collapse?

 

Hobbs




msg:3271240
 6:51 am on Mar 5, 2007 (gmt 0)

There are a number of conspiracy reasons floating around here like earnings cap, Google taking a higher share and such..

One possible reason I can add is being CPM targeted.

I saw overnight a 50% drop in EPC on a stable for years site with too many broad niches to be affected by any of the reasons you are suggesting.

The only explanation is some sort of an algo $crew up or even a manual adjustment, but I am open for suggestions.

Green_Grass




msg:3271249
 7:04 am on Mar 5, 2007 (gmt 0)

Earnings Cap is certainly not a conspiracy theory. It makes eminent sense and can be related to an algo scr*w up/MFA control/Smartpricing. Ofcouse not all agree and G will never clarify. I for one have seen enough evidence on my new site to believe in this 'conspiracy' theory.

incrediBILL




msg:3271255
 7:18 am on Mar 5, 2007 (gmt 0)

Earnings Cap is certainly not a conspiracy theory.

That's true, it's a defeatist theory.

Earnings are factors of traffic, CTR and CPC and it's as simple as that. Like MB says, when you supply too much traffic your CPC may go down as you have a surplus of ad inventory.

So far, I've increased traffic massively 2 times and my average CPC has gone up, not down, as more people scramble to outbid others to advertise on my high traffic juggernaut so if I ever hit that mythical cap, I'll let you know.

Hobbs




msg:3271256
 7:21 am on Mar 5, 2007 (gmt 0)

The last 70 clicks just made me $3
That's less than a small fraction of what I've been making for years

Let's differentiate between a slow decay in EPC and earnings VS. an overnight drop and address each separately.

Green_Grass




msg:3271261
 7:46 am on Mar 5, 2007 (gmt 0)

An overnight drop in earnings cannot be explained by any logical hypothesis.

It is something other than logic... Ofcouse 'conspiracy' theories are ruled out..Because G is perfect and they have a transparent and pure business model. ;-)

BTW , I experienced a dramatic drop in EPC during DEC/Jan. Something like 12-14 cents a click to 1-3 cents a click. Then from 1st Of Feb, 2007, everything turned around and the tap was opened. I keep looking for increasing ad inventory/increasing traffic/ higher CTR / decreasing publishers to explain this sudden change, but I can't seem to find them. Maybe I haven't looked hard enough.

Hobbs- I do belive that G does play around with their algos.. and you could have been hit.

Hobbs




msg:3271263
 7:57 am on Mar 5, 2007 (gmt 0)

Yeah, that's exactly what I believe, a "manual" adjustment, coinciding right with the last "un-announced" maintenance.

How does one become 'worthy' again, that's the question ;-)

potentialgeek




msg:3271281
 8:46 am on Mar 5, 2007 (gmt 0)

* A high-end AdWords Ad Campaign ends. You'll see advertisers throw money at online advertising from their ad budget. They don't necessarily spread it thin, because that's not the most aggressive advertising. They bid high to get top spot. So you see weeks or even months with their ad showing up at the top/left side of the ad units. Then their exhaust their budget, and, poof!

* Google "Quality Skewers" the advertisers and they're off to YPN! :/

p/g

P.S. I suspect SERPs make a difference. If your competitors ahead of you have Adsense ads, and they are similar ad units or similarly placed on web pages, or with similar blending/colors, you won't get as many clicks. (They already clicked on the same ads, been there, done that.)

darkmage




msg:3271398
 11:42 am on Mar 5, 2007 (gmt 0)

Some other reasons:

1. Google, and other search engines, change the type of traffic they send you. This not about volume, but different keywords, countries, different reasons for visitors to be at your site.

2. Glitches. I have seen ads slowly (and rapidly) drift out of relevance. Then snap back. It happens.

3. Google is a chaotic system. In a Chaotic system, data is not perfectly dispersed. It groups - both good and bad.

4. When Google is first implemented on a site, it is novel. Over time, regular visitors turn blind to the ads. Then there is a decline.

