Forum Moderators: martinibuster
low performance mean low ctr or low ecpm?
I thought low ctr means low ecpm... but it is not always like that isnt it?
Does search and adlinks include in this theory of pulling out ads from low performing pages?
newbie pls bear with me if this is not intelligent
It is not possible to work out the earnings per click (EPC) for a specific advertiser showing on your site. But is there are advertisers who you suspect of being low quality Made For Adsense (MFA) sites, then you may reasonablble expect that they pay very little for clicks - it may be worth blocking these sites using your competetive ad filter in AdSense.
If you have a page that achieves a low CTR then you may want to consider changing how your ads are positioned or presented to reduce ad blindness. Removing Ads on your page with a low CTR would not offer any benefit so far as I can see.
If your CTR is acceptable, but your eCPM is low, it may be a function or your niche, or possibly you need to improve your content so as to attract better paying ads.
Hope this helps and good luck.
P
its quick too...
So its "content is king" as always.
Mine is a new website with about 600 ad impressions daily but not with a great ecpm ($3.5). but it is a hot niche.
Does new websites always start on spricing?(I learnt this word here)
Does no. of years domain booked (renewed) have anything to do with spricing?
Removing Ads on your page with a low CTR would not offer any benefit so far as I can see.
Pengi, sorry, but I have to disagree with you. I have personally seen my total earnings go up signficantly on two occasions after removing ads from a page or pages that had a very low CTR (well under 1%).
Smart pricing is supposed to look at converions, but if the advertisers aren't tracking conversions for Google, then Google has to look at something else. There may be several factors, but I think it's arguable that a very low CTR might be one of them. It could be caused by ads that just don't interest visitors, and thus any clicks won't convert well for advertisers. Or it could be caused by a lot of repeat visitors.
Anyway, if you can identify pages with a good number of impressions and a very low CTR, it may be worth a try.
use channels to come up with the highest epc page design, and go from there... don't use optimized page design for the best ctr, use 'em to find the best consistent epc.
Welcome to Webmaster World - currently the best public forum of web publishers.
Currently
I agree with the advice above, but I would add that I don't think a low ctr page with low ecpm is going to make a difference unless it has a disproportionately high percentage of visitors.
IE, if you have page A that has a reasonable amount of visitors, reasonable ctr and ecpm, and page B with low metrics PLUS a low volume of impressions it's not going to impact a great deal on page A. However, if you reverse that and have a page that has a high number of impressions and poor metrics compared to other pages then it might, and you should consider removing the ads to see if it has an impact or not.
Lots of low paying pages can add up to a decent amount, so I'd be careful about what you remove.
As regards blocking ads, if it's a persistant MFA then yes, block it. Many people report good success with blocking them and it doesn't necessarily mean that it's replaced with an even lower paying ad. MFA's are very good at manipulating the system for optimum placement - over, and above proper ads that pay well in many cases.
MFA's are very good at manipulating the system for optimum placement - over, and above proper ads that pay well in many cases.
And the well paying Ads pay one cent above the MFAs ;-)
Google place ads based on assumptions of what they THINK might be the highest paying ads, and they base those assumptions on data from other parts of the Google network - not necessarily your site. For example, and ad that has a good ctr yet low epc in general is likely to be placed above an ad that has a high bid price, yet low ctr. The idea is that 10 clicks at 10c is a better bet than 1 click at 99c. Problem being that ads that are well targeted and historically pay well on your site, can and are replaced by ads that pay very, very low.
Google do not state that the ads they show have bid the highest - they state that they show what they think is likely to represent the best income. So the scummy MFA ads you see won't be bidding 1c higher than the ads from real advertisers. They are getting cheap placement because google "Thinks" they "Might be" the best bet. But in many cases we know that this is simply false. Blocking low paying MFA's is not new - it's been discussed many times here over the last couple of years.
So the scummy MFA ads you see won't be bidding 1c higher than the ads from real advertisers.
Ummm, that's not what I meant (if you are referring to my post above).
You may want to read my post here:
[webmasterworld.com...]
However, does the real ad pay just 1c more than the MFA? In my experience that's not the case. I'd say that the average epc I get is WAY higher than any MFA could possibly afford. So does keeping the arbitrage crowd off increase income? Oh yes :)
does the real ad pay just 1c more than the MFA?
That, I don't know, I was merely guessing...only G knows!
But is it not true that Advertisers pay 1c above the next highest bidder?
Which means that if it's next competitor is an MFA, the Ad pays very low!
This is the reason I'm looking for a tracking tool that gives me the destination URLs of clicked Ads, because the average epc I get is WAY LOWER that many MFAs could afford :)
My rules for blocking are to look at the landing page of ads I see. If they are clearly MFA's beyond doubt then I'll block them. However, as some of the ads only appear once or twice I usually see if they are still there 24 hours later and THEN block them. No sense in clogging the filter up with seen once ads. Genuine ads stay - even if they are badly targeted as again, often I see them just once or twice anyway.
Also, I won't use the preveiw tool to tell me what ads are being shown elsewhere on the grounds that it can't be trusted to show ads I can currently see on my pages. Just block ads you see as MFA's probably target with a splattergun approach. Also look at your search terms and what MFA's are shown on Google's search pages. I do occasionally block ads seen there.
It's fairly clear to see from the landing page what is and isn't an MFA. If there is any doubt I'll usually leave it in place. But do real ads pay only 1c more that MFA's? In my experience yes they do. The business model of the MFA dictates they can only spend the absolute minimum to get placement and rely on eye catching copy to get clicks. Genuine ads may well have copy as dull as ditchwater, but they are the ones with the money, and both us and the MFA's are looking for their clicks.
Don't forget the RandomPricing(tm) algo is supposed to discount the cost of clicks depending on how it feels your site converts. If your site shows mostly MFA's then the genuine ads may well be smartpriced to that level. One way to minimise RandomPricing(tm) is to keep off crappy low paying ads. There are several ways to do this - blocking is not the only way. That way over time the algo places a higher value on genuine clicks. A higher value on the site also means that in effect a lot of the low paying ads are often priced out of being shown in the first place.
afaik, the nature of the conversion is first defined by the advertiser, not google... in the case of mfa's, i would guess that a "conversion" is simply getting the click, no sale involved, so mfa'er traffic should be as good as it gets, lol? every click is a conversion? ack!
all joking aside, i do block all mfa sites, and it has made a big difference in my epc.
There is also another very good reason for advertisers not to use conversion tracking. As it stands, Google don't have a clue as to what I think a conversion is. We do know that Google takes into account a wide variety of factors in determining what is a conversion - hence what the RandomPricing(tm) value is. Unfortunately much of that data is trends and patterns seen elsewhere, the trends and patterns may well be based on other trends as opposed to hard data such as what advertisers state as conversions.
If I tick one of the boxes and tell them, then they will KNOW what I regards as a conversion and it's likely that they will charge more for those clicks. At the moment I can work out what converts for me without too much trouble, and thus amend my campaigns accordingly. By not telling them what a conversion is my costs as an advertiser are possibly lower. Telling Google what does, and doesn't convert doesn't help me - it's just likely to give them a reason to charge me more for the same clicks, so I won't do it. Nor am I alone.
