Skip to main content

Research and Studies

Are Links Still a Powerful Ranking Factor in 2016 ? Resizeimage 2019 06 04t091452.776

Are Links Still Important for SEO?

UPDATE: For the latest version of this study go to Do Links Still Matter for SEO.
You’ve seen the claims:

  1. “So Many New Ranking Factors!”
  2. “Content Quality is Everything!”
  3. “User Engagement is the New King!!”
  4. “Social Media Signals Rock!”
  5. “RankBrain has Taken Over All Ranking!”

Puhleeze, spare me the nonsense! In today’s post, I’m going to share new data on links as a ranking factor. This report will demonstrate links still matter in rankings. I plan to share data that shows that all that talk of a decline in the importance of links as a ranking factor is grossly exaggerated.
In fact, links remain amazingly powerful. Let’s dig in!

Recent Googler Comments on Top Ranking Factors

On Mar 23, 2016, I participated in a Hangout with Google Dublin’s Andrey Lippatsev (along with Rand Fishkin and Ammon Johns). Here is an important snippet of the conversation:

Ammon Johns: We heard that RankBrain is the third-most-important signal contributing to results now. Would it be beneficial to us to know what the first two are?
Andrey Lippatsev: Yes. Absolutely. I can tell you what they are. It’s content and links going into your site.

Do tell now?

What the Moz and Searchmetrics Studies Showed

Both Moz and Searchmetrics have run groundbreaking studies on ranking factors, and each includes a look at links. You can see them here:

  1. Moz Correlation Factors
  2. Searchmetrics Ranking Factors

You can see some of the basic data points in the following chart:
Moz and Searchmetrics on Links as a Ranking Factor
Each of the bars above shows the level of correlation between that factor and higher rankings. In both the Moz and Searchmetrics studies, correlation to the number of links was relatively high, but not significantly higher than other factors they examined. If links are so important that a Googler would call them one of the two most important ranking factors, why aren’t these correlations showing higher numbers?
The key to the Moz and Searchmetrics studies is understanding how the evaluations were done. They did the evaluation of each SERP on an individual basis, and then took the mean of all the results (I’ll call this the “Mean of the Individual Correlations” approach). Both these studies also focused solely on commercial search terms.
So while their approaches have validity, because of our experience at Perficient Digital with the power of links in ranking, we decided to take a deeper look into the power of links. We took some new approaches to expose more about their impact. Bottom line? I believe that the Mean of Individual Correlations approach is not providing a complete picture.

Our Study Results

Based on consultations with a couple of experts (Paul Berger and Per Enge), I did a different type of calculation, based on the Quadratic Mean. The reason for doing this is that it leverages the square of the correlation variables (where the correlation value is R, the quadratic mean uses R squared).
It’s actually the R squared value that has some meaning in statistics. For example, if R is 0.8, then R squared is 0.64, and you can say about that 64% of the variability in Y is explained by X. As Paul Berger explained it to me, there is no meaningful sentence involving the correlation variable R, but R squared gives you something meaningful to say about the correlated relationship.
Here is a visual on what this calculation process looks like:
Quadratic Mean Calculation of Links as a Ranking Factor
In addition to the different calculation approach, I also used a mix of different query types. We tested commercial head terms, commercial long tail terms, and also informational queries. In fact, 2/3 of our queries were informational in nature. This may be one reason why we showed somewhat different results:
Correlation Between Links and Ranking
Our total links correlation result was higher than the correlation scores of DA and PA as ranking factors. I asked Rand Fishkin of Moz for a comment on why that may be, and here is what he had to say:

We use a different, broader corpus of keywords to generate PA/DA algorithms, and thus, it makes sense that on different types of keyword queries, they’ll have different levels of correlation. Very interesting to note that raw link counts tend to do better on these particular corpuses. My hope is actually that, one day in the future, Moz will be able to show how various metrics of all kinds correlate to any given set of keywords (wouldn’t it be awesome if you could see the correlations for the thousand keywords you’re rank tracking, for example?!).

Also of note is how high our total link score correlation was in comparison to Moz and Searchmetrics. The score I showed above used a different methodology, but even when we use the exact same methodology they did (the Mean of Individual Correlations approach), we get higher results. You can see the head to head comparison of all our link score calculations using that approach here:
Comparing Total Links Score for all 3 Studies
Taking a Moment to Give Thanks: all of our data was pulled from the Moz API, and the good folks on the Moz team were quite supportive of us in all aspects of the entire study.

Aggregate Evaluation of Links as a Ranking Factor

I think that the Mean of the Individual Correlations and Quadratic Mean approaches are both highly valid ones, but one of the limits with them is that a small number of results with a highly negative correlation can drag down the total score in a significant way.
For that reason, I chose to take some other approaches to the analysis as well. The first of those was to measure the links in a more aggregated manner. To do this, we normalized the quantity of links. Then we took the total of all the search results by ranking position. The equations for this look more like this:
Using Total Links by Ranking Position to Calculate Correlation
The value of this is that it smooths out the impact of the negative correlations in a different way. When you look at the correlations in this manner, here is what you get:
Aggregated Link Total Correlation
I also took another look at this. In this view, I continued to use the normalized totals of the links, but grouped them in ranking groups of 10. I.e., I summed the normalized link totals for the top 10, did the same for ranking positions 11 to 20, 21 to 30, and so forth. I then calculated the correlations to see how they looked in terms of what it would take to rank in each 10 position block. Those calculations looked more like this:
Calculate Correlation by Blocks of 10 in the SERPs
This gives us a bit more granular approach than simply aggregating all the ranking positions into the SERP positions, but still smooths out some of the limitations of the Mean of Individual Correlations method. Here is what we saw for results here:
Correlations Based on Groups of Ten in the SERPs
I’ve now gone from numbers in the 0.39 range to near-perfect correlations, so what do these numbers tell us? I believe that the aggregated approaches to the calculations tell us that links are far more important than the “mean” based calculations reveal.

