In 2009, Google revealed the exact formula that goes into the calculation of how much an advertisers is charged for a click:
It's an inversely proportional math equation, where the cost-per-click (CPC) is equal to an unknown number that is divided by your known Quality Score. Everytime your Quality Score increases, your CPC decreases.
Google has since then made changes to this formula, without divulging everything that now goes into it.
They have, hovewer, let advertisers know that the formula now includes an evaluation of Ad Extensions, even though in practice, Ad Extensions haven't proven to be of significant influence.
Ad Rank is a number that determines the position of an ad in the auction. A higher Ad Rank means an ad will show up higher in the search results.
The ad with the highest AdRank gets the #1 position, and the ad with the lowest AdRank gets the last postition, or doesn't show up at all.
The actual number is not directly available to advertisers, but understanding it is crucial to advertising on Google.
AdRank is directly proportional to an advertiser's bids and Quality Score. Quality Score plays an important role into how AdRank is calculated.
The only two ways of improving AdRank are to increase bids (MaxCPC), or to increase Quality Score.
This graph shows how AdRank increases when you bid from $0.50, to $1.50.
This graph shows how AdRank increases with Quality Score.
Naturally, advertisers usually opt to improve Quality Score as a cost reduction mecanism that also improves visibility.
In 2016, our friends at Adalysis reverse ingineered the exact theoritical weight that each Quality Score component holds.
As you know, there are 3 factors of Quality Score that Google gives access to along with the 1 to 10 scale:
For each of these factors, Google attributes one of three qualificators:
Brad Geddes and his team analysed their data to discover the actual weights of each factor, and uncovered a "point" system equivalent to "Above Average", "Average" and "Below Average", as follows:
And the corresponding equation that tranlates it into the visible Quality Score: