I've been doing some analysis on the Ad Impressions, Clickthrough, Rates on the Webtrends Application - exporting and manipulating the data. It became clear to me, everyone likes "rates" but they are sometimes a little misleading, especially to the non-analyst. It seems like the lower the visits, the greater chance of rate fluctuations. So I went about trying to develop an algorithm to only concentrate on high value ads - with more user interactions, the rates become more reliable.
I recently noticed Google were thinking the exact same way and they created a new "weighted sort" algorithm - http://bit.ly/9nLyCe
This is extermely useful. If you now have a really low or high rate based on this, it really gives you justification to take action.
I don't think GA has fully rolled this out yet but are webtrends going to have an answer to this as well?
It's a great feature, something that seems to make perfect sense once you see it in action. Many customers are doing similar things that you are -- manipulating the data in Excel or another tool -- but having this in the product interface would call out more meaningful information right away. We've been looking into something like this and will keep you updated on our progress.
If you see other features like this that clearly would add value to your Webtrends experience, please let us know on this forum.
CHAMPI’m wondering if I'm missing the point1I didn't know Google A. had added this. It makes perfect sense. But once again Google has done a black-box implementation where the rules are (as far as I can tell) hidden.
Something very close can be done in WebTrends now without the need for dumping data to Excel. The key is Query. Do a Query for "visits > 1000" or whatever level you think is appropriate. Save it with a bookmark.