Digging Into What’s Going On Here

To help with this analysis, we did some manual analysis of a few hundred results to figure out what percentage or results were likely not as influenced by links. What types of results are these? Here you go:

  1. Local results (not maps results, but results that are locally influenced
  2. Query deserves diversity
  3. In-Depth Articles

Our analysis suggested that around 6% of the results came from these types of results. So that’s one factor right there, albeit not one that explains the difference between the “mean” based calculations and the aggregate calculations.
However, the more important issue is understanding the role of content quality. Since Google has called content and links the two most important ranking factors, you can imagine a greatly simplified equation based on simply multiplying a “link score” times the “content score”.
Now in this completely hypothetical discussion, I would argue that the content score probably counts for more than the link score. After all, if the content isn’t relevant, it shouldn’t rank. That just makes sense. In addition, the level of relevance of a piece of content is highly variable.
Here is a major oversimplification of what the impact of this might be:
Simple Ranking Factors Concept
In the above chart, it’s actually scenario 2 that would rank higher, even though its “Link Score” is significantly lower. Now imagine that the link score runs only from 1 to 100, and the content related scoring factors (relevance and quality) have an exponential decay (were minor variations in quality and relevance have a large impact on the “Content Score”), links will clearly not be able to overcome lower relevance or weaker content.
[Tweet “If your content is not relevant or competitive, links won’t help ranking. If it is, links will make the difference.”]
Important Note: the above discussion was meant to illustrate the basic point about the significance of content scoring and its impact on ranking algorithms. It’s all my own speculation based on the analysis of the data I’ve seen, and the case studies that I’ll share in the next section).

Learn More! Why great earned links and quality content may be the only ranking factors that matter.

Cementing the Point With Case Studies

We do a lot of high-end content marketing campaigns with our clients, many of which are Fortune 500 companies. Here is a sampling of the results across many of our clients:
Sample Content Marketing Results
The sample results shown here have been repeated hundreds of times by us. However, we don’t find that links can rescue poor quality content, or cause low relevance content to rank. Also, all of our efforts focus on getting recognition from, or content published on, very high authority sites. Our data isn’t based on high volume, low-quality link building.


The Google algorithm continues to evolve, and we see many things that are impacting overall organic search traffic. Some of the biggest causes are:

  1. More real estate allocated to paid search
  2. More content from other sources, such as image search, YouTube, and the other factors I mentioned above
  3. Some pages that have less than 10 web results
  4. Portions of the web results that are clearly less driven by links, such as local web, query deserves diversity and in-depth article results

As a result, there are fewer than ten results on the first page of the SERPs which are driven by factors other than links. That doesn’t mean that links aren’t involved at all in rankings for those pages just that they matter less.
But, our study data strongly suggests that links continue to play a major role in rankings. In addition, our case study data makes the whole thing a slam dunk. When you aren’t facing page relevance or quality issues, links can and do, continue to significantly impact rankings.
[Tweet “Study shows minus relevance or quality issues, links still are a very powerful ranking factor. More at” ]
Note: even though our data focused on showing the correlation based on the quantity of links, that doesn’t mean that the quality of the links doesn’t matter, as it truly does. AND (see the emphasis here?), don’t even think about buying links.
So what’s the real upshot of all this? Simple:

  1. DO build great content and user experiences. In today’s digital marketing world, that’s just table stakes.
  2. DO proactively market your business and do the types of things that cause people to write about you and link to you.

Here’s a video summary of the study results:

Or watch Mark Traphagen present this data at the Engage Portland conference:

Thoughts on “Are Links Still a Powerful Ranking Factor in 2016 ?”

  1. What an incredible study. The biggest thing I spotted here was the fact bad content doesn’t rank by links. The crazy thing is that what your study shows is what ethical SEO’s have already been doing for years now.

  2. I think links are still working but if i got confused when i saw a website which is having to many niche and ranking better than the competitor. So i think links are pretty cool working

  3. I often quote that statement from Andrey Lippatsev above regarding content and links. It’s kind of a “duh” answer but if you really focus on those two items you’ll do fine.
    Creating good content is a bit time consuming and requires some energy and effort. Something most people do not want to do. However, I’ve found that even with great content, cutting through the mass amounts of drivel on the internet has become increasingly difficult.
    Articles, tweets, posts and updates have almost gotten to the point of junk mail. Great content is being created but you need to wade through a serious amount of nonsense to find it. Email subscriptions (like the one to Perficient Digital) and select RSS feeds have been my main source for finding good content. And…if it’s REALLY good, I’ll link to it.

