The question of how much value nofollow links give to a website is a hotly
Some studies by Matt Diggity strongly suggest that Google can and does use them as a ranking factor – and that they may even pass some form of Page Rank (or however you want to call ‘link juice’ these days).
Other experts such as Charles Cloake see nofollows as having value only as part of a pillow linking strategy. In this view, nofollows add trustworthiness, some traffic and a natural look to your otherwise spammy (but super-powerful) link profile. But they don’t believe that Google uses nofollows to provide any SEO value – they just help you look less spammy than your rivals when the banhammer comes down.
So, there’s a ton of disagreement in the SEO world about the exact value of nofollows, and if they can pass any juice.
But most agree that a definitive test would be a hard one to create, as you would have to remove all other variables to come to a firm conclusion either way.
Over Christmas 2018 (too much wine?) I decided to create exactly such a test 😊
I built two nearly identical websites in the same niche and built the same backlinks to each site – the only difference is that one site got follow links and the other one got nofollows.
Let’s have a look at the details of the test, and how I controlled each variable as best as I could.
The Test Details
The niche I chose for this was the adult niche. Without going into specifics, the sub-niche has high competition, medium-sized volume but a high value of each conversion (one sale is valued in the hundreds of dollars).
I chose this particular sub-niche because:
1) It’s a spammy topic and even less respectable than casino – so is extremely unlikely to pick up any backlinks by accident (which would ruin the experiment)
2) I have over a decade’s worth of experience in this niche, so know how to get a new website out of the Google sandbox fast (so I can present results quickly)
3) I have built up a lot of relationships with other webmasters in the niche, so I can ask them to build follow or no follow links to my sites
4) It tends to share a similar backlink profile to websites in the iGaming niches, so casino link builders can probably draw similar conclusions
These were new domains and had never been registered before.
They were both .com TLD
The domains were newly created brand names and not Exact Match Domains (EMDs) or even Partial Match Domains (PMDs).
The domains had the same structure.
They were purchased at the same registrar and both had who.is privacy turned on.
The websites themselves were built on an identical codebase (not WordPress), using the same amount of H1s, H2s and images.
The text used two variants of spun content to be similar enough to maintain the same quality, but different enough not to be seen as duplicate by Google.
The two sites were hosted on the same server with the same IP address (in this niche, you can get away with leaving amazingly obvious footprints). The two sites were not interlinked.
By 24th April, backlinks from 30 different domains were built using similar velocity (within 48 hours of each other).
The links were a mix of images, menu links and contextual links.
Most were placed on the homepage, and some on inner pages.
The anchor texts were a mix of brand names and naked URLs. No keyword-based anchors were used.
The links were built only from sites within the same sub-niche, so they were highly relevant.
Here’s is the distribution between follow and nofollow for each site:
The Nofollow website received 100% nofollow links
The Follow website received 98% follow links
Google Search Console was installed on both sites for tracking purposes. Separate GSC accounts were used for both sites. No fetch/render or index requests were made on either site – the Googlebot had to find these sites on their own by following links on the web.
As well as measuring traffic and rankings via GSC, I tracked both sites and their links with LRT and Majestic. I was interested in the values these would give as many SEOs use these tools to judge links.
Traffic and keyword rankings (data from Google Search Console)
The nofollow site does receive impressions and rankings, but only for its brand name
The follow site receives impressions and rankings, for its brand name AND money keywords in the niche
Yeah, I know…. these are low volumes via organic search….but remember that
1) it’s early days and
2) each conversion in this nice generates hundreds of dollars.
For the nofollow site, the power value was passed but trust value was not
For the follow site, both power and trust value were passed (as expected)
For the nofollow site, the citation flow was passed but trust flow was not
For the follow site, both citation and trust flow were passed (as expected)
Conclusions from the test results so far
Google Search Console Results:
Both websites were successfully crawled and indexed by Google, just by the spider crawling the web – so we know that the Googlebot does visit nofollow links.
The nofollow test website is so far only ranking for the Brand name.
The follow test website (the control) on the other hand picked up impressions for some highly competitive money keywords in this niche.
So far, it seems that nofollow links don’t pass on any serious link juice value.
LRT and Majestic values
On the other hand, both the follow and nofollow links DO pass citation value for the purposed of Majestic and LRT, but none of the trust values.
Some SEOs believe that the ratio between Majestic Trust Flow (TF) and Citation Flow (CF) should not be higher than 1:2 (or 1:3). The belief is that too high a ratio suggests that a website has too many spammy inbound links compared to how authoritative it is. Porn tube sites are good examples of this.
This test shows that these trust metrics might also be influenced by the number of nofollow links pointing to the website.
Those who rely on Majestic and LRT trust values to judge a site should also check the backlink profile of a potential link partner to check out the ratio of follow/nofollow links. It might just be that they have a ton of backlinks from Wikipedia 😊
This test is only 4 months old, and we shouldn’t draw any firm conclusions just yet.
In stage 2, I want to start adding keyword-based anchors to both the nofollow test website and the control site.
It will be interesting to see if the nofollow website can rank for money keywords when those are used as anchors.
I will report back later in the year to tell you how the test is going.
Happy link building!
PS: Just because nofollows don’t appear to pass link juice, DOESN’T mean that you shouldn’t build them. In fact, in most niches (including iGaming niches) the average link profile of the highest-ranking sites contains a good ratio of nofollows. If you want to emulate these high-ranking sites and appear natural in Google’s eyes, you should build nofollows. Just try not to pay for them 😊