Forum Moderators: martinibuster
I suggest that ‘Smart Pricing related algorithms’ have contributed significantly to the major fluctuations observed by many publishers.
Depending on the number of click conversion sites and pages are grouped into categories. To make the things easy to understand I will give use some examples:
1) If a site xyz.com has 3% conversion rate---------------------------------------- is placed in a category ‘A’ and gets highest payouts (say $5) from advertisers (and so also from Google).
2) Site yyx.com has 1% conversion rate------------------------- is placed in category ‘B’ and gets less payout ( say $2) from the same advertiser, who paid $5 to the publisher of xyz.com
3) Site zzx.com has a conversion rate of 0.2%--------------------- is placed in category ‘C’ and gets the least payout (say $ 0.3) from the advertiser.
If now over a period of time (e.g. Few days or 1 week) zzx.com’s conversion rate improves from 0.2% to 1% then he moves to category ‘B’ from category ‘C’.
Of course there are other factors like 1) advertisers budget at the time of the click, 2) time of the day when the click is made, 3) geographical region from where visitor made the click and 4) some other unknown factors ....
The publishers, who are witnessing a sudden decrease in earnings despite increased traffic and increased CTR, can blame ‘SP’ algorithms. As and when your site (or web pages) moves from ‘C’ to ‘B’ category or vice versa you will see a change in earnings.
Certain kind of sites e.g. travel related, e-commerce, sites with product reviews have higher conversion rates. Some special “made for Adsense” site also have high conversion rates. While some other sites like ‘text heavy’ descriptive sites have low conversion rates (because visitors at your site come to read and are less likely to buy, unless you are promoting something through your writings).
P.S. These are all hypothetical categories, used to explain one type of algorithm that ‘Smart Pricing’ is perhaps using. I am sure there are more knowledgeable webmasters here who will contribute much more to my simple explanation.
A short while ago, I added extra banners to some pages to hopefully increase earnings. The extra banners didn't get clicked very often, so the site's overall CTR dropped by about 50%.
Then started a slide in cost per click. At the time there were loads of threads about drops in earning so I thought that was affecting me. I let things go for a couple of weeks, then decided to cull all the unproductive banners and hey presto - there was a dramatic rise in ctr to the previous steady level, and earnings shot back up again within 48 hours.
I'll never know for sure, but my suspicion is that SP looked at the ctr of the site, and the change in it and downgraded me. I probably still get $0.03 clicks on occasions, but the average click price over the month is high.
I now monitor banners very carefully, and if they don't produce, they don't stay. I removed some of the the adlinks banners I was testing after 24 hours as they clearly weren't working.
In the past I have tried experimenting by removing ads or channels when they don't seem to be performing well to increase the overall CTR and CPM and hopefully help with smart pricing. But because there are so many other factors involved I was unable to tell if it was beneficial.
I currently run two ad blocks per page, which of course has decreased the CTR and CPM from when I only ran one. Overall earnings have improved, but I don't know if this is because there are now more viewable ads, or because the additional ad unit is better placed, or just due to a seasonal demand.
Unfortunately unless you get a high volume of steady traffic per page it is very difficult to determine what is happening.
On my site, I've seen that the targeting tends to be worse on the second ad block. This may affect both clickthrough rate for impressions as well as overall conversion since the poor targeting results in ads that the users aren't really interested in. If so, the second adblocks may be equivalent to poison.
Affect CTR? Yes, no doubt, but where is the conclusive evidence that Google imposes a Smart Pricing-like penalty for low CTR?
Affect conversions? I don't see how this follows necessarily. Assuming you are not tricking visitors into clicking, why should a visitor be less likely to convert clicking on a lower-placed ad link? It could be that, after reading through the (motivational) page content, the reader is more likely to convert.
If we accept the argument that 2nd (and 3rd) ad units are poison, wouldn't this imply that 2nd, 3rd, and 4th advertisements within a single ad unit are also poison? Hence, that 2-ad banners are better than 4-ad leaderboards and skyscrapers. That 1-ad buttons are better are better than 2-ad and 4-ad blocks?