  4. Vivek, as Eric says in the concluding portions of this study, there are factors that can make a page rank other than links, but in the vast majority of cases links will be the decisive factor.

  5. When the study says number of links pointing towards a site are they referring to follow + no follow or just followed? You should make a supplementary study using this data to see if you can see any nofollow/follow patterns – I for one think nofollowed links carry value; maybe not authority or juice but relevancy and so on.

  6. Thanks for sharing this data, Eric.
    We’re in another cycle of chicken littles proclaiming the death of SEO and this sort of study proves there is, and will always be, value in understanding the nuts and bolts of ranking.

  7. Superb study, & great insight to add to Moz and SeachMetrics’ studies.
    The cumulative aspect of link metrics correlating to ranking is fascinating. However, the idea of a link score plus content score giving rank makes a great deal of sense. Especially as many of the updates in recent years (penalties & filters) would make sense as factors affecting those scores (such as Panda filter discounting some pages for the content score, or adjusting a portion of the overall score somehow).
    Even more than this, theories of Rank Brain being a learning element, adjusting other ranking factors based on machine learnt experience of what will be a good user experience, works with this. Subtly adjust the link x content ratio based on experience of industry, such as for turning down link power for traditionally spammy niches, would help.
    Would you say user experience elements, such as bounce back %s, would affect the content score, or be a separate calculation?

  8. Hi Charlie – It’s not clear how Google uses engagement in ranking, but my guess is that straight bounce rate % isn’t it. However, to broaden your question, let me talk about all user engagement and how Google might treat that.
    To me, I think that’s part of how they measure content quality, so I think of it as part of the content score part of the equation.

  9. We have some clients for whom content is a challenge and in these cases we’ve had great results focusing on building quality links. Yes content is cool and sexy and link building is often dismissed as old hat so it’s great to read your findings. As always you dig deep to get the facts! Appreciated.

  10. The word decisive is very important, and well used to describe the value of links in the larger picture. In today’s digital world, SEO professionals can’t be a one trick pony with just links, or with just great content. You need both to prevail over your competition, and if your competition is doing the same thing, well you just need to do it better.

  11. I am a true believer that nofollow and even pure text links can carry some value, but as you have mentioned more along the line of relevancy and not really in terms of authority. So that is why I always try to look for both nofollow and follow opportunities. I wonder would Perficient Digital ever consider a study that weighs in on nofollow influence?

  12. Here’s a thing, Eric. When new signals are added to the algorithm, that doesn’t mean that they get more attention, or that something else gets less attention, or even removed from the mix. So why, when these studies come around do they always give credit to one factor as the “dominant” signal. With machine learning (or pattern matching or any other description you’d like to use for it) the ranking mechanism becomes fluid. It’s more of a “genetic” process. I believe 100% as Google states, that as they move from pure “retrieval” algorithms to “learners” ( a step towards AI) that end user data can be folded in (and out) at a rate that wasn’t possible before. Basic library science says that, more popular books get taken out more than less popular. And less popular get put in the back store room… Until someone requests them again.
    Usage data is not just based on quantity of clicks when you look at temporal analysis, for instance. Look at my recent column on the past, present and future of SEO. A hugely important point I make (and have done for some long period of time) is how “undemocratic” the web becomes if only people who create web pages and can link to and from them, get to decide what we see on the web via search.
    As I point out, it’s like saying the people who make TVs get to decide what we watch on them. Google’s mission is to create the best end user experience. And the best way to do that, is be aware of what the end user herself deems to be the best content experience. See you in Las Vegas for more of the same 🙂

  13. Great data and insights, Eric.
    With respect, I don’t think many SEOs or marketers would have found these results surprising. Links were the fundamental of Google’s PageRank system and have continued to play a crucial part in their ranking algorithm for the past decade.
    I think an interesting study would be to measure the impact of authoritative nofollow links on fortune 500 company websites. This just might be a big eye opener for most SEOs and marketers.

  14. Very (very very) nice study ! So What about a study on the quality of links next ? 🙂 Thanks for your work anyway !

  15. I tend to disagree with the conclusion that the correlation shows the impact of links on ranking as this study does not take into account the effect of the ranking itself on the number of external links (so the other way around).
    So it has no relevance from statistical point of view as you CANNOT establish by default the relation between the cause (name it ranking OR external links) and its effect (name it ranking or external linking).
    To be more specific (and I have dozens of examples) those ranking high/top positions because of the exposure is normal to have a high number of external links as more and more people in time do refer to them as trustful sources.
    Actually if you take a look from statistical point of view at the other situation where pages without that many external links rank high it would tell that I’m right with the above that links cannot be interpreted this way.
    On the other hand the single thing links can help from SEO point of view is in terms of stability/volatility in time into a top ranking position. For instance what I suggest is that those ranking high (depending on the query of course, as some situations woun’t require a top ranking webpage to get links in time .e.g: some basic info pages) if not gain enough citations in time, they may suffer (hard to say how much in average) as that may be a strong signal for Google that the webpages they rank high are not trustful enough, but again that is a temporary and presumably effect I think as some of them still stay there over time.
    Hope that helps,

  16. Hi Laura – we do plan to do more on this study with a follow-up later this year. We’ll look at adding more on the link quality side of thing then.

