What Deep Learning and Machine Learning Mean For the Future of SEO – Whiteboard Friday

Posted by randfish

Imagine a world where even the high-up Google engineers don’t know what’s in the ranking algorithm. We may be moving in that direction. In today’s Whiteboard Friday, Rand explores and explains the concepts of deep learning and machine learning, drawing us a picture of how they could impact our work as SEOs.

For reference, here’s a still of this week’s whiteboard!

Video transcription

Howdy, Moz fans, and welcome to another edition of Whiteboard Friday. This week we are going to take a peek into Google’s future and look at what it could mean as Google advances their machine learning and deep learning capabilities. I know these sound like big, fancy, important words. They’re not actually that tough of topics to understand. In fact, they’re simplistic enough that even a lot of technology firms like Moz do some level of machine learning. We don’t do anything with deep learning and a lot of neural networks. We might be going that direction.

But I found an article that was published in January, absolutely fascinating and I think really worth reading, and I wanted to extract some of the contents here for Whiteboard Friday because I do think this is tactically and strategically important to understand for SEOs and really important for us to understand so that we can explain to our bosses, our teams, our clients how SEO works and will work in the future.

The article is called “Google Search Will Be Your Next Brain.” It’s by Steve Levy. It’s over on Medium. I do encourage you to read it. It’s a relatively lengthy read, but just a fascinating one if you’re interested in search. It starts with a profile of Geoff Hinton, who was a professor in Canada and worked on neural networks for a long time and then came over to Google and is now a distinguished engineer there. As the article says, a quote from the article: “He is versed in the black art of organizing several layers of artificial neurons so that the entire system, the system of neurons, could be trained or even train itself to divine coherence from random inputs.”

This sounds complex, but basically what we’re saying is we’re trying to get machines to come up with outcomes on their own rather than us having to tell them all the inputs to consider and how to process those incomes and the outcome to spit out. So this is essentially machine learning. Google has used this, for example, to figure out when you give it a bunch of photos and it can say, “Oh, this is a landscape photo. Oh, this is an outdoor photo. Oh, this is a photo of a person.” Have you ever had that creepy experience where you upload a photo to Facebook or to Google+ and they say, “Is this your friend so and so?” And you’re like, “God, that’s a terrible shot of my friend. You can barely see most of his face, and he’s wearing glasses which he usually never wears. How in the world could Google+ or Facebook figure out that this is this person?”

That’s what they use, these neural networks, these deep machine learning processes for. So I’ll give you a simple example. Here at MOZ, we do machine learning very simplistically for page authority and domain authority. We take all the inputs — numbers of links, number of linking root domains, every single metric that you could get from MOZ on the page level, on the sub-domain level, on the root-domain level, all these metrics — and then we combine them together and we say, “Hey machine, we want you to build us the algorithm that best correlates with how Google ranks pages, and here’s a bunch of pages that Google has ranked.” I think we use a base set of 10,000, and we do it about quarterly or every 6 months, feed that back into the system and the system pumps out the little algorithm that says, “Here you go. This will give you the best correlating metric with how Google ranks pages.” That’s how you get page authority domain authority.

Cool, really useful, helpful for us to say like, “Okay, this page is probably considered a little more important than this page by Google, and this one a lot more important.” Very cool. But it’s not a particularly advanced system. The more advanced system is to have these kinds of neural nets in layers. So you have a set of networks, and these neural networks, by the way, they’re designed to replicate nodes in the human brain, which is in my opinion a little creepy, but don’t worry. The article does talk about how there’s a board of scientists who make sure Terminator 2 doesn’t happen, or Terminator 1 for that matter. Apparently, no one’s stopping Terminator 4 from happening? That’s the new one that’s coming out.

So one layer of the neural net will identify features. Another layer of the neural net might classify the types of features that are coming in. Imagine this for search results. Search results are coming in, and Google’s looking at the features of all the websites and web pages, your websites and pages, to try and consider like, “What are the elements I could pull out from there?”

Well, there’s the link data about it, and there are things that happen on the page. There are user interactions and all sorts of stuff. Then we’re going to classify types of pages, types of searches, and then we’re going to extract the features or metrics that predict the desired result, that a user gets a search result they really like. We have an algorithm that can consistently produce those, and then neural networks are hopefully designed — that’s what Geoff Hinton has been working on — to train themselves to get better. So it’s not like with PA and DA, our data scientist Matt Peters and his team looking at it and going, “I bet we could make this better by doing this.”

