NONE: Re: ONLINE-ADS>> Charging for cached ad views
Re: ONLINE-ADS>> Charging for cached ad views
Michael Tanne (michael_at_tanne.com)
Wed, 18 Feb 1998 22:41:20 -0800 (PST)
Tom Hespos wrote:
>IMGIS can also bust cache. I'm not sure whether they have the ability
>to report the cached ad views separately, so I won't comment on whether
>IMGIS-enabled sites and networks bill out the cached activity across the
>board.
Tom summarized well the issues relating to counting cached impressions.
I'm VP of marketing at IMGIS. Tom's correct, IMGIS does
count cached impressions, and has done so for a long time.
Both MatchLogic and IMGIS do this to get to accurate
numbers, which is ultimately where we all need to go. CPMs
and Click-throughs that have embedded error are ultimately
not in anyone's interest even if some are used to it. There
are still ways to negotiate bonus impressions, and CPMs are
always a negotiated figure. Ultimately the supply of
eyeballs and the demand for ad space will determine the
market price, and not whether there is caching.
Some background about how IMGIS counts cached impressions
for those who are interested: Two requests are made for an
ad. The first (very fast) request is made to select which ad
to serve. The second request is made to get the creative
itself (image or other media). Because it's not an image,
the first request is not cached. this ensures the request
is unique and counted. This allows us to track reach and
frequency. The request for the media is allowed to be
cached. This has the benefit of faster serving, since it
may be cached at a proxy server of your company, ISP or
online service, or even by your browser. It also separates
the media request from the typical <img src=myad.gif> tag,
and allows just about any rich media to be sent.
The number of cached ads is hard to measure or predict, but
the number of ad requests can be precisely measured.
Remember, that whether counting cached ads or not, most
systems do not guarantee the image was completely loaded. A
small percentage of users hit the stop button before the
download completes. This is less of an issue than it may
seem: Incorrect counting occurs only if the users stop
after the ad request and before the image is finished - a
relatively narrow window relative to the download of a whole
page. Even most on-site ad servers like NetGravity count
based on the image request and don't wait for a complete
download. The miscount due to stopped loads is tiny compared
to the inaccuracy due to not counting cached ads, and it is
predictable.
Does anyone know of a source for information about how many
people stop ads? Maybe someone running a panel and doing
direct PC-top measurement can give us an estimate.
Hope this helps.
Michael
---------------------------------------------------
Michael Tanne mtanne_at_imgis.com
Vice President, Marketing (408) 873-3680
IMGIS, Inc.
---------------------------------------------------
----------------------------------------------------------------------
This week's Online Advertising Discussion List sponsors:
ICONOCAST and HotBot
"ICONOCAST is without question, the coolest, shortest,
best written Internet thing I read every week."
Seth Godin, CEO, Yoyodyne http://www.iconcast.com
---
HotBot, rated the #1 search engine by CNET, has keywords available now
in these categories: computers, software, shareware, Internet, games,
music, and more. Contact Rick Boyce at 415.276.8440 or rick_at_wired.com.
-----------------------------------------------------------------------
Online Advertising Discussion List To Unsubscribe send UNSUBSCRIBE
http://www.o-a.com/ to online-ads-request_at_o-a.com