  17. Hi Florin – you saw that we included some case study data, right? That should backup the correlation data I would think.

  18. Great 🙂 The definition itslef of what is a quality backlink has always be a discussion with SEO friends… I’d love to have your point of view, I’ll keep reading 🙂

  19. Well, truth be told that one is not relevant as getting a specific number of links for a top ranking webpage is a function dependent on several variables (e.g: the “age” of the webpage ranking and “for how long” that webpage stayed into a top ranking position to get a specific number of external links from people referring to it as trustful source, “the relevancy for the queries it was top ranking” which determine how people can behave in relation to the webpage in question, “the query type” – some of them are suitable for getting external links, while some don’t, the “quality of the content itself and how many sources were used for that which in the end determine external linking from similar sites/those being linked to” etc).
    So how did you assessed these so that to be able to argue your conclusion is supported by data?
    None of these weren’t taken into account in all of this (or I may be missed them), so how can you afterwards say you’ve taken this into account?
    Thus I tend to say the conclusion risen from the study is not strongly argued with data. Or to be politically correct, your conclusion may be a good one but is purely a hypothesis as you did not proved it.
    Waiting for your reply,

  20. That’s a lot of maths! But I can’t find nowhere in your article the word “causation”.
    How can you tell these are “ranking factors” when you do correlation studies?
    I think it’s a very good idea to use another type of keywords corpus.

  21. Hello Eric,
    I am happy to reading this blog becuse some of the people told that, now backlinks are not more helpful to rank the keywords in Google. That’s the article which i need to see them. I happy to read this blog. Get me more inspirationto build backlinks.
    Thank you

  22. Olivier, there’s a good reason why you won’t find the word “causation” in our study: correlation never proves a particular causation. At a base level, correlation simply means that two things occur together. For example, we can see that there is a high correlation between ice cream sales and drowning deaths. Does that mean either one causes the other? Of course not! But they do tend to rise and fall together. In this case, we are reasonably certain of the real cause: seasonal changes in the temperature.
    So what’s important when looking at correlation is how likely is the studied factor to be a cause of the effect. This is typically expressed as a ratio, as shown throughout the article. So you’re always dealing in a percentage of likelihood, never certainty. However, in this case, we can have a higher degree of certainty that there is at least some causal relationship, as Google has told us (repeatedly) that links are a major part of their ranking algorithm. So here we do not have to make much of a leap. The question remaining is: given that links are indeed a ranking factor, how strongly are they associated with results that rank highly? That’s what this study set out to show.

  23. Vasek, Eric Enge is traveling today, so let me share with you an example he shared with our Perficient Digital team yesterday.
    Say you enter into Google the query “jaguar.” That word could refer to an automobile brand, an animal, a type of guitar, and perhaps a number of other things. So since Google can’t be sure which you mean, it will display some results from each of the most popular types of things that represent the query. And that means that there are inevitably results ranking high that are not there because of traditional ranking signals (such as links), but because they might represent what the searcher is actually looking for.
    Now in reality, in many cases Google can take a reasonable guess at which thing you’re looking for. Such things as your personal search history, or even the histories of people whose search profiles are similar to yours, might tip the balance of showing more of one type of possible answer for your own search. But the point stands: query deserves diversity is one example of where things other than link profile are having the major effect of what’s on page one.

  24. Hi Mark
    You’re literally confirming what I’ve said in the above, while Eric was arguing that the study is backed up by appropriate study case data. But after going through your last comment I can say you just confirmed my position on this article. And that is:
    By this study that uses correlation as indicator you can’t answer the question that you promote within the article title which I quote: “Are external links a strong ranking factor.” (by the way look at the tag of this article that sais “Links as Ranking Factor: Study Shows They Still Rule” – which if taken seriously tells us that according to this “new study” they still rule.
    But again within your last comment you say something totally different. By looking at correlation only you can’t show if they rule…honestly. Needless to say there’s no word about how these links are counted so that to have a more accurate output about their influence.
    So the study is useless as it only tries to be “catchy” by simply bringing in the spotlight the same statement Google made that links are an important factor (which is partially true), but what about the value added by the “new” study itself?
    Thank you,

  25. Above all else, one of the hardest things to accomplish as an SEO professional is link building. Typically when someone goes to a bar they don’t want a diluted drink, otherwise it will require more of them to accomplish what they set out for.
    In other words, why perform more searches, or time in search, researching for your answer?
    Google understands the importance of your buzz and wants to make sure your feeling it on the first shot. Now we have the bar…it’s just a matter of becoming the supplier to the people who are looking for the buzz.
    Figured I would throw a little screw driver analogy into and already cloudy topic.

  26. Michael, quality linkbuilding is indeed difficult. That’s why here at Perficient Digital we’ve invested years of effort into building a highly effective content marketing program. We use our highly-tested “Success By Design” methodology to establish the best quality, most relevant, and “reachable” publisher targets and then pitch them with high quality content they are proud to have on their sites. Takes a lot of effort, but our client results speak for the value!

  27. Links had to work, because they are the reference for good site and good content. I think links and content complement each other. Happy to read this post.