This is standing back and the guys at Google just going, “All right machine, you learn.” They figure it out. It’s kind of creepy, right?

In the original system, you needed those people, these individuals here to feed the inputs, to say like, “This is what you can consider, system, and the features that we want you to extract from it.”

Then unsupervised learning, which is kind of this next step, the system figures it out. So this takes us to some interesting places. Imagine the Google algorithm, circa 2005. You had basically a bunch of things in here. Maybe you’d have anchor text, PageRank and you’d have some measure of authority on a domain level. Maybe there are people who are tossing new stuff in there like, “Hey algorithm, let’s consider the location of the searcher. Hey algorithm, let’s consider some user and usage data.” They’re tossing new things into the bucket that the algorithm might consider, and then they’re measuring it, seeing if it improves.

But you get to the algorithm today, and gosh there are going to be a lot of things in there that are driven by machine learning, if not deep learning yet. So there are derivatives of all of these metrics. There are conglomerations of them. There are extracted pieces like, “Hey, we only ant to look and measure anchor text on these types of results when we also see that the anchor text matches up to the search queries that have previously been performed by people who also search for this.” What does that even mean? But that’s what the algorithm is designed to do. The machine learning system figures out things that humans would never extract, metrics that we would never even create from the inputs that they can see.

Then, over time, the idea is that in the future even the inputs aren’t given by human beings. The machine is getting to figure this stuff out itself. That’s weird. That means that if you were to ask a Google engineer in a world where deep learning controls the ranking algorithm, if you were to ask the people who designed the ranking system, “Hey, does it matter if I get more links,” they might be like, “Well, maybe.” But they don’t know, because they don’t know what’s in this algorithm. Only the machine knows, and the machine can’t even really explain it. You could go take a snapshot and look at it, but (a) it’s constantly evolving, and (b) a lot of these metrics are going to be weird conglomerations and derivatives of a bunch of metrics mashed together and torn apart and considered only when certain criteria are fulfilled. Yikes.

So what does that mean for SEOs. Like what do we have to care about from all of these systems and this evolution and this move towards deep learning, which by the way that’s what Jeff Dean, who is, I think, a senior fellow over at Google, he’s the dude that everyone mocks for being the world’s smartest computer scientist over there, and Jeff Dean has basically said, “Hey, we want to put this into search. It’s not there yet, but we want to take these models, these things that Hinton has built, and we want to put them into search.” That for SEOs in the future is going to mean much less distinct universal ranking inputs, ranking factors. We won’t really have ranking factors in the way that we know them today. It won’t be like, “Well, they have more anchor text and so they rank higher.” That might be something we’d still look at and we’d say, “Hey, they have this anchor text. Maybe that’s correlated with what the machine is finding, the system is finding to be useful, and that’s still something I want to care about to a certain extent.”

But we’re going to have to consider those things a lot more seriously. We’re going to have to take another look at them and decide and determine whether the things that we thought were ranking factors still are when the neural network system takes over. It also is going to mean something that I think many, many SEOs have been predicting for a long time and have been working towards, which is more success for websites that satisfy searchers. If the output is successful searches, and that’ s what the system is looking for, and that’s what it’s trying to correlate all its metrics to, if you produce something that means more successful searches for Google searchers when they get to your site, and you ranking in the top means Google searchers are happier, well you know what? The algorithm will catch up to you. That’s kind of a nice thing. It does mean a lot less info from Google about how they rank results.

So today you might hear from someone at Google, “Well, page speed is a very small ranking factor.” In the future they might be, “Well, page speed is like all ranking factors, totally unknown to us.” Because the machine might say, “Well yeah, page speed as a distinct metric, one that a Google engineer could actually look at, looks very small.” But derivatives of things that are connected to page speed may be huge inputs. Maybe page speed is something, that across all of these, is very well connected with happier searchers and successful search results. Weird things that we never thought of before might be connected with them as the machine learning system tries to build all those correlations, and that means potentially many more inputs into the ranking algorithm, things that we would never consider today, things we might consider wholly illogical, like, “What servers do you run on?” Well, that seems ridiculous. Why would Google ever grade you on that?

