First, each ranking is given a click-through rate (CTR), which is then multiplied by its keyword’s search volume.

Then, in the SERP Features tab, these values are added up for every SERP feature in the set, which produces the total number of click-throughs that each feature receives. In the Competitive Landscape tab, all of the weighted values for a site are added up for every keyword in the set, then expressed as a percentage of the total available search volume.

For an example in the Competitive Landscape tab, let’s look at a keyword set of two for “examplesite.com.” For [keyword one] — which has a search volume of 100 — examplesite.com has a number one ranking. For [keyword two] — search volume of 10,000 — the site ranks number five. If we assume that a ranking of one will earn 20% of click-throughs from the SERP, and that a ranking of five will earn 5% of click-throughs, then examplesite.com should earn 520 clicks from these two keywords. In other words, the site’s share of voice for this keyword set will be 5.15% (520/10,100 = 0.0515).

Example calculation using sample CTR percentages:

 Share of voice = Total click-throughs (520) / Total search volume (10,100) = 5.15%

The higher the CTR value, the more share of voice you earn from that ranking.

STAT’s default CTR model factors in the natural curve that click-throughs typically fall into on the SERP and is designed to give you an accurate directional measure of share of voice. This means that the higher your share of voice is, the more visible you are on the SERP, and the more traffic you can expect to drive.

You can also customize the CTR model in STAT. This is especially handy if you already have a site-driven click-through-rate model that your team uses to estimate traffic from a ranking.

Always keep in mind that high search volume keywords will effectively drown out low search volume keywords when it comes to calculating share of voice, and that close-variant keyword search volumes may over-inflate your total. If you want to monitor share of voice for low search volume keywords or for segments more accurately, then it’s a good idea to create tags or data views that specifically exclude high-volume or close-variant keywords.

Note that zero search volume means no contribution to SoV.

Yes, but it depends on the nature of your click-through rate model.

If you have customized your model with a simple rank-scoring system that is not driven by actual click-through rate data, then it is more accurate to think of share of voice as an abstract directional metric rather than a reflection of the actual amount or percentage of click-through traffic earned. What does that mean? Simply that if you are more visible on the SERP than a competitor, then your share of voice will be higher than theirs, and you can expect proportionally more traffic from that keyword set.

If, on the other hand, you are using an accurate, data-driven click-through rate model (especially one that is site-specific), then you can indeed use share of voice to estimate click-through traffic.

For example, if [example keyword A] has a search volume of 100,000 and your share of voice for that keyword is 25 percent, then you can estimate that you’ll capture 25,000 click-throughs from that keyword.

However, this example assumes that 100 percent of a keyword’s search volume results in a click-through. In reality, most keyword sets do not see a 100 percent click-through rate. To estimate click-through traffic accurately, you would have to adjust the available search volume to account for the searches that do not result in a click.

So, for [example keyword B], the search volume is 200,000 but only 60 percent of searches result in a click. That means that only 120,000 click-throughs are available to actually compete over. If your share of voice for this keyword is 25 percent, then you can estimate that you’ll get 30,000 click-throughs from this keyword.