Friday, August 11, 2017

Google (Almost Certainly) Has an Organic Quality Score (Or Something a Lot Like It) that SEOs Need to Optimize For - Whiteboard Friday

Posted by randfish

Entertain the idea, for a moment, that Google assigned a quality score to organic search results. Say it was based off of click data and engagement metrics, and that it would function in a similar way to the Google AdWords quality score. How exactly might such a score work, what would it be based off of, and how could you optimize for it?

While there's no hard proof it exists, the organic quality score is a concept that's been pondered by many SEOs over the years. In today's Whiteboard Friday, Rand examines this theory inside and out, then offers some advice on how one might boost such a score.

Google's Organic Quality Score

Click on the whiteboard image above to open a high-resolution version in a new tab!

Video Transcription

Howdy, Moz fans, and welcome to another edition of Whiteboard Friday. This week we're going to chat about organic quality score.

So this is a concept. This is not a real thing that we know Google definitely has. But there's this concept that SEOs have been feeling for a long time, that similar to what Google has in their AdWords program with a paid quality score, where a page has a certain score assigned to it, that on the organic side Google almost definitely has something similar. I'll give you an example of how that might work.

So, for example, if on my site.com I have these three — this is a very simplistic website — but I have these three subfolders: Products, Blog, and About. I might have a page in my products, 14axq.html, and it has certain metrics that Google associates with it through activity that they've seen from browser data, from clickstream data, from search data, and from visit data from the searches and bounces back to the search results, and all these kinds of things, all the engagement and click data that we've been talking about a lot this year on Whiteboard Friday.

So they may have these metrics, pogo stick rate and bounce rate and a deep click rate (the rate with which someone clicks to the site and then goes further in from that page), the time that they spend on the site on average, the direct navigations that people make to it each month through their browsers, the search impressions and search clicks, perhaps a bunch of other statistics, like whether people search directly for this URL, whether they perform branded searches. What rate do unique devices in one area versus another area do this with? Is there a bias based on geography or device type or personalization or all these kinds of things?

But regardless of that, you get this idea that Google has this sort of sense of how the page performs in their search results. That might be very different across different pages and obviously very different across different sites. So maybe this blog post over here on /blog is doing much, much better in all these metrics and has a much higher quality score as a result.

Current SEO theories about organic quality scoring:

Now, when we talk to SEOs, and I spend a lot of time talking to my fellow SEOs about theories around this, a few things emerge. I think most folks are generally of the opinion that if there is something like an organic quality score...

1. It is probably based on this type of data — queries, clicks, engagements, visit data of some kind.

We don't doubt for a minute that Google has much more sophistication than the super-simplified stuff that I'm showing you here. I think Google publicly denies a lot of single types of metric like, "No, we don't use time on site. Time on site could be very variable, and sometimes low time on site is actually a good thing." Fine. But there's something in there, right? They use some more sophisticated format of that.

2. We also are pretty sure that this is applying on three different levels:

This is an observation from experimentation as well as from Google statements which is...

  • Domain-wide, so that would be across one domain, if there are many pages with high quality scores, Google might view that domain differently from a domain with a variety of quality scores on it or one with generally low ones.
  • Same thing for a subdomain. So it could be that a subdomain is looked at differently than the main domain, or that two different subdomains may be viewed differently. If content appears to have high quality scores on this one, but not on this one, Google might generally not pass all the ranking signals or give the same weight to the quality scores over here or to the subdomain over here.
  • Same thing is true with subfolders, although to a lesser extent. In fact, this is kind of in descending order. So you can generally surmise that Google will pass these more across subfolders than they will across subdomains and more across subdomains than across root domains.

3. A higher density of good scores to bad ones can mean a bunch of good things:

  • More rankings in visibility even without other signals. So even if a page is sort of lacking in these other quality signals, if it is in this blog section, this blog section tends to have high quality scores for all the pages, Google might give that page an opportunity to rank well that it wouldn't ordinarily for a page with those ranking signals in another subfolder or on another subdomain or on another website entirely.
  • Some sort of what we might call "benefit of the doubt"-type of boost, even for new pages. So a new page is produced. It doesn't yet have any quality signals associated with it, but it does particularly well.

    As an example, within a few minutes of this Whiteboard Friday being published on Moz's website, which is usually late Thursday night or very early Friday morning, at least Pacific time, I will bet that you can search for "Google organic quality score" or even just "organic quality score" in Google's engine, and this Whiteboard Friday will perform very well. One of the reasons that probably is, is because many other Whiteboard Friday videos, which are in this same subfolder, Google has seen them perform very well in the search results. They have whatever you want to call it — great metrics, a high organic quality score — and because of that, this Whiteboard Friday that you're watching right now, the URL that you see in the bar up above is almost definitely going to be ranking well, possibly in that number one position, even though it's brand new. It hasn't yet earned the quality signals, but Google assumes, it gives it the benefit of the doubt because of where it is.
  • We surmise that there's also more value that gets passed from links, both internal and external, from pages with high quality scores. That is right now a guess, but something we hope to validate more, because we've seen some signs and some testing that that's the case.

3 ways to boost your organic quality score

If this is true — and it's up to you whether you want to believe that it is or not — even if you don't believe it, you've almost certainly seen signs that something like it's going on. I would urge you to do these three things to boost your organic quality score or whatever you believe is causing these same elements.

1. You could add more high-performing pages. So if you know that pages perform well and you know what those look like versus ones that perform poorly, you can make more good ones.

2. You can improve the quality score of existing pages. So if this one is kind of low, you're seeing that these engagement and use metrics, the SERP click-through rate metrics, the bounce rate metrics from organic search visits, all of these don't look so good in comparison to your other stuff, you can boost it, improve the content, improve the navigation, improve the usability and the user experience of the page, the load time, the visuals, whatever you've got there to hold searchers' attention longer, to keep them engaged, and to make sure that you're solving their problem. When you do that, you will get higher quality scores.

3. Remove low-performing pages through a variety of means. You could take a low-performing page and you might say, "Hey, I'm going to redirect that to this other page, which does a better job answering the query anyway." Or, "Hey, I'm going to 404 that page. I don't need it anymore. In fact, no one needs it anymore." Or, "I'm going to no index it. Some people may need it, maybe the ones who are visitors to my website, who need it for some particular direct navigation purpose or internal purpose. But Google doesn't need to see it. Searchers don't need it. I'm going to use the no index, either in the meta robots tag or in the robots.txt file."

One thing that's really interesting to note is we've seen a bunch of case studies, especially since MozCon, when Britney Muller, Moz's Head of SEO, shared the fact that she had done some great testing around removing tens of thousands of low-quality, really low-quality performing pages from Moz's own website and seen our rankings and our traffic for the remainder of our content go up quite significantly, even controlling for seasonality and other things.

That was pretty exciting. When we shared that, we got a bunch of other people from the audience and on Twitter saying, "I did the same thing. When I removed low-performing pages, the rest of my site performed better," which really strongly suggests that there's something like a system in this fashion that works in this way.

So I'd urge you to go look at your metrics, go find pages that are not performing well, see what you can do about improving them or removing them, see what you can do about adding new ones that are high organic quality score, and let me know your thoughts on this in the comments.

We'll look forward to seeing you again next week for another edition of Whiteboard Friday. Take care.

Video transcription by Speechpad.com


Sign up for The Moz Top 10, a semimonthly mailer updating you on the top ten hottest pieces of SEO news, tips, and rad links uncovered by the Moz team. Think of it as your exclusive digest of stuff you don't have time to hunt down but want to read!

No comments:

Post a Comment