If human beings are putting factors into the algorithm, they never would. But the neural network doesn’t care. It doesn’t care. It’s a honey badger. It doesn’t care what inputs it collects. It only cares about successful searches, and so if it turns out that Ubuntu is poorly correlated with successful search results, too bad.

This world is not here yet today, but certainly there are elements of it. Google has talked about how Panda and Penguin are based off of machine learning systems like this. I think, given what Geoff Hinton and Jeff Dean are working on at Google, it sounds like this will be making its way more seriously into search and therefore it’s something that we’re really going to have to consider as search marketers.

All right everyone, I hope you’ll join me 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!

Reblogged 4 years ago from tracking.feedpress.it

Technical Site Audit Checklist: 2015 Edition

Posted by GeoffKenyon

Back in 2011, I wrote a technical site audit checklist, and while it was thorough, there have been a lot of additions to what is encompassed in a site audit. I have gone through and updated that old checklist for 2015. Some of the biggest changes were the addition of sections for mobile, international, and site speed.

This checklist should help you put together a thorough site audit and determine what is holding back the organic performance of your site. At the end of your audit, don’t write a document that says what’s wrong with the website. Instead, create a document that says what needs to be done. Then explain why these actions need to be taken and why they are important. What I’ve found to really helpful is to provide a prioritized list along with your document of all the actions that you would like them to implement. This list can be handed off to a dev or content team to be implemented easily. These teams can refer to your more thorough document as needed.


Quick overview

Check indexed pages  
  • Do a site: search.
  • How many pages are returned? (This can be way off so don’t put too much stock in this).
  • Is the homepage showing up as the first result? 
  • If the homepage isn’t showing up as the first result, there could be issues, like a penalty or poor site architecture/internal linking, affecting the site. This may be less of a concern as Google’s John Mueller recently said that your homepage doesn’t need to be listed first.

Review the number of organic landing pages in Google Analytics

  • Does this match with the number of results in a site: search?
  • This is often the best view of how many pages are in a search engine’s index that search engines find valuable.

Search for the brand and branded terms

  • Is the homepage showing up at the top, or are correct pages showing up?
  • If the proper pages aren’t showing up as the first result, there could be issues, like a penalty, in play.
Check Google’s cache for key pages
  • Is the content showing up?
  • Are navigation links present?
  • Are there links that aren’t visible on the site?
PRO Tip:
Don’t forget to check the text-only version of the cached page. Here is a
bookmarklet to help you do that.

Do a mobile search for your brand and key landing pages

  • Does your listing have the “mobile friendly” label?
  • Are your landing pages mobile friendly?
  • If the answer is no to either of these, it may be costing you organic visits.

On-page optimization

Title tags are optimized
  • Title tags should be optimized and unique.
  • Your brand name should be included in your title tag to improve click-through rates.
  • Title tags are about 55-60 characters (512 pixels) to be fully displayed. You can test here or review title pixel widths in Screaming Frog.
Important pages have click-through rate optimized titles and meta descriptions
  • This will help improve your organic traffic independent of your rankings.
  • You can use SERP Turkey for this.

Check for pages missing page titles and meta descriptions
  
The on-page content includes the primary keyword phrase multiple times as well as variations and alternate keyword phrases
  
There is a significant amount of optimized, unique content on key pages
 
The primary keyword phrase is contained in the H1 tag
  

Images’ file names and alt text are optimized to include the primary keyword phrase associated with the page.
 
URLs are descriptive and optimized
  • While it is beneficial to include your keyword phrase in URLs, changing your URLs can negatively impact traffic when you do a 301. As such, I typically recommend optimizing URLs when the current ones are really bad or when you don’t have to change URLs with existing external links.
Clean URLs
  • No excessive parameters or session IDs.
  • URLs exposed to search engines should be static.
Short URLs
  • 115 characters or shorter – this character limit isn’t set in stone, but shorter URLs are better for usability.

