NONE: Re: ONLINE-ADS>> Official Bail-Out Rate?
Re: ONLINE-ADS>> Official Bail-Out Rate?
Mark J. Welch (markwelch_at_ca-probate.com)
Tue, 17 Mar 1998 12:30:57 -0800
At 11:09 AM 3/16/98 -0800, Christine Bensen wrote:
>Has anyone out there been measuring % difference between people who click
>on advertising vs. people who actually wait for the resulting content to
>load. I heard a while back that 10-15% "bail out" was average. Any
>official or unofficial reads?
I'm not sure this can ever be accurately measured. What
looks to you like a "bail-out" might be a successful
clickthrough in which the visitor sees your content:
some visitors may see your page because it is loaded
from a cache or proxy server, and thus you won't get
an accurate count. In the other direction, your server
might report that it delivered your page to the visitor,
but the visitor might not have actually seen the page
due to a "bail-out."
In my analysis of my FlyCast ad campaigns during January,
I discovered an 18% difference between the clicks that FlyCast
reported and the pageviews at my site. [During the first eight
days of a campaign (January 13-20), FlyCast claimed they
sent me 2,242 visitors, but my logs reflected that I received
only 1,894 visitors from FlyCast, a difference of 348 clicks,
or 18% of the pageviews I actually received. See my
report at http://markwelch.com/flycast.htm] I have no
way of knowing how much of this variation was due to
"bail-outs," technical problems, or caching.
A more important problem is measuring the "depth" and
"persistence" of visits (meaning the number of additional
pageviews and the duration of the visit). Most visitors
who find my site through a link from Yahoo or another site
will explore several pages. In contrast, people who click
on ads quite often merely look at my home page and then
vanish (possibly they are bookmarking the page, or they
might decide they are not interested). While a single
visitor's failure to explore my site isn't a failure of my
ad campaign, I do watch the trends carefully.
I have found that certain sites consistently generate
clickthroughs that have a depth and persistence of zero:
the visitors NEVER explore any other content at my site.
My suspicion is that these sites are somehow "cheating,"
but even if they are not, the pattern of activity is not one
I want to pay for, and I terminate my advertising on those
sites.
Mark J. Welch
<markwelch_at_ca-probate.com>
----------------------------------------------------------------------
This week's Online Advertising Discussion List sponsors:
Texas Monthly, Wired Digital, and ICONOCAST
Looking for Texans? We've got 'em at the Texas Monthly Ranch,
http://www.texasmonthly.com. Plus travel, food, events, culture and
more. Contact Jena at 512.320.6928 or jenab_at_texasmonthly.com.
---
Brand Impact Studies from Wired Digital deliver mission-critical data
on the branding value of Web ads - for less than $5,000 per campaign.
More info? Contact Rick Boyce at rick_at_wired.com or call 415.276.8440.
---
"ICONOCAST is the best source of aggregated data and insight into Net
stuff that I receive." -- Rita Belle, Marcom Director, Zip2 Corp.
E-mail majordomo_at_iconocast.com with "subscribe iconocast" in body
-----------------------------------------------------------------------
Online Advertising Discussion List To Unsubscribe send UNSUBSCRIBE
http://www.o-a.com/ to online-ads-request_at_o-a.com
