The analytics sector is constantly trying to make tracked activity more aligned to the visitors intentions by enhancing how new landings are detected.
All tracking systems will report slightly differently due to the way they work technically, for example, Google Analytics (GA) uses 30 minute sessions, where as we look at visitor landings from different marketing sources, which we feel is a more useful marketing measurement. For example, if a visitor came in via PPC then went away, looked at another site then returned through SEO 15 minutes later, GA would only record the PPC visit, where we would record both.
In August 2011 GA changed the way they calculate re-land detection, which is more in line with the technique we use as it’s considered a better way of recording visitors.
We like to keep in line as much as possible with GA data to help give a consistent data set for our clients, but GA is not always leading the pack. They sometimes create shortcuts to serve so many customers in a cost effective way (like the midnight session re-set that so many people have complained about).
When a visitor navigates around a web site, sometimes the referrer data they send back indicates a land, when really they are just clicking back to the first page, or their browser is hiding or sending the wrong referrer data.
To avoid generating additional land triggers Infinity Tracking performs ‘re-land detection’, this is the process of comparing the current landing source with the last tracked landing source, and deciding whether or not to count the land or convert it to a page view instead. We are enhancing this to now include a 30 minutes session for visitor activity; similar to what GA do.
The calculation detects a re-land and will always convert to a page view in the following situations:
- If the current landing source is exactly the same as the last land source in the cookie. This helps protect against users clicking the back button to their last landing page.
- If the current landing channel is the same as the last land channel in the cookie, and the referrer is empty. This helps protect against users who have browsers that don’t always send referrers on subsequent pages, but still allows visitors to land from a different channel with no referrer within the 30 minute session window.
- If the current land source is direct. This protects against visitors who’s browser clears the referrer mid-session.
Land and Re-Land Scenarios
In summary this enhancement allows the benefit of detecting multiple marketing sources within a 30 minute session but filters out the mis-landings that some browsers create.