  28. Nice job! I think you missed the largest ranking correlation study related to links based on number of SERP results, 1 million. I oversaw this for Brian Dean and Backinko. We chose to focus JUPerficient Digitalon page one GSERP. Our data was from ahrefs which many people believe has a deeper db of links than Moz or Majestic. Our keyword data was from SEMrush (who leant their stats employee to us). We also used a very different, exhaustive, method for normalizing a random set of 100,000 keywords v. prior studies. The write-up of the study was the most up-voted on and in Q1 & Q2..
    Keep the studies coming guys! Your contributions are fantastico!

  29. I think this is a great study, but what Google does with PageRank, and there is no secret, is a lot more complex.
    You are trying to correlate the sheer number of backlinks with SERP position. In the PageRank model Google was building a graph and count links in a diffrent way, by counting the links to links, to say so.
    Meaning that if a page has 3 IBLs, and one of the IBLs comes from a page with 1mil IBLs that will outrank a page with 5 IBLs that have no IBLs themselves.
    I think that now, when you have the data, you should try, at least for a small partition of the data, to build the graph and run the calculations for each node.

  30. Hi guys – this is totally awesome. In my previous company we built and patented a tool based on Spearman/Pearson correlations.
    For those who have not spent a lot of time looking at that type of correlation data, I thought I would chime in and let them know that a 0.3 is pretty darn strong, and a 0.4 is much stronger than a 0.3. So some of those bigger correlation numbers are incredibly strong. (Just thought I would give it a little explanation/perspective.)
    This is just a really nice piece of research using real research tools. Nice job!

  31. Brilliant, I recently had a discussion with a whole load of people who we’re arguing that links were no longer a ranking factor.
    Despite showing them my own results across hundreds of sites they argued their side to the death.
    I laughed too and now I’m gonna show them this….

  32. Hi Florin – It’s understood that correlation does not equal causation, if that’s your main point, then I agree with that. However, that doesn’t mean that correlation studies are of no value. One just must understand the limitations of what they tell us, but we didn’t limit our study data to that. I’m also still unclear on why you think the case study data isn’t useful here.
    We took many pages which we promoted via content marketing campaigns, and they prospered, greatly.
    In parallel, on the same sites, we saw many pages of a comparable nature which we did not promote via content marketing campaigns, and they didn’t prosper (granted, BTW, that I didn’t make this latter point in the write up, but it’s true).

  33. Hi Eric,
    Keeping in mind the limitations the correlation has (actually its relevancy in this matter, not the fact that correlation as a measure is not useful because it is indeed) what I’m saying is that particularly in this case it is even less useful (specifically measuring the correlation between top ranking pages and the no of backlinks so that to see how often they occur together) because of the following reasons (and these are just basic variables which aren’t taken into account in the above analysis, and by that I mean taken into account by quantitative metrics which can then be measured/stay objective).
    Take the below list as a really shortlist, so the list is open as neither me or you know exactly all the stuff Google takes into account:
    a) For how long the webpage in question stayed on top ranking position so that to have an enough high exposure so that in the end to get a specific no. of external links in time (I repeat the importance of “for how long”). I should emphasize that I strongly believe that each high ranking webpage Google ranks actually has a score related to backlinks that is computed by an algorithm which takes “for how long” into account and to illustrate that I can give the following examples:
    1st situation: have a new webpage that once launched online within less than 30 days gathered a significant no. of backlinks – that can receive a specific backlinks score I’m speaking about (name it “Score A”);
    2nd situation: have a 5 year old webpage ranking high that gathered another amount of backlinks within the given time frame (name it “Score B”). So within the correlation study you make in order to be relevant as a measure you should try and make the study by having a clear split between younger / older webpages ranking high.
    b) The relevancy for the queries the webpage in question was top ranking: needless to say that the broader the terms the page is ranking for the less relevant it may be for some user needs thus, in the end it can result in being less referenced to.
    c) The query type – some of them are suitable for getting external links, while some don’t – I assume I don’t have to give an example for that because you know better than me that that’s the case (e.g: one is to have inside your sample deep analysis/study webpages ranking high which are debated into specific circles thus have a high no. Of backlinks, versus another webpage with some basic info where users don’t have what to actually debate about, thus less backlinsk). But I can bet Google takes that into account because otherwise things would look ugly in the top 10 results (e.g: I can’t judge someone for not dealing with a specific task as I know for sure the fact that person is not supposed to deal with it according to its role within an organisation).
    d) the relevancy of the webpages, niches, queries you took into account (which is also related to point c as some areas are suitable for getting high no. of backlinks, while some other don’t.
    So, in a study like you did once you’ll be able to take into account at least few of the above points and assess their influence in the no. of external links a webpage gathered (respectively try and alter the absolute values of the no of backlinks accordingly) you can in the end use correlation to say something like this: look for x and y correlation is positive, thus top ranking pages do have a high number of backlinks, respectively webpages with high no. of backlinks have a top ranking position, which can then lead to the question: now what is the causation between the two, respectively are backlinks still rulling or not?).
    Since we both agreed that correlation can’t show the causation and that prior to use correlation you first need to alter the figures related to the no. of backlinks (which is a hard job to do, but that is how goes with doing research and produce such studies), you can see why I consider that in this case correlation is even less relevant and can’t be used to argue what we know and try to prove after Google’s statement that they’re an important factor. (we’re on the same boat until a specific point on this, but in all honesty this study is not the best way to prove it like you initially pretended).
    So from correlation to causation, you first need to ensure that the 2 variables you take into account are clear from the influence of other variables, then use them to say they correlate positively, then follow up with the discussion about causation, then use a tag for your article like “Links as Ranking Factor: Study Shows They Still Rule” as you did. That title for this blog article is misleading SEOs and the community around it (especially business owners) because you’re not proving it simply because you can’t do it by correlation.
    Let me know what you think of the above,

  34. This is incredible!
    Although some of the calculations were a little confusing but this is really an eye-opener.
    Just shared it on Twitter. 🙂

  35. I am new here ! Really interesting and blog ! Technical SEO i love it ! I believe in trusted links and good content for Users with a SEO optimization !

