Is Google’s Ranking Algorithm Really That Complicated?

dbk03

Regular Member
Joined
Feb 21, 2012
Messages
360
Reaction score
361
This article says that, if you zoom out far enough, the Google’s ranking algorithm boils down to two main factors: relevance and authority.
Of course, it mentions that there are other factors, too, like: content, links, site optimization, but these are more for improving your authority and they're not primarily focused on ranking (although higher authority should mean higher rankings in my opinion...)
In the end, they do admit that Google Algorithms are quite complex, though.

What do you think?
 
relevance = content
links = authority
In a nutshell... I kinda agree... all you need is relevance and authority and you'll outrank better articles.
 
They are absolute requirements, but then there are others which are also very important - load time, bounce rate etc. relevence and authority will get you noticed, the others will help you stay in view
 
not its not, but it takes constant tweeking and a lot of work, therefore it take investment and addoption google has the adoption making it hard to beat.
 
you can say those are the main ranking factors, yet they entail alot of other factors
 
maybe it's not really complicated, but it definitely has a lot of filters, exceptions and rules that we don't know about... Otherwise there's no logical explanation for different keywords and sites within the same niche gets treated differently, why some new sites get sandboxed while other new sites rank from week 2 (both sites in the same niche I mean) even though they're almost carbon copies of each other... Too many rules and filters to be just a simple algo... at least in the tougher niches....
 
It's not complicated at all.

Pushing out consistent content is a ranking factor. It determines how involved your content is in it's niche.

Links are how the content is received. You've also got other deterministic factors like ads and social media like facebook, twitter, Youtube, etc.

It's just a process. Ranking requires work, effort, and some patience.
 
maybe it's not really complicated, but it definitely has a lot of filters, exceptions and rules that we don't know about... Otherwise there's no logical explanation for different keywords and sites within the same niche gets treated differently, why some new sites get sandboxed while other new sites rank from week 2 (both sites in the same niche I mean) even though they're almost carbon copies of each other... Too many rules and filters to be just a simple algo... at least in the tougher niches....
i agree with this,

even carbon copies but unique enough not to be considered 100% duplicates get treated differently

2 carbon copies can launch at same time and one can rank for keywords the other cant meanhile the latter also ranks for stuff the former cant.

i think sandbox and duration is also very niche specific.

in the pirate field i notice its about 12 months while other niches may be released right off the bat.
 
maybe it's not really complicated, but it definitely has a lot of filters, exceptions and rules that we don't know about... Otherwise there's no logical explanation for different keywords and sites within the same niche gets treated differently, why some new sites get sandboxed while other new sites rank from week 2 (both sites in the same niche I mean) even though they're almost carbon copies of each other... Too many rules and filters to be just a simple algo... at least in the tougher niches....
Plain magic. I can imagine that machine learning model having 2 almost identical samples but it doesn't like one of them and bam! No ranks!

Don't try to understand ML. It's a cheap replication of human brains google engineers use instead of hiring people. That's what happens when you employ neural networks at scale.


Ever thought what would happen if AI ran a business? That's it! That's it!


Google doesn't compare two websites that are alike. I don't research that but I think it's common sense to think this when you consider how neural networks work.

Give AI two identical samples. You will get different (but similar) results every time.

How would that work if google used an algorithm that would compare two sites and give them the same score? I would expect it to rank it the same which is impossible cause one website can take only one spot in SERPs. So this doesn't make sense and google does that differently.
 
Last edited:
"The Google codebase includes approximately one billion files and has a history of approximately 35 million commits spanning Google's entire 18-year existence."

People can't contribute to open source projects developed by a single person without spending months of studying their code. And you ask someone to understand Google? All we have is a generalization.

If something doesn't work, stop thinking why that domain is gone. Get a new one, replicate your content and move on. You won't win with that.

I would see SEO as a medium risk investment.
 
Last edited:
How would that work if google used an algorithm that would compare two sites and give them the same score? I would expect it to rank it the same which is impossible cause one website can take only one spot in SERPs. So this doesn't make sense and google does that differently.
exactly! I, too, keep saying to everyone that there's only 10 spot in google (even less if we consider the fact that only the first 3-5 positions get most of the traffic), but there are literally 1000s of sites wanting those 3 spots for the same keyword, and even if all 1000s of sites are carbon copies of each other and they all follow google's best practices, they simply can't rank all of them. So, google needs to differentiate, but how they differentiate between sites is something that we, as SEOs, have no real clue about. We can only speculate, although, after the last update we don't need to speculate anymore because apparently google - yet again - gave even more power and priority to big brands...
 