"Earnings Cap is certainly not a conspiracy theory."
I think point 1 partly explains why people think they are being 'punished' for higher traffic - and start to believe in the idiotic idea of an earnings cap.

"An overnight drop in earnings cannot be explained by any logical hypothesis."
-See point 2 and 3

Hobbs




msg:3271400
 11:53 am on Mar 5, 2007 (gmt 0)

start to believe in the idiotic idea of an earnings cap

I don't like your "Chaotic system" theory, it makes me feel insecure ;-)

While I've outgrown my glass ceiling a long time ago and many times too, I certainly wouldn't fully dismiss any idea or call it idiotic unless I have full access to Google's data and feel like insulting lesser informed fools, so let's just call it foolish instead :-)

steve40




msg:3271414
 12:24 pm on Mar 5, 2007 (gmt 0)

Well here is another possibility

Advertiser sells and is targetting

Green Widgets

Green Widgets CPC 50cents ( good Conversion )

Widgets CPC 5cents ( Poor Conversion )

Scenario
Publisher Site increases traffic through being higher in serps for widgets or lower for green widgets

So G shows Widgets add exactly the same add but now only paying 5 cents

To the publisher all looks exactly the same but due to advertiser campaign overall quality score same adds but 10$ of CPC

Now this may be the only advertiser doing this on green widgets and the others may well be paying more so new range could be from 10% drop to 60% drop and all would look the same as before to the publisher

PS one other thing this could also then effect may be quality score of publishers site ( I.E. less conversions )

Just my ramblings and difficult to fully explain

steve

europeforvisitors




msg:3271450
 1:32 pm on Mar 5, 2007 (gmt 0)

Two other things to consider:

1) In 2006, advertisers were given the ability to bid separately for the search and content networks. This made it easier for advertisers to buy "content clicks" at lower prices.

2) Advertisers now have unlimited domain filters, so they can block sites whose clicks don't convert well for them (or where they'd simply rather not have their ads appear).

And, coming soon:

3) According to THE NEW YORK TIMES and Google, advertisers will be able to "site-target" contextual ads, not just sitewide CPM ads, within a few months.

incrediBILL




msg:3271494
 2:32 pm on Mar 5, 2007 (gmt 0)

3) According to THE NEW YORK TIMES and Google, advertisers will be able to "site-target" contextual ads, not just sitewide CPM ads, within a few months.

Which is going to make me insane for other reasons off the topic.

Give you a for instance as I just had a blow-out with someone that went off the deep end when her ad on my site ranked above her domain name when you search for it in Google. OK, people don't tend to search for domain names unless you're a brand, and I have no control over who Google ranks first, but if you want off the site, fine. Not only are you off the site, you're banned!

Fast forward a week and the same ads for the same site are back on my site in AdSense.

Would anyone like to hazard a guess what came next?

That's right, competitive ad filter, buh bye, adios. CYA!

FWIW, One last thought on earnings caps for the tinfoil crowd to consider is that it also caps Google's earnings and they're all about the almighty Ad$en$s dollar$.

martinibuster




msg:3271743
 6:33 pm on Mar 5, 2007 (gmt 0)

a "manual" adjustment, coinciding right with the last "un-announced" maintenance.

That's interesting. From certain experiences, I've had the feeling that certain circumstances can trigger an automated review. I am certain it wasn't manual, although it felt like it. The purpose was to self-heal the SERPs. Could there be a self-healing mechanism in the AdSense program that can adversely affect you?

If you think certain sites come in for an "adjustment", what factors or criteria would trigger it?

Hobbs




msg:3271783
 7:31 pm on Mar 5, 2007 (gmt 0)

The purpose was to self-heal the SERPs

I'm confused MB, what would the SERPS have to do with an AdSense adjustment?

If you think certain sites come in for an "adjustment", what factors or criteria would trigger it?