  36. Florin – There is a reason why I included case study data in the post. It deals with the problem of confounding variables (that any correlation study in this arena will face). The case studies basically operated like this:
    1. We started with a page ranking in position 7 (using the first one in the table shown)
    2. We got it 11 backlinks through content marketing campaigns
    3. No changes were made to page content during the process
    4. It moved to position 1
    The process has been repeated many times, with many different keywords.
    That data isn’t from a correlation study, so we’ve not relied solely on the correlation data. I provided:
    1. Google’s comments on the top 2 ranking factors
    2. The correlation data
    3. Some examples of real world campaigns we’ve done for our clients

  37. Mike – so sorry that it’s taken me this long to get back to you. I added a link to your excellent post in your comment above.
    I completely agree with the key point at the end of your post, which is that “Perhaps now really is the time to focus on optimizing for humans and not for machines.”
    Certainly, by one means or another, that’s where we’re headed.

  38. Very well done!
    Googlers have been reiterating that content quality is THE most important factor. Well, while quality is absolute, relevance is relative to the search term and yet unless they both match, one cannot rank high.
    Any study on quality correlation?

  39. I struggle to follow the maths (I sat at the back of the class for that in school). But I think what you are saying is that more of the right links are better. Logic suggests that the more people with authority talk about you, the more authoritative you look as a result.
    There is also another factor – link recency. I have seen sites climb when they have a flurry of new link activity, only to see them slide again when the recency of their links drops off.

  40. It is amazing how many seos only focus on the on page and technical aspects. To them links are not even a thought. Quality is important but what metric is used algorithmically to assess it?

  41. Omg, thank you. This research post is the most impressive and needed one ever, Reading it seems like drinking a glass of fresh, cold water ofter a hike in Death Valley in 120 degree heat without a canteen. No, it’s more like stumbling into a god dang oasis. I only wish I had the verbal skills to give it the praise it deserves. I think the post is momentous and timely since backlinks, even quality ones, have gotten a bad rap in last couple of years. (This praise is not hyberbole. ) There is a myth going around that the importance of backlinks is dwindling. Even though my own professional experience contradicts this, and I find myself wanting to argue with many SEO conference speakers who pronounce this during a workshop. I only had anecdotal proof that it wasn’t so.
    Also most of the highly-rated top contributers on Google Search Console’s forum on ranking are very anti-backlink. They state quite emphatically that almost every backlink looks unnatural and you should get rid of most if not all of them. They told one of my clients to remove an extremely valuable link on a high trust flow site which was natural link from a quality site. It seems to me they’re giving really bad advice and must hurt the people’s sites who follow it. Now there is some research to refute them.
    Anyway, I digress.
    I read in this post somewhere that one site went up from ranking sixth on the first page to ranking first after getting only 3 quality links. I read it in the middle of the night and now can’t find it again. Did I dream it? Lol. No I also heard about it in the podcast from SEO 101.
    What three links were those? Lol. Wikipedia? .edu sites? What?
    And don’t get me wrong I don’t think for one second that backlinks are the only or even most important ranking signal. The first step must always be an awesome site with compelling content. Throw in some quality backlinks and you’re cooking.

  42. Hi Cathie – I’m glad you found the study to be valuable. It reflects our experience as well (obviously ;->). As for the 3 links, keep in mind that every situation is different. You might rank in position 6, and need very little to catapult you into position 1, depending on the “gaps” between your ranking signals and those in front of you. Or, you might be in position 2, and need LOTS of links to get into position 1, if the gap between you and your competitor is large.
    No doubt, it’s rare that you need only 3 links (even when they are great ones) to go from position 6 to 1, but it does happen.

  43. If there is a contextual relevancy between the linking page, linked page and the anchor text, and if the source site is an authoritative site then that link is bound to do good for the lniked-site. The important thing is quality of the link is more important than the numbers of links. Anchor text diversification is another important thing which we must practise. Thanks for this post it strengthens that ethical or organic links are still as important as those were used to be.

  44. This is awesome, so sick of people saying links aren’t important.
    Of course “links” can be loosely defined. Obviously on your chart where they are only getting a handful of links you are getting high quality editorial type style from authors of credibility and probably not link exchanges, and blog comments 🙂 or could be industry link roundups etc type links etc.
    I know links will always be a factor because that’s that hardest form of quality signals of trust and authority to manipulate. Well not impossible but a lot harder than every other metric out there.