I really don't think the algorithm is that complicated at all. Providing high-quality, relevant content on a consistent basis to your audience - that is linked to - is 2/3'rds of the game in SEO.

If you run a legitimate business, that provides value, SEO is quite simple.
 
maybe it's not really complicated, but it definitely has a lot of filters, exceptions and rules that we don't know about... Otherwise there's no logical explanation for different keywords and sites within the same niche gets treated differently, why some new sites get sandboxed while other new sites rank from week 2 (both sites in the same niche I mean) even though they're almost carbon copies of each other... Too many rules and filters to be just a simple algo... at least in the tougher niches....

Exactly. The algorithm is complicated in terms of having many variables, but SEO itself is not complicated.

I hired several SEO consultants last year to help me with some SEO audits. Usually, I always do them myself but I wanted to see if they could speed up the process for me. What I found was that everyone I hired knew all the on-page SEO rules and technical SEO factors to look out for, but the great divider was when it came to choosing keywords, mapping those keywords to the site, and using those keywords in the content.

Firstly, they ALL chose something different.

Secondly, none of them chose what I viewed as the best options.

FYI, we were dealing with difficult niches, such as eVTOL and Telematics insurance.

It gave me some perspective on my approach to SEO and helped me realise how to find better SEOs to hire.
A successful SEO needs to not only know the fundamentals of SEO, but they need a flair for marketing copy (making people feel), a keen understanding of target audiences (fast to study and adapt), and the buyer's journey.
These are areas you rarely see discussed on forums as compared to "how to get links?" and "how much content to write?", and they certainly complicate SEO for people that aren't broadening their skillset.
 
There is a journey thread how someone ranked for high competition keywords in 8 mo or so with pbn.

The idea is to build night quality relevant group of websites linking to your money website. Each website should be a real one - good content, updated regularly with good links and social signals. This is the key.

I did something similar and it works, I can confirm and I see my competitors are still doing it. I just got bored with all that d*cks around. But yes this the way to rank for anything.
 
A successful SEO needs to not only know the fundamentals of SEO, but they need a flair for marketing copy (making people feel), a keen understanding of target audiences (fast to study and adapt), and the buyer's journey.
You know why there is a shortage of SEOs in Europe?
Cause being in Europe means you have to deal with people who take 4 week vacations and drop all responsibilites
Some of these people can work a $1k job and not complain.

You won't learn many skills that way. And SEO is where many skills will make you excel.

https://www.blackhatworld.com/seo/this-business-is-so-huge-my-mind-is-blowing.1412885/
 
I have a very simple advice for you -

Keep publishing good quality originally written well researched content rich with keywords and keep on building manual backlinks on a daily basis, and more importantly - don't stress yourself on thinking how search engine algorithm works. The algorithm is way too complex to be understood by any mortal like us and all is in our hand is to keep doing good Karma so that we will be in good books with Google God.

Your efforts will be rewarded only if you keep consistency in your work. If you stop in between, all your work will be gone.
 
I have a very simple advice for you -

Keep publishing good quality originally written well researched content rich with keywords and keep on building manual backlinks on a daily basis, and more importantly - don't stress yourself on thinking how search engine algorithm works. The algorithm is way too complex to be understood by any mortal like us and all is in our hand is to keep doing good Karma so that we will be in good books with Google God.

Your efforts will be rewarded only if you keep consistency in your work. If you stop in between, all your work will be gone.
What do you mean by keep on building manual backlinks on a daily basis? What kind of backlinks you are referring to?
 
exactly! I, too, keep saying to everyone that there's only 10 spot in google (even less if we consider the fact that only the first 3-5 positions get most of the traffic), but there are literally 1000s of sites wanting those 3 spots for the same keyword, and even if all 1000s of sites are carbon copies of each other and they all follow google's best practices, they simply can't rank all of them. So, google needs to differentiate, but how they differentiate between sites is something that we, as SEOs, have no real clue about. We can only speculate, although, after the last update we don't need to speculate anymore because apparently google - yet again - gave even more power and priority to big brands...
My theory would be that Google wants a range of different answers within the top 10 for any given search term, to give it the best chance of satisfying that search when it isn't quite sure of user intent. So it's not going to put 10 versions of the same answer in the top 10 in that case. It will maybe try to have some sites in there that provide some variety. That would be why two sites that are carbon copies can never rank both together. Instead, some other site will get in there which doesn't compete with the other results on authority, but does provide a variation on the naswer to the search query. That would be why sometimes you will look at the top 10 results and wonder how the hell that one site ranks in there when it's got like, 20 RDs and nowhere near as many articles as the others, or the articles are all less than 500 words, or whatever. You think it shouldn't rank, but it does because it's just different enough.
 
Back
Top