Right after that last maintenance, it was as if AdSense lost all records on my pages targeting and started serving totally non related ads all over. Also one of my channels that was the top performer in ctr and earnings stopped displaying ads all together although it has been working for years and targeting was right on..

Then gradually it was as if the targeting started to improve gradually, but that channel never recovered and remained MIA, all ads removed from there now and every few days I test to see if there is any improvement with no success so far.

It still looks like a glitch to me as nothing else changed other than the maintenance.

alephh




msg:3271796
 7:45 pm on Mar 5, 2007 (gmt 0)

Major advertizers leaving adwords/adsense (as reported in the recent news) for a better platforms.

If a major player takes his 10,000$-100,000$ budjet away (from a certain sector), it takes more than four 100$ budjets to balance it out.

Pengi




msg:3271819
 8:17 pm on Mar 5, 2007 (gmt 0)

These have been touched on, but perhaps are worth restating.

1) Weekly, Seasonal and Holiday variations - if you haven't collected data over a large enough range, the first time that one of these effects you it can seem like an earnings collapse.

2) Ad [mis]targeting - if your ads go off target for some reason (as happened to me for a week or two). Click Through Rate will plummet.

fearlessrick




msg:3271847
 8:35 pm on Mar 5, 2007 (gmt 0)

AdSense - for the content network at least - may be suffering from oversaturation of publishers (and MFAs, scrapers, etc.) and some flight of advertisers.

Personally, I ran some numbers and found, to my distress, that I've given up pretty much the best spots on my web site for a mere 3.00/1000 page impressions.

To put this in perspective, imagine paying $60 for a full page ad in a 20,000-circulation newspaper or magazine.

That would mean, if 20 of the say, 32 pages were ads, the publisher would GROSS $1200, which would be far less than just the printing cost. Add rent, utilities, writers, ad sales commissions, design, etc. and you understand the real costs involved. I've been out of the newspaper business for some years, but I recall CPM having a baseline of roughly $30, not $3.

Sure, the production costs are different between the web and print, but the only different costs are printing and distribution. Web publishers still need writers, ad sales, designers, and pay rent and utilites.

I am certain that what Google pays is far less than what a publisher needs to survive because without other revenue streams, I would have been forced to have given web publishing up a long time ago.

The solution is simple and evident for me. Sell my own space at prices with which I can live, and I've already begun doing that.

The days of questioning the value of internet advertising vis-a-vis that of print, or any other media are OVER. The internet is a proven media and content publishers are being shortchanged badly because Google, the very same company that organizes and uses OUR sites for their benefit on their search pages for massive profits, has devalued those very same sites when the money comes directly to us.

I would love to know what the difference is between an ad campaign on the search network as opposed to the content network. I'd wager that the search network costs 3X more or better.

martinibuster




msg:3271895
 9:15 pm on Mar 5, 2007 (gmt 0)

>>>I'm confused MB, what would the SERPS have to do with an AdSense adjustment?

It's an example of a possible automated review that was triggered, in this case by certain conditions the SERPs. Several years ago a site of mine raged into the SERPs riding on scores of great keywords, then ten days later went Poof. I thought it was a manual review but from conversations I had with certain people it sounded like it was an automated process that over-zealously took it out.

The example is to illustrate that your notion of a manual review (albeit automated) is quite possible.

Content_ed




msg:3271906
 9:27 pm on Mar 5, 2007 (gmt 0)

I think panic at a temporary earnings (or traffic) collapse often induces webmasters into frantic activity trying to reverse the trend. In cases where this activity includes any changes in site design, all bets are off, because the reason we were doing "well" to start with isn't necessarily the same reason Adsense sees.

Pretty much every time we've attempted any sort of redesign in the wake of an Adsense crash it's jut confused the issue, and I'm sure we've abandoned (for advertising purposes) sections of our website that were just seeing temporary inventory challenges or seasonal blips.

