homepage Welcome to WebmasterWorld Guest from 54.226.235.222
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

    
Statistics on small sets of numbers
tony873004




msg:1332873
 5:08 am on Jun 2, 2006 (gmt 0)

I'm tired of trying to speculate how well my site is doing. I earn an average of ~$1 / day since I joined in Feb 06.

May started out as my best month ever, and ended as my worse. I had a 2 week drought of nothing but $0.00 or < $0.20 / day, following a week of > $2.00+ / day.

You can't do meaningful statistics on small samples of numbers and it's frusterating me.

 

martinibuster




msg:1332874
 5:34 am on Jun 2, 2006 (gmt 0)

You left out important information like how much traffic you are trying to monetize.

Then there is the issue of whether your site is appropriate for monetizing. Joke sites etc. reportedly don't generate much revenue. Similarly, many forums are difficult to monetize.

So which is it? Could the problem be your niche or the amount of traffic you are trying to monetize?

youfoundjake




msg:1332875
 5:38 am on Jun 2, 2006 (gmt 0)

I too get small numbers, been going for about 4 months and a 2.50 day is the lotto... but i got the 2.50 day by putting a channel to each page, and monitoring which page got the most clicks, and then i changed the color of the ad and made sure more traffic landed there and it went from .20 days to 2.50-3.00 days, now i just need more traffic. So if you haven't set up channels yet, you might want to check it out. on a side note, i get about 25 impressions a day with 3 or 4 clicks.

eeek




msg:1332876
 5:57 am on Jun 2, 2006 (gmt 0)

You can't do meaningful statistics on small samples of numbers and it's frusterating me.

So stop trying. Instead of worry about meaningless statistics work on increasing your traffic.

berto




msg:1332877
 12:47 pm on Jun 2, 2006 (gmt 0)

This is one reason, perhaps the biggest reason, why it is so hard (for some) to move from $1/day to $10/day.

At < $10/day, the sample sets are so small that statistical noise predominates. Meaningful testing (over a reasonable time period) is next to impossible.

It's often been said that it's easier to move from $10/day to $100/day than from $1/day to $10/day. I think it's because, at $10/day and beyond, the noise factor diminishes, and real cause and effect become clearer.

rden17




msg:1332878
 1:10 pm on Jun 2, 2006 (gmt 0)

Martinibuster is 100% right about joke sites. I started one of these in Feb 06 myself in a niche occupation for fun and added adsense. CTR is pathetic on this site. I'm finally starting to get better SERPs, and am #1 on MSN, and climbing on yahoo and google, but clicks are real hard to come by. If you are going to do a joke site, do it because you enjoy it, not in the hopes of making some money with it. What I have done is started creating a few other sites in my field with a more serious angle- quality content that people want to read and search for. This is getting better results with adsense. While these are new sites as well, it has taken several months to get in the SERPS, but now it is picking up. I thought it never would. Just be patient, and like you read all over this board, provide quality content and they will eventually come.

tony873004




msg:1332879
 9:34 pm on Jun 4, 2006 (gmt 0)

I think Berto said it best. Statistical noise is just too overwhelming.

It's a science-based website. It's been online for 5 years, but only with Google ads for 4 months. I average about 200 page impressions/day.

I get the impression that people who click often click multiple ads. Perhaps they know that it makes me money, or perhaps they're just click-happy people. So less important than page impressions is how many click-happy people visit my site. And that number is way below the statistical noise.

ronburk




msg:1332880
 5:03 pm on Jun 5, 2006 (gmt 0)

You can't do meaningful statistics on small samples of numbers

Even low-traffic sites have enough numbers if you wait long enough. Load all your past daily stats into a spreadsheet. Calculate the moving average. Start with a window of 1 week. Keep expanding the window until you see a relatively smooth curve. Then you have trends you can look at.

If you had ten times the traffic, you still wouldn't look at a 30-second decline in income and get upset. Since you have low traffic, you can't pay attention to numbers averaged over one hour or even one day. But you probably can pay attention to numbers averaged over some number of weeks. It's just a matter of scale relative to your traffic volume.

This is an example of people taking the numbers handed to them even when they're not appropriate for your use. You can make your own stats that are appropriate -- you don't just have to take the analysis that Google hands you.

berto




msg:1332881
 6:38 pm on Jun 5, 2006 (gmt 0)

Trouble is, the longer you wait, the more things change. Who has the patience and iron will to apply just one change to a set of test pages, then leave those pages unaltered for weeks and months as you gain new insights and good ideas in the meantime? Moreover, exogenous factors come into play, like monthly cycles, seasonal cycles, changes in the Adwords and Adsense programs, SERPs fluctuations (or having half your pages dropped from the index) that change the volume and nature of your visitor traffic, etc., etc. With so many changes going on during the testing period, with so many variables to consider, it becomes difficult to impossible to determine true cause and effect.

ronburk




msg:1332882
 12:12 am on Jun 6, 2006 (gmt 0)

Who has the patience and iron will to apply just one change

When you have that little traffic, you should be spending your time expanding your content, not twiddle-tweaking AdSense page placement and the like.

21_blue




msg:1332883
 12:16 am on Jun 6, 2006 (gmt 0)

For a description of the stats you can do, see my Golden Gate post [webmasterworld.com] (in the library).

berto




msg:1332884
 1:14 am on Jun 6, 2006 (gmt 0)

When you have that little traffic, you should be spending your time expanding your content, not twiddle-tweaking AdSense page placement and the like.

I tend to agree with you. You should emphasize content creation and traffic building, but don't neglect ad tweaking entirely.

When I first deployed Adsense, my "ad placements and the like" were so poor, my CTRs, EPCs, and eCPMs were so bad that I would have had to increase my then (not minuscule) traffic levels over a hundredfold to earn decent money.

I chanced upon one golden tweak that raised my CTRs and revenues by 5X. Great! Now I "only" had to increase my traffic levels by 20X in order to earn decent money--not nearly so impossible a task.

That golden tweak motivated me to continue. After that, content creation and traffic building no longer seemed so pointless.

Garry4Ads




msg:1332885
 2:15 am on Jun 6, 2006 (gmt 0)

Its interesting topic though. I am also newbie to AdSense. I am running ADSense for last 1 month. It started with .5$ day and it gone up to 14$ a day also. It now varies from 0.0$ to 3$ per day. I realized that it is just because I had a 'good content' which makes people to visit my site. Now, it purely depends on are you displaying 'interesting ads' which encourage visitors to click on them or/and the ad placement. I have another travel site running AdSense for last 3 months, I tried all my skills including making it 5th ranked in Google, but it fetches 1 $ a day because I know that the content is not good enough for the visitor to go throw various pages. Need to try out various things to make 1$ a day to 10$ a day.

Good Luck!

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