How the Google Search Algorithm Works \ Blog \ BH&P \ Creative Venture Marketing Agency in Reading

How the Google Search Algorithm Works

 

Last week, BH&P attended the first ever HubSpot Agency Content Bootcamp. We were privileged to hear from HubSpot’s Principal Technical SEO, Victor Pan, who gave us a whistle stop tour of the Google Search Algorithm, and what we as content marketers need to know.

Now, I’m happy to confess that much of what Victor had to say was way too technical for me – but there was lots that DID make sense.

Here’s my interpretation of what Victor had to say (thank you Victor for reviewing this too!)

Why the Google Search algorithm is important

The Google Search algorithm is one of those great mysteries.

If only we knew the secret to the code, we’d be like Keanu Reeves seeing the Matrix, right? With this superior knowledge, we could make sure that our sites, videos and content rank at the top of the first page on Google and YouTube, every time.

Or would we?

If only, if only, Google would spill the beans.

Facts, Leaks and “off the record” chats

There are three forms of information we get about the Google Search algorithm.

They look a bit like this:

Google Algorithm - facts, leaks and interviews

Facts

Here are the bare facts: what Google does to your site, every day:

  1. Google crawls your website

  2. Google renders your website

  3. Google indexes your website

  4. Google ranks your website

  5. Google uses machine learning

You can read more about how Google does this on their handy “How search works” guide

But the playing field has changed. The number and type of searches that people are making, and the content available to be searched, is immeasurably different from what it was ten years ago.

SEO is Dying, Right - Victor Pan

So is SEO dead? Far from it. The number of new queries does not equate to the total number of queries.

More and more people are searching online every day. When there is more and more content floating around up there in the cloud, then SEO can only become more important.

 

Pandas and penguins

Panda and penguin are spam filters within the Google algorithm introduced in 2011 and  2012 respectively, each designed to prevent sites with poor quality from working their way into Google's top search results.

Penguin, for example, introduced in April 2012, was designed to better catch sites deemed to be spamming its search results, in particular those doing so by buying links or obtaining them through link networks designed primarily to boost Google rankings.

Google are on a mission to ensure that the very best content ranks above the mediocre.

The trouble, of course, is that every time Google updates the algorithm, site masters and SEO agencies race to find ways to rank higher than the competition.

For the most part (see the caveats in the leaks and speculations below), Google has continued to evolve their algorithm, penalising those who use trickery in their attempt to rank, and to move towards an approach that allows the very best content to rise to the top of the search rankings.

Here are some of the key facts:

  • The information that Google share about their algorithm is typically very top line – it is designed mainly to reassure users that Google is working to ensure that high quality content ranks above what marketers want you to see

  • Google made 3234 changes in 2018 (8.9 changes every day), many based on machine learning (AI) so no one person can ever know absolutely everything about the algorithm

  • If you are confident in your content strategy, but are not ranking, there are many tools you can us to review search volumes, how competitive the search term(s) are, and to review your strategy

  • If you want to get ranked faster, you can “ping” your site in Google Search Console (i.e. ask Google to recrawl your URLs)

(Source: Google search: rigorous content)

The latest news

In September 2019, Google officially announced a change to the rules around rel=nofollow.

HubSpot published a blog about this recently, in which Victor Pan explains the change and why it is important.

To find out more, read the article on hubspot.com

Leaks

In August 2019, Google whistleblower Zachary Vorhies leaked information that showed how Google allegedly manipulates search result – “twiddling”.

The “Twiddler” system – assuming that it is a real thing - lets Google/YouTube workers rank search results based upon what Google want emphasised or de-emphasised.

Allegedly, twiddler makes use of a list of keywords called the “Controversial Query Blacklist.” This list of flagged keywords will cause a video or site to be reviewed as to whether it should be promoted to the first page of search results, demoted, or possibly removed entirely.

