Re: online targeting

From: Clint Ballard <clint_at_accelerationsw.com>
Date: Thu, 4 Feb 1999 08:09:26 -0600 (CST)

DOUG BATES WROTE:
>Interesting analysis, but the premise is very limited in
>scope. It assumes a situation in which the targeting is
>within the *exact* same universe as an untargeted ad. That
>can be true in some cases, but it is of limited
>applicability. "Targeted" and "untargeted" are vague terms.
>Perhaps you were assuming in your post that you were talking
>about targeting within the exact same universe. I didn't
>make the same assumption.
>
>Moreover, the analysis assumes that all potential buyers are
>within the targeted universe. I don't think that's likely
>to be true. If it is not true, then you cannot say that
>then the value of the| targeted advertising is N times the
>untargeted. You must subtract from N the potential buyers
>you've excluded by your targeting.
>
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Doug,

Actually, the Linear Valuation Theory does not have ANY
population overlap assumptions and in fact "Targeted" and
"Untargeted" could be from entirely disjoint sets of people.
 My intent is to define "Targeted" and "Untargeted" based on
MEASURABLE characteristics that are of primary importance.
For the study that started this thread, the conversion rate
of clickthroughs from Targeted was 30 times that of
Untargeted. I hope you can see that it doesn't matter what
level of overlap the two sets of people have, if people are
doing the ultimate desired action (purchase, etc.) it is not
relevant what web site they happened to be on when they
clicked on the advertisement. I guess if you are saying
that the value of a conversion of someone who was not at the
Targeted site is somehow more valuable than someone who was
at an Untargeted site, then what you are saying could make
sense. The only case I can think of where this would be
true would be if the conversion being measured is not the
ultimate goal, but an intermediate step and Targeted had a
higher rate of subsequent post-conversion value. My
suggestion would be to measure this and compute N based on
the ultimate conversion desired, but if this action takes a
long time to develop, then the premium for Targeted can be
estimated and factored into the current Linear Valuation
equation:

TP(1/N) = (N*P)/2, where P is the multiple of desirability
of a conversion from Targeted over Untargeted.

The value of the potential buyers in Untargeted is built
into the conversion rate, so it is not necessary to subtract
from the value of Targeted. Any conversions happening in
Untargeted has already increased its value and therefore
reduced N. As explained above, unless you care about the
origin of your customers in a significant way, a conversion
is a conversion is a conversion and it doesn't matter what
population group they are from. Basically, all I am saying
is that if a clickthrough from Targeted has 30 times the
conversion rate, then it is worth 30 times more. The only
relevant question is how much more did such a clickthrough
cost and the above equation produces the relative value of
impressions from Targeted and Untargeted.

Clint Ballard
C.E.O., Acceleration Software, clint_at_accelerationsw.com
"http://www.clicksales.com/index.html?oa", pay-per-click advertising with a
100%
satisfaction guarantee




========================================================================
------------------------------------------------------------------------
       This week's Online Advertising Discussion List sponsor:
                                Conducent

   Don't Buy Hope! Get a 94% User Response Advertising with Conducent!
     Place Dynamic Ads in Downloadable Software. Invest in Results!
 http://www.conducent.com/advertisers.shtm or eisenstadta@conducent.com

------------------------------------------------------------------------
========================================================================
Online Advertising Discussion List To Unsubscribe send UNSUBSCRIBE
http://www.o-a.com/ to online-ads-request@o-a.com



Received on Thu Feb 04 1999 - 10:53:29 CST