What they did
Zyppy crawled 23 million internal links across about 520,000 URLs on 1,800 sites and joined every URL to its clicks in Google Search Console. They then checked how link count, anchor text variety and exact-match anchors lined up with search traffic, re-running the anchor analysis three times to see if it held.
What they found
Pages with 0 to 4 internal links averaged around 2 clicks from Google; pages with 40 to 44 links pulled roughly four times that. Past about 45 to 50 links the gain reversed. Pages with at least one exact-match internal anchor had at least five times the traffic of pages without one, and wider anchor text variety tracked with more traffic on every run.
How much weight to give it
The dataset is large and tied to real Search Console clicks, so the patterns are worth attention. Re-running the anchor analysis three times is a fair robustness check.
What it doesn't prove
It is correlation. Important pages tend to gather both more internal links and more traffic, so you cannot read the 5x as 'add an exact-match anchor and win 5x'. Clicks come from Search Console, the sites are not a random sample, and the page is updated over time, so use it to prioritise, not as a causal rule.