The Twiddler Quick Start Guide explains to Google employees how to rank results such as how to “demote a result to the second page” and so on, or even hide them altogether.

And it is speculated that a similar algorithm will move sites up or down the Google rankings regardless of the quality of the content, or of site optimisation.

As most people using Google or Youtube statistically go to the first few results that pop up when they search for a topic, this means that some sites effectively have a “glass ceiling” above which they will never rank.

Despite being controversial, the main problem with twiddles is that not only are they entirely speculative and highly technical, but they are also top secret, and virtually impossible to compensate for in your content or SEO strategy.

So is there anything you can do about twiddles?

Our suggestion is that unless you are an SEO genius, your best bet (for now at least) is to leave them alone. If they are real (and Victor Pan suspects they are), then they may not be fair, but they are designed to lower the ranking of controversial and politically biased websites.

Since typically, these are not topics that most digital marketers (B2B or B2C) are focused on, twiddles are probably not something we need to worry about unnecessarily.

However, if you do want to find out more, here’s a handy link!

 

Interviews

Victor Pan shared the outputs from a number of recent interviews, with Googlers whom he believes to be reliable sources.

The insights he shared are listed below – and both are useful to search and content marketers.

Interview 1, May 2019

When a client or agency works on a site or content, they typically focus on the things they thing are most important to help the content “rank” through a bidding system.

Yet sometimes you can work hard on a factor that you think you are being penalised for (such as moving from HTTP:// to HTTPS://, for example), but not improve your ranking.

Marketers should not be disheartened by this.

You need to continue!

The reason for this is that the Google algorithm is designed to work like a score sheet, with a multiplier which leads to a bid score.

So to rank high, you need to optimise for all the important factors, not just one.

It is better to do everything quite well, rather than one thing brilliantly (e.g. page speed, content dates, use of image tag, etc.).

Interview 2, with Gary Illyes, September 2019

When Google crawls the web, it has to break down the words it sees into numbers its algorithm can understand.

So it does not read sentences.

Instead, the text is converted into “tokens” that Google uses to infer user intent.

Because of the way Google sees the content, content marketers need to remain mindful of the developments and optimisations.

All the things that are recommended by your technical SEO experts are worth doing because the bids stack as multipliers. Stay on track of all the things that are recommended versus what still needs to be done. Continue to stay on track of how Google manages user intent, see how that evolves, and adapt to that change with tangible action items in your team’s day-to-day workflow.

Great enterprise SEO is invisible to the untrained eye because collaboration happens between designers, writers, front-end developers, and business stakeholders; we’re aligned to solve business problems, at the core of which are real human needs and not our egos.”

Victor Pan

 

Follow @victorpan on twitter

5 Hints and tips for SEO in 2020

  1. Video adds to your Google “score” (especially YouTube).

  2. Remember the fundamentals are important for video too, as they are for any other content - the title, the description, backlinks to the video itself, the transcript, the text on the web page above and below the video, anything that stops people from dropping off mid-view.

  3. A good thing to look at when exploring what subtopics to consider in relation to a campaign is Google’s “People also asked” listing. But where does this come from? Are they real questions? They are the amalgamation of thousands of real searches and real questions – questions people have really asked and that you should try to provide an answer in your content.

  4. Google does not provide data on audio search – so if this is something you are featuring, you may need to double down on featured snippets. This is something that may change in the coming months and years, as marketers continue to create content in audio format.

  5. Rename stock images so the alt text does not match the name on the image library and is relevant to your content and user intent. Ideally, make modifications to stock images (colour filters, layover text, photoshop) so that it becomes unique.


Final word on content and SEO

Remarkable content starts with empathy.

The content we create needs to be approached from a standpoint of how we can solve our users’ problems, while also considering when and where they will search for and find the answer to their problem.

The reality is that when you create original, high quality content, focused on topics that people are searching for, and supported by links from high ranking sites, you are likely to improve your search visibility – but this is by no means guaranteed.