A short history of backlinks
Since the very beginning of the internet, backlinks (links pointing from other sites to yours) have been used to analyse a page’s importance. It’s a foundational principal of a network. The quantity and quality of inbound links to a node can be used to determine the likelihood of its importance. The more quality links, the more importance. PageRank, the algorithm Google was built upon, was based on this principal.
In 2000 Google released a toolbar extension that showed the PageRank value for any page on the internet. Pandora’s Box had been opened and people immediately started to reverse-engineer PageRank. If you could figure out how Google was calculating rankings then you had the power to optimise your site. Considering that on average the 1st position gets 10 times the amount of traffic as the 10th position for any given search term, an optimised site could lead to huge amounts of organic traffic.
It didn’t take long for the importance of backlinks to permeate through the SEO community and all sorts of unintended niches formed. Link farms, high authority domain acquisitions, and link networking all moved to the forefront of SEO. There are stories of pages solely filled with links, others were designed to render completely different sites for crawlers and users. Forums and content aggregators became oceans of link spam and the art of PageRank ‘sculpting’ was all the rage.
By 2005 Google decided that something needed to be done. In a flurry of updates, collectively named ‘Jagger’ and ‘Big Daddy’, the rel=”nofollow” tag was released. The tag simply tells crawlers that the link it’s attached to is not endorsed by the domain it’s found on and no PageRank value is to be passed on. Webmasters implemented the tag on all user-generated content, and the majority of link spam became yesterday’s problem.
The tag was such an effective and easily implemented solution that it’s usage spread beyond pure user-submitted content and into almost all content-based sites. It’s now rare to find outbound links without the nofollow tag outside of pure editorial content, and even then it can be difficult. Analysis of this Forbes article shows that the main content has eight outbound links, all to various news organisations’ websites. Forbes’ referencing of other sites should contribute to their scores in PageRank’s original model but the blanket use of nofollow prevents that.
Google now recognises that nofollow tags have become a problem. In September 2019 they introduced two new tags to add more nuance to outbound linking and announced that they will be taking nofollow tags solely as ‘hints’ from March 2020 onwards. Despite the new tags, it’s hard to convince sites to leave an easy blanket approach and adopt more tedious categorisation.
From a growth standpoint
As young companies like us look to gain ranking authority in today’s search landscape, a quality backlink profile remains a key piece of the puzzle. There are a myriad of strategies and tactics that people will recommend but the fundamentals of most of them boil down to quality content means more attention.
Kickstarting the process can be a difficult first step. For the more editorially inclined sites, you’re at the whim of pitched writers and editors, but there are plenty of authoritative sites that feature direct user-submitted content. Are any of them viable options for building a backlink profile?
Here’s our rundown:
It seems like a natural place to begin for a link building strategy due to its size and potential user reach, but in actual fact its size counts against you. There’s no way for Facebook to review everything posted on the platform so every single outbound link gets an instant nofollow tag attached to it. Not the place to build backlinks.
Unsurprisingly, Twitter follows a similar line to Facebook but executes it in a different way. No links contain the nofollow tag and instead Twitter uses their robots.txt to tell crawlers not to cover most pages. Interestingly, they specifically allow the crawling of search and hashtag queries, presumably to be able to rank for current trending events, which leaves open the possibility that links in those pages will have at least some authority passed on to them. The page’s ephemeral nature and high content churn rate will definitely count against you though and success is far from guaranteed.
LinkedIn is a much ‘cleaner’ platform in terms of general spam compared to Facebook and Twitter, making the potential for getting a followable backlink theoretically higher. They follow the same methodology as Twitter but go even harder with their robots.txt, only allowing crawling of their settings and help pages. Unfortunately, no luck here either.
Now it has to be said that this doesn’t make large social media platforms useless in the quest for backlink growth. They’re an amazing way to reach your audience and amplify your content. Your readers may even go on to write about or link to your content in places you didn’t even know existed. So don’t boycott them, but be wary that you almost certainly won’t be earning any direct authority.
Reddit is a more direct form of user-submitted content than the above three, and functions as hundreds of thousands of smaller communities. Recently Reddit underwent a drastic redesign that generated some kickback and caused Reddit to be hosted as two concurrent versions, Old and New. Oddly, the two versions (despite having identical content) handle outbound links and nofollow tags differently.
A little investigation of Old Reddit confirms it. All link posts start with a nofollow tag, and if a self-post contains an outbound link then the link and the self-post are also given a nofollow tag. After checking a few subreddits the mysterious algorithm seems quite simple; if a post has more than 10 upvotes then the nofollow tag is removed. Seems easy enough to get a backlink then? Unfortunately not. After a check of the robots.txt it looks like Reddit has stopped literally everything from crawling the Old Reddit subdomain.
On to New Reddit then. It’s robots.txt lets crawlers access the homepage and subreddits which is much more promising. After diving into specific links the nofollow tag is nowhere to be seen. Even the newest posts with zero engagement don’t show the tag. Either Reddit is happily passing on small amounts of authority to every link posted with no regard for content, or something is happening behind the scenes to prevent authority attribution. Without knowing what’s behind the curtain, it would seem like it’s technically feasible to get a followable backlink from New Reddit. The complete lack of nofollow tag does raise a red flag however, and I wouldn’t recommend investing significant resource on the site.
Hacker News is as simple as user-submitted content gets. Submit a link, people comment and vote on it. Easy. Its robots.txt allows essentially everything to be fair game for crawlers, and a quick scan of the day’s front page shows that they’re definitely followable links. A deeper inspection reveals that links on the New page aren’t followable though. So when do they change?
In order to establish when the change occurs, I kept browsing through the new submissions until I found a link that didn’t have the nofollow tag. The first I found had 6 comments and 43 points. Then another, with 11 comments and 65 points. I then checked the front page – they were the 1st and 2nd posts. After going 10 pages deep into New, I finally found a link that was followable, and not currently on the front page. It had 2 comments and 20 points.
It looked as though the front page doesn’t seem to be directly important in changing a link, it’s only an indicator of an engagement algorithm that converts nofollow to dofollow.
What’s the minimum level of engagement required for the change? After trawling through popular posts the most engaged nofollow link had 2 comments and 10 points. But there was another link. It had zero comments, 7 points, and was followable. Engagement isn’t the only factor for the tag change.
What else could it be? I started to look at the users who submitted the posts, the followable post’s user had 256 account karma (earned by interacting with Hacker News content) and the nofollow post’s user had 4. So reputation also matters?
Another post: zero comments, 7 points, 22692 karma, nofollow. There appears to be a lot more going on behind the scenes than can be easily gleaned and at a guess I would say that it’s a blackbox of points, comments, personal karma, karma of those engaging, and possibly even how quickly a post is engaged with.
Don’t despair though, the vast majority of links that achieved more than 7 points became followable. Combined with the large amount of traffic that Hacker News can generate, it’s a fantastic place to post your content. Quality content, that the Hacker News audience finds genuinely interesting, will almost certainly do well and you’ll be rewarded with a backlink.
What’s the trend?
There appears to be a general trend to the analysed sites: the larger and more varied the submitted content, the more necessary it is to prevent link spam. If a site grows beyond the ability to be directly moderated then it’s near impossible to not use a mixture of nofollow tags and strict robots.txts to limit the spam. The chance of a backlink from these sites is close to none. Hacker News is a more niche site, has a strong ability to moderate content posted there, and with it the potential for backlinks is higher.
Since niche sites with strong communities and moderation look like the way to go, we’ve already begun an upcoming post that dives deeper into some of the more obscure sites from across the web and the chance of generating backlinks from them.