Unless you have a great configuration change control system in place (and we don't) you can end up without a path backwards, beyond old backups and prayer. After a software upgrade incidental redesign a few weeks ago tanked the earnings on a major portion of our site, I had to go all the way back to a back-up from last summer to find one that returned things to normal. Part of the problem was that the other backups were done RIGHT AFTER other changes that didn't work out.

Olney




msg:3272029
 11:40 pm on Mar 5, 2007 (gmt 0)

Wouldn't this thread just show there is a problem?
I thought it was just me but ads on my Japanese site are usually seem to be the lowest of the barrel. For themes & markets I know that have plenty of advertisers.
I'm reading more & more people say that the ads are untargeted.

ronburk




msg:3272164
 3:18 am on Mar 6, 2007 (gmt 0)

a list of what causes earnings to dive.

End of fiscal year. If you're attracting ads from companies that do serious advertising, you may find a highly predictable once/year drop in sales. A couple of big advertisers get within a couple of months of the end, see that sales predictions have not measured up, and cut the advertising budget to help soften the blow for the final report.

Industry-wide belt tightening. Most of your revenue comes from ads for widgets. Newspaper headline blares "EU puts a tariff on widgets.". Advertisers pull back on their ads.

An overnight drop in earnings cannot be explained by any logical hypothesis.

Heh, that's what more than one person in this forum said before discovering that they had made a site-wide change that screwed up their display of ads.

According to THE NEW YORK TIMES and Google, advertisers will be able to "site-target" contextual ads, not just sitewide CPM ads, within a few months.

If that really happens, watch out -- WebmasterWorld will need an additional forum just to contain all the new weeping and moaning threads! Alas, I can't find any announcement of this online.

MsHuggys




msg:3272171
 3:28 am on Mar 6, 2007 (gmt 0)

I've been hearing alot of chatter about earnings decline, not related to traffic decline. I'm hearing it from clients, and I am reading as I lurk about the web forums.

I don't see this, and I have no idea why. My earnings both elevated and declined, since the beginning, have all been related to my decline and increases in organic traffic.

In fact, my traffic is way up as I got credit for my major websites' longevity in the last Google update. My traffic almost doubled on several websites. I believe it was not just the longevity credits, but the back links credit as well.

Anyway, what I saw was an increase in earnings. I hear you say, "so what, that is a no-brainer.". Not so. What I saw was an increase in my cpc.

I have a very wide nitch, and rank for thousands of keywords as a result. Site wide, on my major site, I average .22cpc over a long period of time...years. Suddenly, after the increase in traffic, I saw dozens of more refined nitches with cpc between $1-3. Many of these, had never gone over $0.50 in the past, on the best day. Some never over $0.10.

The first few times it happened, I thought newb advertiser. Now, it has happened hundreds of times and I am more puzzled. How does a low budget nitch(s), suddenly become a high budget nitch? I have no idea.

If I had to guess, I would say the increase in relevancy in Google and the increase in traffic, brought many more unique visitors to my site. Maybe, when you have stagnant traffic, from the same visitors over and over, they are not unique, have clicked before, perhaps on the same ads, they are heavily discounted or even ignored by Adsense. With a wide range of increased uniques, they are considered better converters, more highly coveted by Adsense and advertisers.

In some of those sudden performers, my CTR is 20% or more in recent months.

Not only that, I have been heavily updating the sites since the increase in traffic. Maybe there is a new content, recently updated content bonus in percentage of share. This might explain why a low budget nitch, of low budget products suddenly pays more. It is not higher bidding but higher share of revenue?

Speculation, but I would say decline in earnings, even cpc, may be the result of stagnant traffic and/or stagnant content, if it is a long term problem.

However, around the same time .... I did something else with my Adsense ads... it could account for some higher cpc bids. That... I am not sharing, now that I have identified several head to head competitors on these Adsense threads.

europeforvisitors




msg:3272178
 3:52 am on Mar 6, 2007 (gmt 0)

According to THE NEW YORK TIMES and Google, advertisers will be able to "site-target" contextual ads, not just sitewide CPM ads, within a few months.

