Our supposed test on nofollow vs dofollow links ended up being a test on building trust flow. In that test, we have taken a peak on the inner workings of Majestic and its trust flow metric. For this test, we check if we can determine how many trust flow 10 links a site needs to increase its own trust flow to 10.
For this test, we just had 1 test page where all the links are going to. It was tricky finding a reliable source that was willing to provide one TF10 link at a time.
Link One
We sent one link (comment) on April 9 and submitted it to Majestic crawler. The test site where the link will be sent to had TF0 and CF0. The site where the link was placed had TF10 and CF14. The back link showed up in Majestic four days later (April13), and the data for the site changed on April 18, 5 days after the link was registered by Majestic. The updated metrics of the site once the link has registered was:
- Root Domain – TF2 CF4
- URL – TF4 CF14
Link Two
Another link (comment) was sent one April 14 and submitted to the Majestic crawler. The metric for this site was TF8 and CF9. The backlink showed up in Majestic on April 20, 6 days later after it was added. The Majestic data for the test site changed 8 days after the link was registered (April 28). The updated metrics of the site once the link has registered was:
- Root Domain – TF2 CF5
- URL – TF5 CF16
Link Three
Another link (comment) was sent one April 14 and submitted to the Majestic crawler. The metric for this site was TF10 and CF21. The backlink showed up in Majestic on April 21, 7 days later after it was added. The Majestic data for the test site remain unchanged and the third link has disappeared.
Link Four
Another link (comment) was sent one April 30 and submitted to the Majestic crawler. The metric for this site was TF10 and CF20. The backlink showed up in Majestic on May 9. The updated metric for the site is:
- Root Domain – TF2 CF6
- URL – TF5
Results
Majestic showed 4 links going to the site, only 2 of them are live. Seems to be some sort of resorting comment pages on one of the backlinking sites.
The site went from 0 both in Trust Flow and Citation Flow, to now having a TF4 and CF16.
Conclusion
Full link juice is clearly not passed via the links we acquired. In fact, Majestic may have some sort of spam identifier where a low value link will pass most, if not all its CF, but will only pass a portion of TF. As spammy sites are often identified as a site that has the ration CF 2:1 TD, it seems possible that Majestic’s algorithm has the ability to account for spammy links and passes metrics accordingly.
Another interesting point is the concept of link echoes since the target URL which also happens to be the home page has only dropped a little from its height of TF of 5 and CF of 17, despite losing 75% of the links. As for Majestic, we see that the URL and domain metrics grow at a similar paces but start at different points. The URL TF and CF grew at larger pace than the root domain metrics.
Notes
- This kind of test is super tricky – tracking to see when the links indexed, when the tf and cf change, can be hard to pinpoint as Majestic doesn’t immediately update their scores and will change original crawl dates.
- Additionally, the vendor of the links can lose interest in finding one TF10 link. If that original link changes after it has been placed and we request for a fresh one, it is difficult to get the vendor to replace the link quickly.
- The purpose of going through a vendor, rather than pointing links we controlled was to simulate the types of links that are available to purchase.
- During the test, at least one site closed their comment section – we guess because they noticed the spammy comments.
- Not having complete control of the links makes this a very messy test – which should provide some insight on purchasing links that you don’t control and pointing them to your money sites.
We have more tests on Trust Flow and other metrics that may interest you. Check out our test articles for more details.