Loading ...
Sorry, an error occurred while loading the content.

Counting View-Based Activity

Expand Messages
  • sally krumholz
    Hi all, I have a question for you folks who deal with media-related activity. As you know ad servers report on view and click-based activity/transactions. I m
    Message 1 of 2 , Sep 10, 2009
    • 0 Attachment
      Hi all,

      I have a question for you folks who deal with media-related activity. As you
      know ad servers report on view and click-based activity/transactions. I'm
      running into an issue where a client is completely discounting any
      view-based activity/transactions. Unfortunately because Omniture charges a
      boat-load for integration with our ad server, we're not able to get blended
      reports. Client felt the cost was too high - which I agree it is.

      However this leaves me with a problem. Client is assigning a low % for
      "crediting" view transactions or completely ignoring them. Rather calling
      the numbers we are seeing in our ad server as "discrepencies".

      While I'm accustomed to click load discrepencies between ad servers & web
      servers, I take issue with completely disregarding view-based transactions
      since they cannot be directly tracked to a click from an ad.

      Any insights here or am I completely not understanding the problem.

      Thank you,
      SK
    • David Simmons
      Hi Sally, By view, I m assuming you mean a banner ad impression? Would you say that this is a direct response campaign? And when you say that the client is
      Message 2 of 2 , Sep 10, 2009
      • 0 Attachment
        Hi Sally,

        By "view," I'm assuming you mean a banner ad impression? Would you say
        that this is a direct response campaign?

        And when you say that the client is discounting view-based transactions
        are you saying that they don't acknowledge that exposure to the ad, in
        the absence of a click, has contributed to any transaction? If so,
        what's the client's basis for this argument?

        Have you done any ad effectiveness research for this client, either for
        this campaign or during previous campaigns, using other methodologies
        (e.g. DL surveys, focus groups)?

        Alternatively, with the caveat that I am not an expert in banner
        effectiveness research, are there any periods in the ad server &
        transaction data during which, assuming you have a large enough sample
        (and after adjusting for the usual disparities among data sources e.g.
        time zones, geo-IP filters, etc), you can demonstrate that a
        statistically significant uplift in transactions took place without a
        concurrent increase in ad clicks? Typically, a lot of visitors who are
        exposed to banner and online video campaigns never click on the ads or
        other associated links and instead immediately use search or type in the
        site's URL in their browser (this is very common for email too actually).

        Or else, using a full, clean dataset, you could run some regressions to
        determine to what extent ad impressions & clicks predict transactions on
        an hourly, daily, or weekly basis...

        HTH

        David

        sally krumholz wrote:
        >
        >
        > Hi all,
        >
        > I have a question for you folks who deal with media-related activity.
        > As you
        > know ad servers report on view and click-based activity/transactions. I'm
        > running into an issue where a client is completely discounting any
        > view-based activity/transactions. Unfortunately because Omniture charges a
        > boat-load for integration with our ad server, we're not able to get
        > blended
        > reports. Client felt the cost was too high - which I agree it is.
        >
        > However this leaves me with a problem. Client is assigning a low % for
        > "crediting" view transactions or completely ignoring them. Rather calling
        > the numbers we are seeing in our ad server as "discrepencies".
        >
        > While I'm accustomed to click load discrepencies between ad servers & web
        > servers, I take issue with completely disregarding view-based transactions
        > since they cannot be directly tracked to a click from an ad.
        >
        > Any insights here or am I completely not understanding the problem.
        >
        > Thank you,
        > SK
        >
        >
      Your message has been successfully submitted and would be delivered to recipients shortly.