Content

Homepage content is optimized
  • Does the homepage have at least one paragraph?
  • There has to be enough content on the page to give search engines an understanding of what a page is about. Based on my experience, I typically recommend at least 150 words.
Landing pages are optimized
  • Do these pages have at least a few paragraphs of content? Is it enough to give search engines an understanding of what the page is about?
  • Is it template text or is it completely unique?
Site contains real and substantial content
  • Is there real content on the site or is the “content” simply a list of links?
Proper keyword targeting
  • Does the intent behind the keyword match the intent of the landing page?
  • Are there pages targeting head terms, mid-tail, and long-tail keywords?
Keyword cannibalization
  • Do a site: search in Google for important keyword phrases.
  • Check for duplicate content/page titles using the Moz Pro Crawl Test.
Content to help users convert exists and is easily accessible to users
  • In addition to search engine driven content, there should be content to help educate users about the product or service.
Content formatting
  • Is the content formatted well and easy to read quickly?
  • Are H tags used?
  • Are images used?
  • Is the text broken down into easy to read paragraphs?
Good headlines on blog posts
  • Good headlines go a long way. Make sure the headlines are well written and draw users in.
Amount of content versus ads
  • Since the implementation of Panda, the amount of ad-space on a page has become important to evaluate.
  • Make sure there is significant unique content above the fold.
  • If you have more ads than unique content, you are probably going to have a problem.

Duplicate content

There should be one URL for each piece of content
  • Do URLs include parameters or tracking code? This will result in multiple URLs for a piece of content.
  • Does the same content reside on completely different URLs? This is often due to products/content being replicated across different categories.
Pro Tip:
Exclude common parameters, such as those used to designate tracking code, in Google Webmaster Tools. Read more at
Search Engine Land.
Do a search to check for duplicate content
  • Take a content snippet, put it in quotes and search for it.
  • Does the content show up elsewhere on the domain?
  • Has it been scraped? If the content has been scraped, you should file a content removal request with Google.
Sub-domain duplicate content
  • Does the same content exist on different sub-domains?
Check for a secure version of the site
  • Does the content exist on a secure version of the site?
Check other sites owned by the company
  • Is the content replicated on other domains owned by the company?
Check for “print” pages
  • If there are “printer friendly” versions of pages, they may be causing duplicate content.

Accessibility & Indexation

Check the robots.txt

  • Has the entire site, or important content been blocked? Is link equity being orphaned due to pages being blocked via the robots.txt?

Turn off JavaScript, cookies, and CSS

Now change your user agent to Googlebot

PRO Tip:
Use
SEO Browser to do a quick spot check.

Check the SEOmoz PRO Campaign

  • Check for 4xx errors and 5xx errors.

XML sitemaps are listed in the robots.txt file

XML sitemaps are submitted to Google/Bing Webmaster Tools

Check pages for meta robots noindex tag

  • Are pages accidentally being tagged with the meta robots noindex command
  • Are there pages that should have the noindex command applied
  • You can check the site quickly via a crawl tool such as Moz or Screaming Frog

Do goal pages have the noindex command applied?

  • This is important to prevent direct organic visits from showing up as goals in analytics

Site architecture and internal linking

Number of links on a page
Vertical linking structures are in place
  • Homepage links to category pages.
  • Category pages link to sub-category and product pages as appropriate.
  • Product pages link to relevant category pages.
Horizontal linking structures are in place
  • Category pages link to other relevant category pages.
  • Product pages link to other relevant product pages.
Links are in content
  • Does not utilize massive blocks of links stuck in the content to do internal linking.
Footer links
  • Does not use a block of footer links instead of proper navigation.
  • Does not link to landing pages with optimized anchors.
Good internal anchor text
 
Check for broken links
  • Link Checker and Xenu are good tools for this.

