Forum Moderators: martinibuster
I own a site that I recently implemented adsense on. Previously it was more of a hobby site, but I am finding Adsense to work quite well.
I have implemented in-content ads, ad units and the skyscraper in various situtations.
The EPC of one of the ad units is considerably less than the others. Is this because I see it is getting less click throughs AND roughly the same amount of impressions?
If so, would removing this ad help or hurt overall revenue? Will the EPC of other ads on a given page go up if one ad that does not perfom well is removed.
Thanks,
Jaid
This is a big question and one that many wish they knew how to answer (and others who think they know, but don't).
EPC can vary for all sorts of reasons, over time, by page, by ad position, by different keyword, etc., and each site is different. Two of the main factors that affect EPC - bidding rates and SmartPricing - are to differing degrees outside your control or based on information you don't have. Although some advocate that low CTR has an impact, I and others disagree - your lower CTR ads may also be the highest converting ones for the advertiser. Low EPC might indicate poorly converting ads, but it may be due to other things - and a poorly converting ad block this week may convert well next week.
Because each site is different, the main thing to do is experiment on your site, but in a scientific way. Eg:
Using this continuous testing process you'll gradually learn what works well for your site. We did this for about 4 months before finally settling on a format which has been the basis of our site design ever since. We occasionally tinker with it, but overall we think we've maximised the potential of our existing pages, and the main way of increasing future revenue is by adding new content.
Good luck.
21_blue, I agree with your testing model
After going through this process through several iterations, what kind of improvement did you see in site performance?
I should have noted that I didn't intially realise the importance of waiting a couple of weeks (eg: because SmartPricing updates weekly) for a proper analysis. But there were some changes that had an immediate impact, and others that needed time to feed through into the results.
Nevertheless, true to pareto, most of the growth came in the first month, when we roughly doubled earnings. Over the next couple of months we increased it by a further third, hitting a peak of around $6,500 before the summer decline in business started to peg it back. Our highest month in the Autumn period was around $8,500, and December 2005 was around three times the earnings of December 2004.
Not all of this growth is due just to changes in site design or ad positioning; we have added a lot of content as well.
21_Blue...that sound like a fantastic return on your efforts. Thanks for the info.
It really is important to let the changes simmer for a couple of weeks to see what happens.
One thing to remember is that while we're all probably looking for growth, that growth can come in smaller increments that still add up to a pretty big number in the end. It's nice when you do something that gets a 10 or 20% increase, but that doesn't happen that often.
It's a lot easier to get the 1 - 5% increments.
Jaid