So, to optimize both CTR and conversion rates, we should have just a single 1-ad button on each page. But is that a strategy to optimize CPM and overall revenue?
What am I missing here?
[edited by: jimmyboy at 4:07 pm (utc) on Mar. 19, 2005]
Has the market gone soft? HAve the advertisers seen a reduced ROI and reconsidering other options or pricing models.....or is it simply the downturn from the Christmas season?
What can't be ignored is that the situation is always fluid and results differ by industry. As pointed out by respected members not all markets have been affected equally.
So what are the options? Ride it out until the pricing model drops below alternatives then change. Add more content pages in an attempt to generate more click throughs? Work on ad placement to attract click throughs....OR....spend your time typing in this forum that Google is trying to rule the world while punishing publishers.
As a sidenote the targetting on ads since January has been an issue. In almost two years in adsense the first time we actually had to use the filter was mid January and has been continuos since then. Have emailed support in regards to this but no improvement at this time. Some ads are totally off base and I have to think the effort was to reduce the number of PSA's delivered.
Smart pricing depends not just on CTR but on how those clicks are converting for the advertisers...
I know that Smart Pricing depends on conversion rates. But also on CTRs? Low conversions cost the advertisers. Low CTRs cost the advertisers nothing. (If users don't click, advertisers don't have to pay.) Low CTRs do cost Google, however, in terms of increased ad server load. But where is the authoritative evidence that Google penalizes us for low CTR?
Regarding having multiple blocks on one page - The second and third block usually have lower paying ads than in the first block. If multiple blocks are placed in such a manner that people are more likely to click the second or third third block instead of first block then you might be losing revenue.
The first ad units are probably above the fold. The 2nd and 3rd ad units may be below the fold, but after (if you've done it right) motivational editorial content. Whether the user is more or less likely to click on higher or lower ads depends on the situation.
Why is ad clicking an either/or proposition? Do we suppose that visitors have an implicit "click quota," as it were? That they are good for just one click per page, so we'd better make it count. What about the (no doubt) smaller pool of visitors who are good for two or more clicks?
Consider this:
--Gaming magazine A runs ads for the top three best-selling games at the beginning of the magazine, no more after that. Magazine A has many readers.
--Gaming magazine B runs ads throughout the magazine, with a reasonable mix of ad pages to editorial pages. Magazine B has many readers, maybe not quite so many as magazine A (or maybe as many, because ads can supplement editorial content, not just be nuisances).
--Gaming "magazine" C is 99% ads, 1% editorial content. Magazine C (more like "advertising flyer"--or scraper site?) has few readers.
Which magazine is likely to have the greatest ad revenue (a product of readership and numbers of ads)? Between the two extremes of A and C, my guess is that magazine B maximizes its ad revenues. Which may be one reason why most gaming (and other) magazines are like B.
Yes, different strokes for different sites.
Just don't assume that maximizing CTR and conversion rates will maximize your revenue, or is even best for advertisers. If in an attempt to maximize CTR and conversion rates, I make a single high-bidding advertiser supremely happy, how have I benefitted the lower-tier advertisers and advertisers in general?
On my site, I've seen that the targeting tends to be worse on the second ad block.
Sorry, but I have to make this follow-up observation:
On *my* site, I've seen that targeting is often worse in the first ad block. The more on-topic ads appear in the second and third ad blocks. (Or would, if I could just get second and third ad blocks to appear, but for some reason, I can't. But that's a different story for a different message thread.)
That is to say, it's often the crappers and the spammers who outbid the on-topic, "legitimate" advertisers on my pages. I can see from the Adsense Preview Tool that quality on-topic advertisements are available, but for whatever reason the advertisers usually seem to low-bid.
So, another reason why 2nd and 3rd ad blocks don't necessarily imply lower visitor interest, lower CTR, and lower conversion rates. Lower EPC? Yes, but not always lower ad relevancy.
