Welcome to WebmasterWorld Guest from 18.207.136.184

Forum Moderators: martinibuster

Message Too Old, No Replies

I find it amazing that earnings are so stable

No, I have not gone insane, read inside:

     
11:47 pm on Sep 12, 2003 (gmt 0)

Senior Member

WebmasterWorld Senior Member 10+ Year Member

joined:Sept 16, 2002
posts:1404
votes: 0


While I am in the same boat as everyone else when it comes to wild variations from day to day. I just found something that was really shocking:

My daily averages for July and August are exactly the same, while the daily average earnings for September started way below the previous two months, but now it is catching up to the same level.

Guys, I am not lying the daily average is to the penny... And I got less visitors to the site on August when compared to July. However, this was compensated by a higher CTR. It couldn't have worked better if I had planned it.... Hmmm now that I think about it, it would have worked better if traffic had doubled LOL.

I guess is time to write more content and actually promote the site. I might do it for a couple of days while I am on vacation next week. Althoug, I always say that and I always slack. Making money might be just the right incentive to do it for real this time.

1:51 am on Sept 13, 2003 (gmt 0)

Senior Member from GB 

WebmasterWorld Senior Member 10+ Year Member Top Contributors Of The Month

joined:Mar 9, 2003
posts:2099
votes: 10


It's not anywhere near so statistically relevant but I noticed something similar in that my impressions are way up this week compared to last week, my CTR is (bizarrely) way down and my CPC is much higher, giving... approximately the same earnings (not to the penny, mind).

My initial reaction was: hang on a sec, does this mean as I get more visitors to the site, AdSense starts displaying more generic adverts from higher paying companies so that the CTR drops but when people do click the CPC is higher?

'Cuz if that's the case, I've got a lot of work on my hands to get past the inertia of the relative lack of interest in generic ads compared to specific ads.

2:51 am on Sept 16, 2003 (gmt 0)

Senior Member

WebmasterWorld Senior Member 10+ Year Member

joined:Sept 16, 2002
posts:1404
votes: 0


The earnings were improving to historical levels and then the weekend happened. :(

I hate the weekends. Give me my office workers surfing from work anyday.

Still down for September when it comes to daily earnings average, not by much, but still down.

4:26 am on Sept 16, 2003 (gmt 0)

Full Member

10+ Year Member

joined:May 8, 2003
posts:292
votes: 0


wat is it with weekends any way? i have the misfortune of not only loosing out on traffic but also CTR!
4:47 am on Sept 16, 2003 (gmt 0)

New User

10+ Year Member

joined:Aug 24, 2003
posts:34
votes: 0


wat is it with weekends any way? i have the misfortune of not only loosing out on traffic but also CTR!

I am only theorising but I would imagine people who work with computers all day and are not computer nerds like those who frequent WW (myself included) choose not to go near a PC at the weekend :)

Also, weekend (aka home) surfers will be more 56k based compared to weekday (aka office) surfers and are probably a little more choosey about their clicks. Especially if they're paying for internet call charges still.

2:11 pm on Oct 1, 2003 (gmt 0)

Senior Member

WebmasterWorld Senior Member 10+ Year Member

joined:Sept 16, 2002
posts:1404
votes: 0


Just a little update, average eranings per day for September came in within 1% of those exhibited in July and August. Therefore, earnings stability continues.

I have to wonder if this is because of the volume that my site has. One would think there would be even more stability on a larger site where spikes in ctr and traffic would be smoothed out by the sheer amount of data. But my little site continues to show consistency.

Who knows?

4:26 pm on Oct 1, 2003 (gmt 0)

Senior Member

WebmasterWorld Senior Member 10+ Year Member

joined:Nov 27, 2002
posts:927
votes: 0


Following a similar discussion offline with a major Adsense publisher (well into three figures per day) about consecutive days showing figures within a few cents of one another, I did some analysis of the figures. Calculating the differences between consecutive days over a six week period and ranking in size order from most negative to most positive, the graph produced was an almost exact straight line. In this case, though the perception was of some kind of figure manipulation given occasional 'surprising' results, the analysis turned out to prove that day to day differences actually varied normally and the number of figures between -5 and +5 was the same as the number of figures in any other 10 unit range.

It might be interesting to do a similar analysis on your data - is this a one-off coincidence or when you look at a large quantity of data, is there a bias towards numbers that are close together?

4:46 pm on Oct 1, 2003 (gmt 0)

Senior Member

WebmasterWorld Senior Member 10+ Year Member

joined:Feb 25, 2003
posts:972
votes: 0


It might be interesting to do a similar analysis on your data - is this a one-off coincidence or when you look at a large quantity of data, is there a bias towards numbers that are close together?

