After having used some diagrams from the SISTRIX Visibility Index in this blog and seeing how the monthly IndexWatch is based on this information, I would like to explain a few things about this value, the idea behind it and the way it is calculated in a bid to make the data more transparent.
The idea this value is based on, is the fact that, in most cases, simply collecting some keyword-positions and displaying them does not paint a realistic picture of the success or failure of a domain in the (Google-)SERPs. The problem is that, either the number of keywords that are being watched is too low, or – and in my opinion, this happens more often – “wrong” keywords are implemented: siteoperators have different ideas on keywords than those searching which causes any observation to “miss the market”.
To solve this problem, the consideration was to regularly monitor a (larger) number of keywords and generate an index from them. For the SISTRIX Visibility Index, we use a keyword-base of a quarter million keywords of which we use the first 100 results each – giving us 25 million data points. The keywords are chosen as to give us a good selection of the German search-behavior: we include large “blockbusters” like “credit” or “real estate” as well as searchqueries from the (extreme) longtail of the spectrum and everything in between. A small number of keywords are regularly determined based on current trends and results – that way, we will also show queries that are extremely potent for a while like “USA elections”.
Depending on the expected clickrate for the different positions (here we differentiate between the keywords, seeing how navigational-searches like ‘ebay’ have a much higher clickrate on the first positions than information-driven searches, for example) and searchvolume for these keywords, we are calculating a weighted value for each of the 25 million data-points. The sum of these values will then make up the SISTRIX Visibility Index. We considered including more values into the index-figure, like the click-price of the keywords in the SEM-area for example but refrained from doing so after tests showed that, on the one hand, these values can not be properly compiled and, on the other hand, they were shifting the significance of the resulting index-figure in a direction that was often unintended.
The nice thing the SISTRIX Visibility Index enables us to do, is judging the success of a domain (or subdomain) in the SERPs at a single glance, as well as observing the development thereof over time (we are already collecting these data for slightly more than a whole year at this point). For the sake of clarity, here some examples: Yigg botches the relaunch, Webnews temporarily suffers from a duplicate-content-problem but is able to get a grasp on the situation. The “Peoplesearchengine” Yasni.de has a head start but their competitor, 123People.de, is able to grow at a faster rate in recent month and weeks, which caused them to pass Yasni in the Visibility Index. At the launch of the SISTRIX Toolbox, I forgot to block large parts of the tools.sistrix.de subdomain – Google was quick and bestowed plenty of traffic to the site after the indexing but I still threw it out of the index. The scope of this index is manifold and I am already playing around with some ideas which I will realize in the coming weeks.
You will find even more information about our SISTRIX Visibility Index in our SEO knowledgebase Ask-SISTRIX: What is the SISTRIX Visibility Index?