Your Mileage May Vary.
This is one of the reasons why I am so excited about AdLinks. On my site, for the first time the low-bid but "quality" advertisers get their chances.
Most of the threads where people have been complaining about 'drop in income' there are always several others who report an increase or stable income. If the low advertisement price was the only issue then there would have been a decrease in income for all publishers.
I think three factors may come into play (for some publishers, at least):
1) Seasonal variation. My CPM and earnings are up, but that stands to reason, since my topic involves leisure travel and this is the time of year when many readers are planning their vacations. If I were writing about retail products that get bought during the Christmas season, or computer-related items that SOHO businesses might buy in December to maximize their tax deductions, I'd expect my CPM and earnings to be down at this time of year.
2) Supply and demand for ads on popular keywords and keyphrases. (There are a lot of sites competing for home-mortgage ads, but not so many competing for ads about expedition cruises in Alaska.)
3) Changes to the smart-pricing and compensation algorithms.
Smart pricing:
We've already been told by Google that smart pricing looks at the type of content that a click is coming from (e.g., a camera review vs. a page of photo tips). Over time, the sheer quantity of conversion-tracking data has obviously increased, and Google may have determined (for example) that forum clicks convert even worse than was originally thought, that clicks from special-interest sites with focused themes convert better than was thought when smart pricing was introduced, or that certain other charactistics go hand in hand with a lower or higher conversion rate.
Compensation:
Some members have reported that their EPC has slipped as their clicks have increased. This could be pure coincidence, but then again, who knows? Adsense was launched as a "people's ad network" for publishers of all types and traffic levels. Maybe Google feels that, instead of using an upward-sliding scale of payments, it should be using a downward-sliding scale that ensures a higher minimum for small publishers whose loyalty and positive word of mouth will help grow the network. If Yahoo or MSN comes along and starts cherry-picking, then Google can always tweak the compensation algorithm to make the rich even richer again. (Disclaimer: This isn't a theory or even a hypothesis; it's just "what if?" or "try this on for size" speculation.)
"Google can always tweak the compensation algorithm to make the rich even richer again". The term ‘Matthew effect’ is used to describe a phenomenon in which ‘the rich get richer and the poor get poorer’. The ‘Matthew effect’ derives its name from a verse in the Gospels of Saint Matthew from The Bible “For unto everyone that hath shall be given, and he shall have abundance; but from him that hath not shall be taken away even that which he hath”(Matthew 13:12 and 25:29).
Click to view more ads, eh? Maybe they're ok for some sites, but on a lot of content sites, these things just don't pass the smell test.
Click to view more ads, eh? Maybe they're ok for some sites, but on a lot of content sites, these things just don't pass the smell test.
One of my sites is an info portal, where much of the editorial content is, in effect, Google-like ads but without the monetization. The actual Google ads, far from being a nuisance or necessary (for me, the hard-working publisher) evil, instead have the potential of being an additional valuable info resource. (I say "potential" because, apart from theory, in actual practice it is sooooooo hard to tweak things to get Gooooooogle to serve the "goooooood" ads.)
For another of my sites, people come to the site with questions that need answers, with problems to solve. Google ads are *potentially* another route to answering their questions, solving their problems.
So, yes, I can see where--in an ideal world with an ideally run contextual ad program--users would gladly click to view more of my sites' ads.
Seems logical.
Regarding the feel of adlinks, maybe they'll improve this over time. I wonder if it would have been better to use a phrase like "To view ads for products or services, enter a search phrase". They could have accompanied that with an "ads by google" in the box (unless that's already there and I missed it). Or who knows, maybe this approach would suck more.
I think it's supply and demand, plus some google problems , plus advertiser concern about fraud keeping campaigns in close check, etc.. etc...
I do adwords too (very cheaply I might ad) and when content network kicks in, google overdelivers every time - incredibly cheap clicks coming in on some very good keywords.
