DMMA

Would be nice to have historic data for comparison. Also, not to seem dim-witted, but the first set of data refers to “unique browsers” – if so, what would the default browser be, or does it refer to sa unique users based on IP? 

Thanks for the comment Terence. The historic data can be found when you click on the research tab on the site. UB refers to a Unique counted once on each platform/browser used to access the site.

Please note that the SA Unique Browsers number is merely an addition of all traffic across DMMA sites and doesn’t take duplication across sites into account.

Hi Justin

This is actually incorrect figures are not totalled – look out for the official response from Effective Measure to explain.

The DMMA

Unfortunately Mr McCall’s comment is incorrect. Effective Measure does not simply add all the websites unique browsers to get a total industry figure. January 2012′s unique browser figure for South African traffic was 14,148,373 while if we sum up all websites we obtain a figure of 27,346,589. This shows that the total industry figure is de-duplicated and calculated completely separately from the individual websites.

Hi Justin,

It seems there is a misunderstanding regarding the Effective Measure methodology. I am the global Data and Methodology manager for Effective Measure and would like to take the time to explain how some of our figures are calculated.  Effective Measure has tags on each and every DMMA associated website. When a user arrives on one of these website the Effective Measure tag is executed and a combination of first and third party cookies are placed (or updated) on this user’s device. This ensures that Effective Measure is able to count the widest spread of devices as well as ensures the longevity of the cookies because there is less chance of rejection by the device. Each cookie is labelled with an anonymous but unique ID for the user. This ID is sent to Effective Measure for each and every page view and session across all DMMA websites.

At the end of each reporting period (daily, weekly and monthly) Effective Measure then associates these ID’s with the country where the request originated from. This way it is possible to split out South African traffic from the rest of the world. Once all the traffic has been classified to a country, a list of all South African ID’s is created. This list has many duplications because a user could have visited multiple websites on different days. To deal with this an algorithmic sort is performed on the data and then all duplicate ID’s are removed. This leaves a list of unique ID’s for the entire South African market that visited DMMA websites. This list is then counted and becomes the unique browser figure for the period in question.

Once this has been completed each website’s URL definition is then applied to the raw data. A list of all South African ID’s that went to a website is then created. Using the same algorithmic sort and de-duplication function a list of unique ID’s is obtained for each and every DMMA website. Again, the count of all these ID’s becomes the unique browser figure for each website.

The methodology for calculating uniqueness is common across each of different sets of data. However, the actual process is done in isolation for each data set. If the website totals are summed up they are much larger than the market total. This shows that the market total has all the duplicated ID’s across websites removed to ensure that the figure is truly unique.

If any of this is unclear please let us know and I will be more than happy to go into more detail with you. The Effective Measure team is also available to do training sessions if needs be.

Ryan, you have completely missed the boat on this one.  Check you email.