If that really happens, watch out -- WebmasterWorld will need an additional forum just to contain all the new weeping and moaning threads! Alas, I can't find any announcement of this online.

It's on page two of a February 26th article [nytimes.com] at NYTimes.com (free registration required). The article quotes a Google spokeswoman as saying that, in In the next few months, Google’s advertiser reports will begin listing the sites where each ad runs. Advertisers will "soon be able to bid on contextual ads on particular Web sites" instead of taking potluck as they do now.

ronburk




msg:3272186
 4:08 am on Mar 6, 2007 (gmt 0)

thank you for the link, efv! That's a really interesting article. Must've slid by the forums here when I wasn't looking. I've got at least one affiliate that I would like to talk into switching to AdSense, and this new mechanism may give me the tools I need to do it.

Goes to show how sensitive Google is to competition for their bread and butter, when tiny Quigo can make them change policy after all these years.

The fact that Forbes can make more money by switching away from AdSense is really interesting.

Adding to the list:

New features that let advertisers bid on just the websites they want. Watch your EPC collapse when a couple of big, savvy advertisers figure out they can get more bang for the buck by paying higher EPC to just a handful of websites, and dropping out of the rest of the content network that happens to include your website.

Green_Grass




msg:3272200
 4:27 am on Mar 6, 2007 (gmt 0)

Thank you ronburk . It is heartening to note that adWords advertisers will have a choice in picking and choosing the publishers they want to go with. This will obviously lead to higher CPC for selected adSense publishers while others will lose out.

However I would like to point out, that it is not necessary that only very hight quality authoritative sites will gain. It is quite possible that smaller sites that presell better than authoritative sites may actually gain, at their expense.

Things are going to change. I fell quite a few so called high quality sites are going to be hit hard as ROI will take precedence over SERP positioning and quality of traffic will be more important than quantity.


MsHuggys




msg:3272202
 4:32 am on Mar 6, 2007 (gmt 0)

I speculate channel targeting has already been going on...in a beta stage. I'm seeing signs of it. That could be because my channels were already set up for this, long before I envisioned they would be used for this purpose.

Content_ed




msg:3272560
 1:50 pm on Mar 6, 2007 (gmt 0)

What else, apart from Smart Pricing, do you think will cause EPC to collapse?

And getting back to the question, just this morning, I noticed another factor that can contribute to panic changes in site design that really mess things up. I was looking at our earnings for the first five days in March thinking, "This is awful, I wonder what's going wrong." Then I pulled up the same five days from last March, and the results were within a couple percent. Sure, a lot of work has gone into the site in the past year that isn't generating additional Adsense revenue, but we're more of a running in place crowd than exploring brave new worlds.

So, I'm proposing a bad memory or wishful thinking as another reason publishers may rush into trying to tweak their sites and ending up shooting themsleves in the foot.

Process




msg:3272635
 3:07 pm on Mar 6, 2007 (gmt 0)

Tend to agree with EFV:

1. Dec and Jan started seeing the deep decline in our EPC and the only logical conclusion on our part, seems to be the ability of advertisers to bid separately on the content network.

2. Hoping that the "site targeting contextual ads" feature will favor us when it does get released.

Sally Stitts




msg:3273207
 10:29 pm on Mar 6, 2007 (gmt 0)

Today, my EPC is at an all-time 3-year low for the month.
Started in January - down significantly
February - further drop
March - drop accelerating
I find this is concerning, and unexplained.

My impressions, clicks, and CTR continue to rise, over pages on vastly differing subjects.
Yet my earnings are headed south. This causes me great concern, such as, maybe I should be doing something else. Or will have to be, very shortly.

Yahoo is even doing worse, with dropping CTR, due to hopelessly untargeted ads. The publisher game is a rocky road.

[edited by: Sally_Stitts at 10:34 pm (utc) on Mar. 6, 2007]

This 131 message thread spans 5 pages: 131 ( [1] 2 3 4 5 > >
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