RE: Killing caches (was RE: Pay-Per-Click Question)

From: Alex Chudnovsky <alexc_at_jungle.com>
Date: Wed 06 Jun 2001 12:31:05 -0500

JUSTIN W. HITT <Justin.Hitt_at_hittpansophism.com> WROTE:

> As edge routing becomes more affordable your local log
> files become more useless.
> Most content is static.

No way, at least not for us (jungle). More and more
sites are dynamic, and this trend will continue.
There will be more personalised content, which isn't
feasible to make static. Images will probably
remain static for foreseeable future as its a bit
too expensive to manipulate them in real time at
the moment (I am talking about highly trafficed
sites here).

> Browser side measurements of traffic are always going to
> be better than anything you can measure on your server.

Not really, definitely not for dynamic sites.
To be honest, I don't really care about aggregated
traffic data, its no longer the THING to track for
IPOs, if you know what I mean. Log files give me a
LOT of information, and most importantly they give
me personalised data which I can use in my future
campaigns.

Tracking _ALL_ browser's do will give even more
information (nice), however they will be less reliable
because log files contain whats actually HAPPENED on
my site. Caching is not _THAT_ big issue for dynamic
sites, even though IE5 is abusing page caching big
time.

> It took 14 days for my computer to crunch the logs, and
> a few more days to check the math.

Well, there are solutions to big log files, perhaps
just like you we found that WebTrends just doesn't
cut it. Personally I prefer to have _ALL_ log files
in data warehouse which enables me to run customer
queries which would link use actions to actual sales.
Sure, it takes time to process log files, but a custom
programming helps a lot
;)

Again, let me stress -- I care little about aggregated
stats, they are nice to have and see how your site is
doing, but not as good as SALES data that I get anyway.

Sure if you don't sell anything, then perhaps traffic
data is the only thing you can talk about, but to be
fair days when this was happening right and left are
gone.

Bottom line:

Whilst Alexa's data can be used for some useful things,
they are of questionable quality and can't be taken as
absolute numbers, only as relative (MAYBE). Perhaps the
fact that these were free made it worthwhile, but last
time I checked they wanted money for that, and hey, if
pay money, then the data better be good!


Best regards,

Alex Chudnovsky



Received on Wed Jun 06 2001 - 12:31:05 CDT