  45. For SEO purposes, photos give added content and also interest to a website,
    achieve natural position on their own on a SERP, and also are particularly searched for on online search engine
    – every one of which are web traffic electrical generators.

  46. Amazing study!
    There are many factors that helps a page to rank better in search engines and quality links is one of them.
    Thanks for this detailed study.

  47. I had several discussions about the relevance of links to the Google algorithms and I think Google was much too successful with their backlink-relevance-strategy then being now able to just switch away from that strategy. I guess that is still a topic for the future.
    Nice article and study.

  48. Hey eric.what nice content this was.It really amezing.You researched a lot when you written this.I know that linking is always a big factor for ranking.But now i also thing that not only linking we have to more care on social media services and the content also.Then we can get the high ranking easily.thanks for giving this much data

  49. Backlinks is still the top most factor for ranking website in Search Engines. With over 64% of Google Local ranking factors qualifying as backlinks. Thanks for sharing nice blog with us, it is really helpful for SEO webmasters. Keep it sharing with new tips and tricks.

  50. If anything, I think links are even a more powerful ranking factor than they were when this article was published – Google seemed to back away a bit from ‘authority’ and ‘trust’ and appears (at least to me) to be relying on link signals more than ever.

  51. Hello Eric
    Thanks for sharing this data
    Great post done! happy to know links are still helpful for ranking with good content as well
    I am new here ! Really interesting and blog ! Technical SEO i love it ! I believe in trusted links and good content for Users with a SEO optimization.

  52. In summation, links matter! You must also keep in mind what you are ranking for? Page relevance or quality issues may just be a bigger weight for SERPs than a link for the particular page you are optimizing. Once again, great study.

  53. I LOVE that you mention.. Link Score * Content Score. I think it’s so undervalued to correlation that these two things play together. They MUPerficient Digitalplay together to achieve ranking in the high/more competitive keywords. So often times we see SEO’s/Agencies saying “Links are the answer!” or conversely “Content is King!” when in fact they both play an equal role and one can offset the other depending on the quality factor. Like Mie Grehen said… There is no reason to assume one factor is MORE dominant than the other especially now that we move to AI.

  54. Eric
    I’m curious what your feelings are about what Google deems as bad back links. Is it wise to disavow back links originate in directories?

  55. I would still disavow poor quality backlinks. At this point, that would include pretty much any web directory, with possible VERY rare exceptions. Note: when referring to web directories, I do NOT include in that business listing directories such as,, Yelp, and other sites of that type.
    The directories I’m referring to are sites that simply are directories of web sites across the webs.

  56. I’m thinking that Google’s ever evolving algorithms are slowly (or quickly) forcing our hand as SEO’s towards some use of PPC. Google’s AI appears to be favouring the searchers first click. Therefore if a user clicks on a paid ad (for the first time), then after this, my tests are showing that this site is displaying higher in organic listings.
    As you say, the real estates for paid ads has increased and Google is forcing our hand more and more… what next?!

  57. Yes, of course backlink always be the main factor for good SERP.
    Google will always change the algorithms. This March, Google introduced new update named Google Fred.

  58. Continuing to give links a prominent role in rankings now and in the future makes a lot of sense to me, and I can’t see why this would change. Links as a ranking signal have worked so well for Google for a long time, and now that spammy links aren’t effective anymore, relevant authoritative links are hard to get without some serious effort so should be valued as a trusted metric.

  59. Eric,
    very nice study. Question regarding “We got it 11 backlinks through content marketing campaigns”
    Were all the new 11 links going to the home page or some (or all) to inner pages?
    un-natural link building is too many links to the homepage. I’d think some were to inner pages.

  60. I agree totally with this. I experienced it when I created my website newly and didn’t pay attention to link building but as time passed I discovered my problem and solved it. Today I can boost of appearing first in search results for few keywords.
    Thanks and I learnt new thing

  61. I actually find it interesting how many SEO agencies these days just rely on getting content shared across social media platforms as their link building campaign and ignore other tired and true ways of cultivating a more robust link profile.

  62. Hi Eric Enge,
    It is nice study really. I saw Links are still valuable for better ranking and High Page Authority is working well better than low Page Authority. Got something special from you.
    Thanks 🙂

  63. Definitely link building is an effective factor of ranking and you describe all the necessary points of this topic.
    Thanks for sharing this article.

  64. Mark Traphagen

    no-follow links pass no ranking authority, so they aren’t even part of the equation.

  65. Very good and thank you for the data you give. But to get a good link like what ? Do we have to visit a site that already has a high ranking page ?

  66. Mark Traphagen

    Great question, Bawang! You should always try for a link from the most relevant, highest authority page you can get, but especially when starting out, don’t set your standards so high that you can’t get a link anywhere. Any link is better than no link, as long as it is from a good quality site that doesn’t appear spammy and doesn’t engage in black hat schemes (such as soliciting or accepting paid links).

  67. Halo sir.., I want ask.
    How to try links from sites that have the highest quality? I mean, if we want to see a good quality site, how can we find it ??

  68. Mark Traphagen

    I would start by searching for sites that are relevant to your site’s main topics. Then a good first qualifier is using a domain authority metric such as Moz’s DA score, or Majestic’s Trust and Citation Ranks, or Ahrefs’ Domain Rating. Obviously, the higher any of those numbers are, the better, but you will need to work your way up to being able to place content with links on higher authority sites. Start by pitching first to sites with decent (but not super high) authority that don’t appear spammy.