Technical issues

Proper use of 301s
  • Are 301s being used for all redirects?
  • If the root is being directed to a landing page, are they using a 301 instead of a 302?
  • Use Live HTTP Headers Firefox plugin to check 301s.
“Bad” redirects are avoided
  • These include 302s, 307s, meta refresh, and JavaScript redirects as they pass little to no value.
  • These redirects can easily be identified with a tool like Screaming Frog.
Redirects point directly to the final URL and do not leverage redirect chains
  • Redirect chains significantly diminish the amount of link equity associated with the final URL.
  • Google has said that they will stop following a redirect chain after several redirects.
Use of JavaScript
  • Is content being served in JavaScript?
  • Are links being served in JavaScript? Is this to do PR sculpting or is it accidental?
Use of iFrames
  • Is content being pulled in via iFrames?
Use of Flash
  • Is the entire site done in Flash, or is Flash used sparingly in a way that doesn’t hinder crawling?
Check for errors in Google Webmaster Tools
  • Google WMT will give you a good list of technical problems that they are encountering on your site (such as: 4xx and 5xx errors, inaccessible pages in the XML sitemap, and soft 404s)
XML Sitemaps  
  • Are XML sitemaps in place?
  • Are XML sitemaps covering for poor site architecture?
  • Are XML sitemaps structured to show indexation problems?
  • Do the sitemaps follow proper XML protocols
Canonical version of the site established through 301s
 
Canonical version of site is specified in Google Webmaster Tools
 
Rel canonical link tag is properly implemented across the site
Uses absolute URLs instead of relative URLs
  • This can cause a lot of problems if you have a root domain with secure sections.

Site speed


Review page load time for key pages 

Make sure compression is enabled


Enable caching


Optimize your images for the web


Minify your CSS/JS/HTML

Use a good, fast host
  • Consider using a CDN for your images.

Optimize your images for the web

Mobile

Review the mobile experience
  • Is there a mobile site set up?
  • If there is, is it a mobile site, responsive design, or dynamic serving?


Make sure analytics are set up if separate mobile content exists


If dynamic serving is being used, make sure the Vary HTTP header is being used

Review how the mobile experience matches up with the intent of mobile visitors
  • Do your mobile visitors have a different intent than desktop based visitors?
Ensure faulty mobile redirects do not exist
  • If your site redirects mobile visitors away from their intended URL (typically to the homepage), you’re likely going to run into issues impacting your mobile organic performance.
Ensure that the relationship between the mobile site and desktop site is established with proper markup
  • If a mobile site (m.) exists, does the desktop equivalent URL point to the mobile version with rel=”alternate”?
  • Does the mobile version canonical to the desktop version?
  • Official documentation.

International

Review international versions indicated in the URL
  • ex: site.com/uk/ or uk.site.com
Enable country based targeting in webmaster tools
  • If the site is targeted to one specific country, is this specified in webmaster tools? 
  • If the site has international sections, are they targeted in webmaster tools?
Implement hreflang / rel alternate if relevant
If there are multiple versions of a site in the same language (such as /us/ and /uk/, both in English), update the copy been updated so that they are both unique
 

Make sure the currency reflects the country targeted
 
Ensure the URL structure is in the native language 
  • Try to avoid having all URLs in the default language

Analytics

Analytics tracking code is on every page
  • You can check this using the “custom” filter in a Screaming Frog Crawl or by looking for self referrals.
  • Are there pages that should be blocked?
There is only one instance of a GA property on a page
  • Having the same Google Analytics property will create problems with pageview-related metrics such as inflating page views and pages per visit and reducing the bounce rate.
  • It is OK to have multiple GA properties listed, this won’t cause a problem.
Analytics is properly tracking and capturing internal searches
 

Demographics tracking is set up

Adwords and Adsense are properly linked if you are using these platforms
Internal IP addresses are excluded
UTM Campaign Parameters are used for other marketing efforts
Meta refresh and JavaScript redirects are avoided
  • These can artificially lower bounce rates.
Event tracking is set up for key user interactions

This audit covers the main technical elements of a site and should help you uncover any issues that are holding a site back. As with any project, the deliverable is critical. I’ve found focusing on the solution and impact (business case) is the best approach for site audit reports. While it is important to outline the problems, too much detail here can take away from the recommendations. If you’re looking for more resources on site audits, I recommend the following:

Helpful tools for doing a site audit:

Annie Cushing’s Site Audit
Web Developer Toolbar
User Agent Add-on
Firebug
Link Checker
SEObook Toolbar
MozBar (Moz’s SEO toolbar)
Xenu
Screaming Frog
Your own scraper
Inflow’s technical mobile best practices

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!

Reblogged 4 years ago from moz.com

Blog Optimization Service – Page Speed Optimization

Page Speed Optimization,Blog Optimization Service – http://www.seoblogpackages.com.

Reblogged 5 years ago from www.youtube.com