How Google calculates Quality Score
06 Mar 2019 - Steve Robinson
There are many things that can be done to improve the efficiency of your paid search spend; budget allocation, bid modifications, extensions, bid changes, negative keywords and keyword mining to name a few. Arguably the most important is working to improve your quality score.
Google have always been very guarded as to how they calculate quality score, however over the years they have gradually given more guidance as to the important factors. They now provide, for each keyword, an assessment against three components; expected click-through-rate, ad relevance and landing page experience. This makes perfect sense as, if you are doing the right thing by searchers, you want to encourage ads that are so relevant that they are clicked on often, and when the searcher arrives at your website they also find content relevant to their search.
These assessments, above average, average and below average, certainly give you some guidance where to focus your efforts. However one question remains — how do these relate to the quality score number?
Over years of research we have come across many theories as to how Google comes up with the quality score number, mostly from large paid search software companies in click-bait white papers, often quoting comments that have reportedly come from Google themselves. Examples are:
- A company that quoted the % influence by factor; landing page 39%, expected CTR 39% and ad relevance 22%.
- Another company that devised a formula using the assessments and assigned a score against each one. With an assessment of “average” in all three components giving a score of 5.5 it’s a non-starter.
Our analytical investigations led us to the discovery that, far from being of complex origin, the quality score has much simpler roots. The quality scores of 1 to 10 can be directly mapped to the 27 combinations of assessments of expected CTR, ad relevance and landing page experience. We discovered this through access to multiple data sets, across a number of clients, and we know that for each of the 27 combinations only one quality score exists. The diagram below demonstrates this:
Check this for yourself. In Google Ads create a custom report, add the three components and quality score and you will see the combinations above.
This makes sense with some of the things Google has previously said. All keywords start with a score of 6, which is an average grade for all three components, and it is only after Google has time to grade these that you score starts to move up or down.
We believe this insight could make a big difference to your PPC optimization strategy. By understanding the contribution that the three component parts make to quality score, and your current grading against these, you can focus your efforts towards the areas you think are easiest to impact in relation to the cost saving they will generate.
Our deep knowledge of Google Ads, and our world class analytical capability, led us to create a suite of over 60 reports to manage paid search optimization. These reports are in three groups:
- General Analytical Reports
- Quality Score Focused Reports
- Task Based Reports
Specifically looking at quality score we have reports that track average quality score and weighted quality score over time as well as the component parts:
We also have reports that very visibly show our insight in action for individual keywords: