Low, low impressions today

anyone else

[Powdork]

So far today i am noticing ridiculously low impressions, well under a quarter of the typical amount for this time of tuesday. I checked my highest traffic keywords and the rankings haven't changed. Income seems normal. I loaded up a lot of impressions on a rarely visited channel to see if the impressions would show up and so far they haven't.
It's probably the standard delay of stats we see during periods of change but I'm just curious to see if anyone else is experiencing this before I investigate the traffic (or lack of it) further.

[winglian]

been looking at my earnings... it seemed to have picked back up over the afternoon, but there has been no impressions/earnings updates over the last 3 hours (exactly the same as 3 hours ago).... I hope this isn't a sign of anything bad. My logs show normal website activity, and the pages are showing ads... anyone notice anything similar?

[mivo]

Mine had no updates in the last 6 hours or so.

[hfwd]

Normal today, but we had a large drop on Saturday - then a good Sunday / Monday.

Can't tell w/ Adsense - it's here today, gone tomorrow, back the following day.

[incrediBILL]

UT OH.....

I think I know what's happening to my site today.

Remember what I've been saying said about SUPPLY AND DEMAND?

There are many pages that don't seem to be able to fill both a skyscraper and a leaderboard today.

So either google has too many advertisers in my topic, or the advertisers have pulled back, or both.

The site did rebound a little at the end of the day, but it's definitely about 20% under par for a weekday.

[BeeDeeDubbleU]

It's now 08:35am here and I finished yesterday with earnings of about a third of what I had the day before. This is not a concern because my earnings do tend to fluctuate wildly. I don't know why but that's the way it has been since I started last August.

My site tends to serve an industrial audience so it bombs at the weekend but yesterday was a lot worse than normal for a while. The fact that I had a record day the day before made it look even more serious. I think this may have been down to delays in reporting of the stats?

[fearlessrick]

All metrics still down for me, about 60%-80% lower than high day for the month. Earnings: the emerging pattern is the lowest 11 days of the month are the 1st - 5th, 21st-26th. That's fairly obvious.

Could be spending patterns by advertisers, seasonal, google or other, or a combination of all factors. I'm going with the combo theory since I'm just guessing.

Got the typical canned response from G yesterday evening. As usual, no specifics, no help.

Only solution seems to be to get back on the treadmill and stay there until the lowest earnings day doesn't hurt. Continue to solicit advertisers directly.

Also, stay away from that competitive ad filter. Learned it the hard way after seeing 5 clicks earn 4 cents yesterday. Boy, was I wrong on that one!

Also, I should mention that stats seemed to be updating slowly for me as well, but nothing worse than an hour.

My apologies to anyone I offended recently. No excuses. I am just a nutcase. mea culpa...

[BeeDeeDubbleU]

I see my lowest returns on 1st to 3rd but no problems elsewhere apart from yesterday, which is now beginning to appear like a glitch.

[incrediBILL]

Whatever went bonkers yesterday seems to have corrected itself today and income is back up and slightly ahead of the norm.

Off to the pub for lunch to celebrate :)

[incrediBILL]

Learned it the hard way after seeing 5 clicks earn 4 cents yesterday.

It occurs to me as we speculated that EU 0.05 bids would make a bit more than US $0.05 that our friends in Canada and Australia would generate much less based on the conversation rate.

Maybe that's what you're seeing for $0.01?

[hunderdown]

fearlessrick, FWIW, one of the effects of the work I've done to optimize has been to damp down the wild swings. Not the intention, but it happened. Last year, my best days could earn more than ten times my worse days. This year, the difference is more like three times, and that's due to the worst days improving, not the best days declining... Now that the floor's higher I want to start pushing the ceiling up!

[Tropical Island]

Whatever went bonkers yesterday seems to have corrected itself today and income is back up and slightly ahead of the norm.

Yesterday was bad - today is better than ever - headed for a record if the trend continues. Yesterday was 30% lower than Monday.

I think the ups & downs are meant to give us something to worry / cheer about every day. Otherwise we would become complacent and have nothing to complain / rave about :-)

[incrediBILL]

Maybe Google's AdSense team monitors this board and plays AdSense Fantasy Whiner where they pick the 3 people who will post first and in what order about earnings increase/decrease and then tamper with our eCPM. Whoever gets the trifecta wins the pool for the day.

They probably have a side pool for the largest reported change in earnings too.

How's that for entertaining paranoia?

[cyanweb]

LOL - yes - it seems the more i type on these forums the lower my eCPM goes... think i'll shut n lay low..