Dude, go outside and get some fresh air. Trust me on this one, I've been right where you are in the past.

5:08 pm on Oct 1, 2003 (gmt 0)

Senior Member

WebmasterWorld Senior Member 10+ Year Member

joined:Sept 16, 2002
posts:1404
votes: 0


Dude, go outside and get some fresh air. Trust me on this one, I've been right where you are in the past.

Give him a break, when you are talking well into three figures per day, then it is worth doing what he is doing. My analysis is limited to looking at the numbers, saying how quaint! and posting the observation here.
7:03 pm on Oct 1, 2003 (gmt 0)

Junior Member

10+ Year Member

joined:Aug 31, 2003
posts:101
votes: 0


I'm finding an increase in earnings of 10-15% per month since I joined in June.
7:33 pm on Oct 1, 2003 (gmt 0)

Junior Member

10+ Year Member

joined:Jan 27, 2003
posts:87
votes: 0


Dude, go outside and get some fresh air. Trust me on this one, I've been right where you are in the past.

Trust you on what? What on earth is wrong with analyzing revenue data? Or any data at all? Bring it on, I say! The alternative is ignorance.

(And don't believe the hype -- it's not so blissful)

7:58 pm on Oct 1, 2003 (gmt 0)

Senior Member

WebmasterWorld Senior Member 10+ Year Member

joined:Feb 25, 2003
posts:972
votes: 0


I know, I'm obsessed with data too to the point that I know the most useless stats about my revenue possible. Excel and data to me is like pot and a bong to some people.
8:22 pm on Oct 1, 2003 (gmt 0)

Preferred Member

10+ Year Member

joined:Aug 1, 2003
posts:392
votes: 0


Therefore, earnings stability continues.

Isn't that of concern to you? I would have thought the aim would be to increase daily earnings each month, where possible.

By tweaking the ad size, position, and colour, plus some increase in traffic, plus presumably better ad targeting, my daily average revenue in Sept was slightly more than 3 times that of August (I started Adsense Aug 1st).

Looker closer, most of this was due to doubling the average CTR Aug to Sep plus a 10% gain in EPC. The rest was due to increased impressions.

So, to get back to some comments earlier in the thread, increasing impressions (Sep was 54% above Aug) doesn't seem to have hurt the CTR or even the EPC at all.

10:23 pm on Oct 1, 2003 (gmt 0)

Preferred Member

10+ Year Member

joined:Mar 23, 2002
posts:602
votes: 0


Have re-read the above posts with interest. Let me see...higher page views same revenue....lower clicks, higher CPC....flat revenue month to month.

So is there a cap? or have I extrapolated this data too far? <really hate myself for jumping to this conclusion> but is that what is being implied here?

10:55 pm on Oct 1, 2003 (gmt 0)

Senior Member

WebmasterWorld Senior Member 10+ Year Member

joined:Sept 16, 2002
posts:1404
votes: 0


Isn't that of concern to you? I would have thought the aim would be to increase daily earnings each month, where possible.

Not really July got me X impressions, August got me about 80% of visitors from July, and September got me about 110% visitors from August. Overall, my pageviews are down, and the effective CPM is still better than the first full month, but not as good as the second month. In addition, I would have never thought I would make any more money than to break even in hosting costs running ads so is no biggie.

The funny thing about the observation is how predictable return can be. I am not going to talk about specific CTR, but that also bounces around in a very, very narrow range. This tells me that fraud detection can not really be that hard to work out.

11:40 pm on Oct 1, 2003 (gmt 0)

Preferred Member

10+ Year Member

joined:Mar 23, 2002
posts:602
votes: 0


Have always thought that myself loan, that if the numbers are as stable as we are seeing, on a monthly average then fairly easy to spot the spikes over time. (previous posts regarding psa serving days)

Interesting that your basic data and results tracks so close to others that have posted, including the ups and downs in impressions, effective CPC and relativly flat earnings. That is why we asked the question as to is there another calculation going on in the algo, to generate this?

Would expect that as page views go up that so would revenue (based on <somewhat> predictable click through rates). The balancing off of CPC seems to offset this effect, is what you have seen.

This could be due to ad blindness (same ads on multiple pages), but in looking at earnings/visitor ratios it is also following the same trends, so don't account it to that.

Still think the % of psa's served over a period of time comes into all of the above.

This is way more fun trying to figure out this than the search engine algo....lol.

 

Join The Conversation

Moderators and Top Contributors

Hot Threads This Week

Featured Threads

Free SEO Tools

Hire Expert Members