  69. Sir your article arose a question in my mind.
    Are no follow links of no value ?
    Then how can I manage a do follow link from high authority sites

  70. NoFollow links do not have SEO value, but can have a marketing and reputation value.
    Getting links from high authority sites is a big topic that I can’t easily cover here. But, the heart of it is to produce content ofextremely high value and do the right things to get noticed from those sites. That’s a very simplified explanation, and there is a LOT more to it!

  71. WOW!!! What an analysis, thanks for the detailed study. I have few query. Suppose we get a good backlink for “x” page of our website from an authority site can we take link from same site for “y” page of our site, will that be same value or different? Another query is it wise to work for home page to increase its DA or work for internal pages for ranking?

  72. If you get more than one link from the same site, it does continue to add more value. Of course, if you have 1,000 links from a site, and you get the 1,001st, that probably does not add that much value. But, getting multiple links from one site is a very good thing to do, particularly if it’s an authoritative web site.
    Keep in mind, in any given market, there are a small number of sites that are the most important. Getting strong levels of endorsement from those sites, and more than one of them, is incredibly valuable.
    As to your second question, it depends on where you are. If your business is completely unknown, building general authority (and credibility) makes sense. As you get a bit more established, you may want to begin to focus on specific areas of the business. But, beware, there are many different factors that determine the best answer for your particular business.

  73. mind blowing postmortem on RANKING Factor. Really appreciated. Thanks for sharing. Remaining doubts got cleared in comment section. Keep up the good work. 🙂

  74. very informative and nice article on how to write a blog post. You shared lots of things on how to select the topic through Buzzsumo, best practices to be followed while writing blog titles to draw the attention and the importance of Images. Thanks for sharing awesome stuff

  75. This post is really an eye-opener. I have a local news website that ranks highly without many links. I always wondered why Google loves this site so much to give it so many high rankings, and my other sites have to struggle. Now I understand. Thanks for the great info.

  76. Thanks for sharing this with us. Just one question why buying links is not a good option?
    If they are providing you a good quality backlinks then what is the problem?

  77. Google doesn’t like them because purchased links are about your budget, not whether or not you have the best content. They rely on editorially given (not purchased) links to help them find the most authoritative content. In addition, Google has gotten increasingly better at detecting paid links, and punishing sites that buy them. It’s best that you don’t do it.

  78. Finally a structured link study with a pragmatic approach.
    It would be interesting to consider the nofollow links to see if they have any impact as ranking factor.

  79. i still don’t understand about social signal, because i dont have any significant power from that backlink. Can u help me to explain this… sorry for my bad english

  80. Not surprising with the end result, but always good to know we as SEOs need to be focusing on the basics: content and links, and focusing on quality for both.

  81. Thanks Careen. We’ve found that being generous with the knowledge we’ve gained is not only a great way to support and foster the industry we love, but helps us build business as well.

  82. Linking building is the most important factor in SEO to increase domain authority on any website. Thank you for sharing with us more linking information, its’ really very useful for us.

  83. Well the blog is great and it seems that you have analysed a lot of keywords. But
    I have some questions & would love to get an answer from you.
    Can you name a few keywords that you have tracked?
    Which type of link you consider as a quality backlink, like blog commenting, article submission, guest posting, etc etc?

  84. We used a mix of informational and commercial queries. These covered many markets, including health, information, and technology, but also more pure informational queries.
    As for the type of backlinks, stay away from blog commenting and article submission as a way to get links. With guest posting, you also need to be very careful to not implement a campaign just for SEO purposes. Guest posting can be a part of a larger scale marketing program, including PR, social media, and other forms of promotion. If you think about yourself as implementing a branding campaign online, and you get a lot of links as a result of that, you are more likely headed in the right direction.

  85. Hi Eric, thanks for the reply.
    Actually after reading your blog I’m kinda figuring it out which strategy will suit my website i.e. As it lands in the category of Educational websites. We tend to help students with their assignments and homework.
    Can you suggest me, which path to follow?

  86. I don’t think there is such a thing as a right balance. If you market yourself properly on the web, you’ll end up with a certain number of NoFollow links, for sure, but I don’t think you need to hit a perfect ratio.
    That said, NoFollow links probably still add some value. Not in the PageRank sense, but in the general buzz about your brand on the web sense, especially when they appear in contexts that are not likely to be compensated.

  87. Yes, Links are the most powerful option for ranking factor! whether it is a follow link or no follow link. It will improve your overall domain authority.

  88. Quite informative. I still think quality content will attract links. But the degree of your visibility will depend on your link-building efforts. What’s your experience? What approaches work best?

  89. Link building is very effective way to rank your website but you should also have look after other factor such as content , internal linking, social signals etc.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Eric Enge

Eric Enge is part of the Digital Marketing practice at Perficient. He designs studies and produces industry-related research to help prove, debunk, or evolve assumptions about digital marketing practices and their value. Eric is a writer, blogger, researcher, teacher, and keynote speaker and panelist at major industry conferences. Partnering with several other experts, Eric served as the lead author of The Art of SEO.

More from this Author

Follow Us