I don't look at daily numbers. I don't understand why any veteran would bother, since we know that daily fluctuations mean nothing, that one cannot draw any conclusions from daily numbers, etc.

I know you often post when your numbers are bad for a day, and good for a day, but what's the point?

Is there one? Or is it just for entertainment.

I agree with coachm that daily figures per se are not relevant.

The variance within a set of daily data points (e.g. looking at 30 consecutive daily data points) may be interesting and quite telling; rolling averages (e.g. 7 days, 30 days, 50 days, 90 days, 200 days) are much better for identifying inherent problems and trends than looking at daily values, or (heaven help!) at intraday values.

Our long-term data set tells me that the variance within the daily data is so big that it can be called "random". It does not make sense trying to analyze a single throw of a dice IMO.

I don't understand why any veteran would bother

Addiction ;)

What is wrong with some of you people constantly criticising what others post? No wonder so many have left this forum board.

but what's the point?

It's merely an observation note:

Unusual Thursday 8th Jan 2009 Metrics

Do I have to have a special permission now to post:

Result? Best day for 2 months and actually my average daily earnings from January to October 2008.

I never wrote that they were or were not relevant, just unusual considering the new lows of the past couple of month...yes, you've seriously pissed me off now!

I asked the question in case I was missing some purpose that might be useful or valuable. You didn't really answer my question as to the point, other to say it's an observation, which isn't what I asked.

...which is fine.

It's like this. If someone posts messages every few days saying something like: "Today the sun rose at 6:43. How unusual, that's three minutes earlier than last week. wouldn't you be curious as to the point?

Would you ask the person why they felt it was important to post such an observation, or would you silently draw some conclusion.

You've probably got my share HuskyPup. Definitely been unusual for the last few days, but when some go up others go down by 50%. But I'm going to wait to properly analyse figures and will post here in June to ask if anyone saw a fluctuation in January. ;-)

Still, there's often an increase in traffic in January traffic for me that holds into the year, and there are good signs of that. But less money to pay for more server. I'm diversifying this year.

My Thursday statistics were similar to the numbers for Monday, Tuesday, and Wednesday of this week (and to the daily average for January).

My stats (and I definitely look at them daily as any sane person would) are slightly on the up compared to last year's.

I am concerned on a daily basis in comparing stats to last year and I can see HuskyPup's view point 100%. We are told that possibly there is a financial disaster round the corner (or maybe its happening right now) and I want to be absolutely certain that I identify any change in earnings / stats as soon as I can. Forget waiting for weekly or monthly trend analysises. They may well be too late to stop the rot.

Keep posting HuskyPup.

My stats (and I definitely look at them daily as any sane person would) are slightly on the up compared to last year's.
I am concerned on a daily basis in comparing stats to last year and I can see HuskyPup's view point 100%. We are told that possibly there is a financial disaster round the corner (or maybe its happening right now) and I want to be absolutely certain that I identify any change in earnings / stats as soon as I can. Forget waiting for weekly or monthly trend analysises. They may well be too late to stop the rot.

Thank you for answering my question as to the point. I think you are dead wrong in terms of business strategy, and I will also suggest that in terms of business decisions, focusing on dailies will more likely lead to to make wrong ones than right ones. Daily numbers mean nothing.

It's simply a bad strategy, since, in business, by the time you see the impact of a problem, it's too late. I've functioned, for the last 6-12 months on the assumption that adsense income would shrink substantially, and have take steps to deal with that. If anything we acted too slowly, and too weakly, so now we have to struggle a bit.

The point being that it will be too late if you wait until you see impact on dailies, and again the dailies, as zett has pointed out are so random in terms of their variance, you can't use them for anything.

As for being sane, I'm sorry, but there is NO positive reason to worry about a daily number, NO MATTER how many visits you get. Period.

..unless you think it's entertaining. And, it's damaging because it causes you to speculate about a result which is vapor.

I check my stats daily and I believe value can be gained from doing this. Of course over a month the stats will be more represent a truer picture but daily stats can be useful to analyse, especially if any anomalies occur, which is identified in the thread.

Coachm you talk about business strategy and how you have planned ahead, I presume you are aware that in business not everything goes as planned and when this occurs you need to identify the reasons behind it. You however seem to suggest huskypup should carry on regardless as he is 'already too late' ?

As for describing daily stats as 'vapor', that's ignorant. Daily stats may not be as important as monthly but they still need attention. I don't suggest you spend time analysing results all day but give a little time to consider what made today a success/failure with anomaly results.

I check my stats daily and I believe value can be gained from doing this. Of course over a month the stats will represent a truer picture but daily stats can be useful to analyse, especially if any anomalies occur, which is identified in the thread.

Coachm you talk about business strategy and how you have planned ahead, I presume you are aware that in business not everything goes as planned and when this occurs you need to identify the reasons behind it. You however seem to suggest huskypup should carry on regardless as he is 'already too late' ?

As for describing daily stats as 'vapor', that's ignorant. Daily stats may not be as important as monthly but they still need attention. I don't suggest you spend time analysing results all day but give a little time to consider what made today a success/failure with especially with anomaly results.