There are a lot of potential pitfalls when it comes to using the nofollow attribute, I’ve audited many sites and helped unravel this tag for a significant amount of sites. Diagnostics of the nofollow tag can be problematic, some SEO tools can flag a nofollow URL but because they often adhere to the tag, they can provide little information on if it should be nofollow or not.
I often see people leveraging the nofollow tag to try and mask an infrastructure issue with the website (for example, functional but non-SEO friendly orphaned pages or handling excessive pages). The best way to handle such instances is to eliminate the bloating at the website level; ensuring that the pages are handled correctly and are not excessive.
I’d also like to remind people that there are effectively three ways to handle nofollow, this includes the X-Robots tag, meta robots tag and also the robots.txt file, if you are having any issues with the text file, I do have a dedicated robots.txt SEO guide.
How do I diagnose nofollow problems on my site?
I’d run a crawl on the site, using the crawler of your choice, be that SEMRush, Screaming Frog, SiteBulb etc… but I’d run it twice, one adhering to the robots.txt file and one without for comparative measure. Sometimes the crawler can be too light when probing sites if you have it set to follow the robots.txt file. I’d also consider checking in on the robots.txt file to see which sections of the site (by agent) you are removing from crawling, Google has a really robust robots.txt file testing tool too which can validate key areas of the site you may want to allow search engines to crawl.
What is the best use of nofollow on my site?
There is a website-specific approach to handling nofollow but I do see some consistent and useful applications for it. This includes acting as a deterrent for spam, whereby users who try and inject links into a site for SEO purposes will see very little benefit (only referral). Another use I’ve seen is secure pages, so you may want to try and prevent search engines from indexing checkout-sensitive pages but again, if a search engine can access anything sensitive on the site so can a user. Don’t deploy nofollow as a defensive measure; fix the root issue but by all means, use this as a fallback.
If you have a staging website a good deployment of this tag can be used here, alongside the noindex attribute. Production environment sites can often cause a bit of a conflict for SEO and also if a competitor has Google Alerts setup on your brand terms, they can often see the staging website as it becomes indexed. I’m on the fence with nofollow external links, sites that you don’t want to pass any credit to but are publicly linking from (with the exception of spam deterrent), you may as well pass some credit over. Unless the site is an example of something particularly bad e.g. if you have a website that showcases websites with a history of scamming people, this could be a good use of nofollow.
How to prevent incorrect use of nofollow?
Additional notes on nofollow
- Unless you know what you are doing then I’d always recommend caution when using this tag. Seek the advice of a SEO expert like myself to build confidence around the decision to use this tag.
- Unless you are deterring spam, the nofollow tag shouldn’t really be applied to your site. If you need it to do something above excluding credit to third party sites it could be a symptom of an incorrectly structured website.
- Use it as a last option, I also find using it at robots.txt level much easier to manage especially if people hard code this onto the site – as you can forget where it is at.
- Don’t forget to toggle your SEO crawling tool to adhere to and not-adhere to the robots.txt file / nofollow to see a comparative crawl – which can often highlight issues with nofollow.