If we think about this logically, using Google Analytics data as a
ranking factor would be really hard to police.
For instance, using filters could manipulate data to make it seem
like the site was performing in a way that it isn’t really.
What is good performance anyway?
High “time on page” might be good for some long-form content.
Low “time on page” could be understandable for shorter content.
Is either right or wrong?
Google would also need to understand the intricate ways in which
each Google Analytics account had been configured.
Some might be excluding all known bots, and others might not.
Some might use custom dimensions and channel groupings, and
others haven’t configured anything.
Using this data reliably would be extremely complicated to do.
Consider the hundreds of thousands of websites that use other
analytics programs.
How would Google treat them?
Verdict: SEO myth
HOW TO SPOT SEO MYTHS: 20 COMMON SEO MYTHS, DEBUNKED