• Chris Reynolds

    It seems a bit like reinventing the wheel to build this network of redirectors yourself – what about Atlas, Doubleclick, Mediaplex etc. Not to mention all of the bid management tools out there.

    At least that way if it does go down you have someone else to blame!

  • http://www.rimmkaufman.com George Michie

    Hi Chris,

    The point is that SOMEONE needs to make sure these critical details are handled. Whether it’s RKG or another agency, or a third-party tool, it’s important for advertisers to ask hard questions about the integrity of the data. If the data is wrong, the bidding is wrong and the program won’t be clicking on all cylinders as it should be.

  • http://www.fuelinteractive.com/ briancarter

    George, love your columns. Hey, can you give a source for the 10-30% of lost visits? Are you saying 10-30% of javascript conversion code doesn’t execute? I’ve never heard it that high. Is that number from data on sites you work on, or an industry standard?


  • http://www.brickmarketing.com nickstamoulis

    Hi Chris, great post, but I too would like to know where you are pulling the 10-30% lost visits metric? I have never heard or experienced that much of a discrepancy.

    I normally use analytics and conversion tracking that the search engines provide and have achieved excellent results for many years. I have tested many of them out there and have found issues with stability, less accurate data and a learning curve as I try to learn many of these PPC management applications…. Anyway are there any 3rd party products you would recommend that might fill in this 10-30% gap?

  • http://www.rimmkaufman.com George Michie

    Nick, Brian, thanks for your questions and feedback.

    The 10 – 30% figure comes from data audits we’ve done with clients and from our own use of both technologies on the same site. It’s a big number, and a big range within that number based on page load times.

    We built a decent tool to measure page load times and found that on days when the page load time is particularly slow due for whatever reason, the discrepancy between what we see and what the javascript tag shows spikes as high as 50%.

    Again, this is beyond just credit allocation discrepancies, and it doesn’t impact all pages by the same amount. Sometimes the loss is greater when the destination page is a search results page, sometimes it’s less. Same with product pages. Much depends on the size and number of images on the page.

    Nick, I’m not sure what the third party tools use for tracking results. I recommend asking hard questions about whether they use a redirect (good), whether the redirector has to do a database lookup (bad), and whether there is redundancy among redirectors in case of network outages (essential).

  • http://www.brickmarketing.com nickstamoulis

    Hi George – Thanks very much for the additional information and info regarding the third party tools, I will keep this in mind as I continue to search for a good PPC tracking provider in the future.