Let me begin by stating that I do not put an enormous emphasis on SEO when I’m training a link builder. Generally speaking, my team of link builders knows the basics of SEO, but they’ve been taught that they can’t rely on metrics alone in order to judge whether a link is going to be good or bad for our clients. My background is in general and technical SEO, but I realized early on that for the work we usually do at my agency, most of what I knew didn’t really apply — at least, not in a very significant way.
Of course, if you’re doing high-level analysis of any sort, you do need to have a great deal of SEO knowledge. The reason I don’t train anyone to do this (on an advanced level) is that I don’t want them to ever lose their ability to think like humans. Though they must run all link-building activities by me for approval, I still trust my link builders to do good work on their own — and they don’t disappoint.
Approaching link building as a human, not a bot
When you approach link building as a human being — not putting the sole emphasis on “Will Google like this?” — you are able to see websites as users see them. Unless they’re SEOs, users aren’t going to be bogged down with thinking about Domain Authority or whether the robots.txt file is properly set up.
Take Google-indexed pages, for example. I think it’s a terrible signal if a page is not indexed in Google. However, I can also make the case that if this imaginary page is the most popular page on a site that ranks well, and it gets a lot of traffic, a link there is probably still going to be good for traffic to your site.
Maybe that particular page got dropped from the index because it’s a big fat list with tons of links, even though it’s still a good page. A human user would click on a link there, but the search engines don’t like it. Would you rather have a link from that non-indexed but popular page, or would you rather have a link from a page that is indexed but has a fraction of the traffic?
Clients fighting back
We do a lot of talking about how getting links that will actually get clicked on is a good idea, but we’re still getting pushback from clients who focus solely on numbers. This link is on a page with a low Moz Page Authority. This one looks great everywhere, but the Majestic Trust Flow is low.
I love tools like Moz and Majestic. They are incredibly useful, and there is no way I would be able to conduct any link analysis without them. I just don’t think that any single metric can paint a full picture of whether or not a link is going to be useful to your site.
Just recently, I’ve come across a lot of pages with very high Page Authority (PA) and great Domain Authority (DA) that are not indexed in Google. I’ve come across some great pages via social channels that Moz hasn’t even picked up yet, so they have no DA or PA. I’ve seen pages ranking well that have a Moz Spam Score of 4 out of 17.
Many link builders would reject these pages as potential linking partners for those exact reasons. What would you do? What human user is going to know or care about the Moz Spam Score?
Is someone going to click?
When SEOs think, “Is someone going to click on this link?,” oftentimes they are simply not thinking as a human user. They are still thinking about other factors and other signals.
- Are there competitor links in the post? Some users love to see an alternative.
- Are there too many links? Who defines “too many?”
- Are there some misspellings? Does one image seem broken although the others are fine? We’re all human, and we all make mistakes. Why can’t we let the webmasters make a few mistakes?
Again, for the issues listed above: Do you think a human user really cares?
The granularity of what we can dig into is mind-blowing at times.
The importance of referrals, not rankings
As a personal example, I have gotten clients from interviews that I’ve done on fairly new websites that, at the time, had very low DA (or, even farther back, a Google PageRank of 1). I’ve written articles or been mentioned on sites with amazingly good authority and metrics and not received any requests for information. I’m sure they’ve helped me rank, but in my case, a great deal of my business comes from referring URLs — not search engines.
I rank well for certain keywords and get business that way, but the main driver of clients for me is a referral, whether it’s from an article I write, a fellow SEO or a social mention.
In my own writing, if I give a link, it’s because it’s useful and relevant. When I build links, I strive to obtain links that are useful and relevant. If a link is useful and relevant, you’d imagine that it would be clicked on and good for traffic, correct?
Just do a few Google searches and examine the metrics of the top-ranked pages. You’ll find many with very poor metrics. They rank well, though! Since most people outside of SEO do view the search results as being a true representation of which pages are best, you should be better able to get some traffic from one of those pages than from one buried on page 11, even if it does have a DA of 50.
Social is still huge
Then we have the issue of social platforms, which are critical to online success for many businesses these days. You don’t have to rank well in search to do well on Facebook, for example. If your article is on a site with low metrics, and it gets retweeted on Twitter by 1,000 people, you should be happy. In terms of local search, Facebook is a massive player that might eventually compete with Google.
I’m definitely not saying to ignore all metrics, to pretend SEO isn’t important, and/or to just slap a link anywhere you like because surely someone will find it and click it. I’m simply pointing out that when you analyze things to death, you leave a lot of potentially wonderful opportunities on the table.
Contributing authors are invited to create content for Search Engine Land and are chosen for their expertise and contribution to the search community. Our contributors work under the oversight of the editorial staff and contributions are checked for quality and relevance to our readers. The opinions they express are their own.
