ShareThisPiggy backing conversion tags and discrepancies
Hi Admonsters,
I'm trying to work out the likely size of discrepancies when piggy backing tags. The situation I am trying to analyse is this:
Tag 1 - I have a tag on a site i.e. a doubleclick floodlight tag. This gets fired when the page loads and has an expected discrepancy as all tags do.
Tag 2 - This tag is sent by tag 1 and loads once the response to tag 1 has fired.
My question is what factors affect the discrepancy and does anybody have any data that would show how varying these factors affects the discrepancy.
the discrepancy should be made up of:
Time taken to look up the address in dns
Transit time to send the request to the servers
Response time of servers.
How would the discrepancy be affected if tag 2 is on the same domain as tag 1 or if its on a different domain? If they are on the same domain the tag2 discrepancy doesn't need to do an extra dns lookup which saves 20-120ms waiting for a response.
Regards
Pete
Comments
it's a bit hard to diagnose as people can be looking at the page through different connections. Basically it goes with the same thing as most discrepancies in the industry - up to 10%, usually closer to 5%. It also depends on the method of firing the other tag, is it being written to the page and fired or is it a 302 redirect from the first page? other things that are outside the technical realm which can show up on reports is the post action window, filtering methods and whether the reports are on the same timezone.