111
 [ the surest route to our common good]

January 2006

An Internet Advertising Primer
By
Lance Cory Frank
Palm Canyon Advertising


The most basic aspect of advertising is exposure. Long ago, broadcast, print, and outdoor
adopted statistical conventions for measuring exposure. These are ratings, circulation and
GRP showings respectively.

The nascent Internet has yet to establish such conventions. "Hits" is the term commonly used,
but there are at least four different definitions for a hit, each with a significantly different meaning.
They are:

  • Total Requests:

    This is the total number of files requested from the web site's host computer. It is the
    largest, most impressive sounding number and the one most often quoted. Unfortunately,
    it is also the least accurate in determining exposures. It should never be considered as an
    indication of a site's popularity or the value of advertising on it. You only want to care
    about how many times your message will be delivered, not the total number of exposures
    for every advertiser on the site.
  • Page Requests

    This is the number of complete pages requested. If your ad appears on every page, it is a
    conservative measure of total exposures.  If it doesn't, you need to know the page
    requests for the specific pages it does appear on.

    Unlike the linear, two dimensional print model on which most advertising is based, the
    Internet adheres more to a three dimensional sphere. An ad on the home page of a web
    site usually yields the most exposure, but not necessarily because people can enter a site
    from just about any page.  Search engines index every page of a site and return results
    based on the relevancy of the search criteria. For example, at the time of this writing, if
    you were to type "Rancho Mirage Country Clubs" in Yahoo! or MSN Search, the
    HWY111.COM page containing country clubs located in Rancho Mirage will be on the
    first page of the search results.  The user will enter HWY111.COM from that page,not
    the HWY111.COM home page. If you ad only appears on the HWY111.COM home
    page, it will not be seen by the user entering from the Rancho Mirage Country Clubs
    page. Fortunately,you can buy ads on HWY111.COM which appear on every page
    of the site simultaneously

    How big your ad is and where it appears on the page can also influence its effectiveness.
    Generally, the bigger and higher on the page the better.
  • Unique Visitors (Hosts)

    The number of distinct visitors to the site. Taken as a ratio to page requests, it can be used
    to calculate a conservative estimate of how many people viewed your ad and how many
    times they viewed it.
  • Click Throughs

    These discount the value of traditional exposures in favor of the number of times your ad is
    clicked on. This implies all other advertising media are obsolete since they lack interactivity.
    They aren't, and your ad certainly has value whether or not it is clicked on.

There are technical factors which may over or understate these statistics. Caching is a process
which stores pages in individual computers, servers and switching  stations across the global net.
This speeds up the time it takes to recall a page, but since the page is not requested from the host
computer, it does not register as a page request. Unique visitors are pretty much a wash in that
many users may appear to connect as one host, but conversely, one user may appear to connect
as many hosts. Allegations of  "click  fraud" suggest that click throughs can be significantly
overstated. For this reason, Palm Canyon Advertising's rates are based only on the exposures we
know we can deliver, not "pay per click"

If a salesman quotes you "hits," always ask exactly which statistic they are referring too. It's always
a good idea to ask for a print out of their server report before making a decision. There is no
reason but deception for them to withhold this information.

It is important to note that these statistics are meaningless without a defined time frame. There is a
big  difference between 1 million page requests a year and 1 million page requests per month.

Don't get star struck by big numbers, fast talk, and techno babble. Ask questions. If they don't
know, don't buy.

More information:

How The Web Works by Stephen Turner, PHD

A word about demographics:

Who is seeing your ad can be just as important as how many see it. The US Census Report on
Computer and Internet Use
just released is unfortunately already two years old and does not take
into account market penetration of broadband. Other reports indicate a rapid rate of adoption,
particularly among high net worth households.


Click then relax. I'll take care of it.

Table of Contents