[fearlessrick]

Bill, your last post hit my funny bone. I have to watch myself and the paranoia inherent in deling with things I know little to nothing about. (you wouldn't want to know what I wwas like after the 2004 election... it got a little crazy!).

Anyhow, nothing will convince me that they don't play games with numbers, whoever "they" are.

And to clarify, those 5 clicks went for 4 cents TOTAL, less than .01 per click. And I'm not joking about that. I saw that with my own two bloodshot (pollen allergies) eyes.

[fearlessrick]

Hunder, per your last post about establishing a floor: my sentiments exactly. I'm actually close to doubling last month's earnings, so I suppose I shouldn't complain. My floor is now three times what it was a month ago, so it's basically a case of adding more pages, more code, doing more seo, marketing, etc.

Kinda like work ;-)

The beginning and ends of months are showing a distinct downtrend for me. On the other hand, from about the 6th through the 20th are pretty good.

[hunderdown]

Yeah, that's funny--I've seen that too. I can understand that the end of the month would be down, as advertisers exhaust their budgets, but why the beginning as well?

[oddsod]

There seems to be a lot of talk about advertiser budgets in WW.

I'd like to point out that in many cases, in many sectors, and for a large number of advertisers - budget isn't the problem at all. Spend some time in the Adwords forum; the complaint is that Google doesn't have enough clicks for them. Personally, and as an advertiser, I have no budget. Let's say I'm selling software via a download from my site (just as an example - I don't sell software). For every $1 I spend in Adwords my website makes a $4.00 profit. Why would I stop spending if I'm making $4 for every $1 I spend? If I run out of cash I'll borrow, or get a partner and offer him a 1000% annual return.

There is no limit to how much I'd be willing to spend with Google every month at that ROI, and no budget that runs out by the end of the month. Also, if you do set a low montly budget Google spreads that out over the course of the month.

There is, of course, a factor of shoppers having money at certain times and not at other times. But, that will be reflected in impressions as well as EPC. Shoppers are more likely to go researching a price when they have the money to make a purchase.

[Powdork]

There are other factors as well. I am finding in my niche that Monday and Tuesday have been performing better in terms of Adsense and affiliate sales. Some campaigns I only run on those days.

[hunderdown]

oddsod and Powdork, thanks for those insights. So if we are seeing ups and downs over the course of a month in earnings, you would attribute that to variations in traffic or smart pricing? (I'm not talking about purely day-to-day variations, by the way. I mean seeing jumps up or down from one week to another.)

[Powdork]

So if we are seeing ups and downs over the course of a month in earnings, you would attribute that to variations in traffic or smart pricing?

Its very difficult to tell unless you are looking at longer range trends. There are just too many variables especially when we have very little idea as to the variables used for smart pricing.
Another variable Google can alter is the length of delay among the updates for various stats. For instance, there is no doubt that impressions, clicks, and earnings are all updated independently of one another. Now suppose they do that, but have a policy of not letting one stat fall more than 6 hours behind the others. Now they change that to not letting any stats fall more than 24 hours behind the others. That change alone could explain alot of the posts in this thread.
And then you have to throw your traffic mix into the equation. Different topics, with different eCPM's, will perform better at different times.

[hunderdown]

Powdork, so you would say that there's no point in comparing data week to week? I know that day-to-day stuff is all over the map. Do you only look at trends from month to month?

[oddsod]

Rolling averages may be worth looking at. But with any stats it's an appreciation of its limitations that's the most valuable skill.

[Powdork]

Do you only look at trends from month to month?

I look at everything, act on almost nothing. For most of my sites the location and size of the ads is dictated by the page's layout, rather than the profitability of a certain location, size, or color. The only determination I need to make is whether to have adsense on the page at all. That part is pretty easy.
The daily and weekly fluctuations are simply part of the program I have come to accept.

Yesterday I went to my first Adsensaholics Anonymous Meeting....

Google grant me the Serenity to accept the things I Cannot change…
Courage to change the things I can
And Wisdom to know the difference…

[hunderdown]

For most of my sites the location and size of the ads is dictated by the page's layout, rather than the profitability of a certain location, size, or color. The only determination I need to make is whether to have adsense on the page at all.

Good point. I don't use banner ads because I don't like the way they look. When I started, I settled pretty quickly on a single skyscraper unit, and tested only to see if left or right worked better. Left won.

Others may take a completely different approach, of course, and jam as many units onto the page as possible. I don't like to do that because I want people to browse around my site.