Profile cover photo
Profile photo
Ammon Johns & Company
293 followers -
Internet Marketing Consultant, Innovative Web Marketing and SEO Pioneer
Internet Marketing Consultant, Innovative Web Marketing and SEO Pioneer

293 followers
About
Ammon Johns & Company's interests
Ammon Johns & Company's posts

Post has attachment
*Google Q&A (#5) 'Crawling and Rendering' *

http://webpromo.expert/google-qa-crawlingrendering/

This series of Q&A Hangouts with Googlers has been both fun and fascinating.

Fun because all of those involved are good humoured, genuinely good people to have a conversation with. And they still laugh at my dancing pandas backgrounds, or whatever else I decide to bring along.

Fascinating because, let's face it, Google itself is absolutely fascinating on so many levels. The sheer size and scope of what Google does is utterly amazing. That some of their absolute finest technical attainments are virtually unknown, completely unappreciated by most of its users, even those who consider themselves fans, only adds to that for me.

Well, this time around we get to touch on one of those truly amazing parts that are rarely appreciated - Crawling and Rendering.

Google crawl and index millions of pages per day, and billions, perhaps even trillions per month. The sheer scale of this is utterly mind-boggling, and the logistics of it - to simply grab all of those URLs without flooding any one server, even when it is virtually hosting hundreds of different sites, is truly fantastic.

Imagine having 10 billion pages to index. Imagine having to do that in a month, every month. Lets say there are 30 days in each month. Multiply that by 24 to get a number of hours, then multiply that by 60 to get the number of minutes in a month. Finally multiply that sum by 60 to calculate the total number of seconds in the month.

30 x 24 x 60 x 60 = 2,592,000

Now divide 10 billion by that number to work out how many pages Google have to grab and process in every single second, just to index those 10 billion documents once per month.

A little over 3,858 pages per second.

Now, remember that Google need to index many pages multiple times each month, and news sites multiple times per hour. They also like to have the freshest possible idea of what's on the homepage and category pages of important/popular online stores too, of course.

Then think about how few pages actually load in one second or less.

Then add to all of that, the whole thing about how no server could handle that kind of load from a spider, and how Google have to intelligently split up that whole task, carefully scheduled, so even with hundreds of sites on one server, Googlebot isn't taking servers down all over the web day in and day out.

Seriously, I'd rate this as one of the most complex pieces of logistics imaginable, and Google do it with such ease and grace that most people will never even think about it.

In our discussion, +Ammon Johns+Bill Slawski, +Eric Enge and +Jennifer Slegg once again get to put their questions to Google's +Andrey Lipattsev and +John Mueller  - hosted by +Anton Shulke of +WebPromoExperts (дистанционное обучение) 

We'll be discussing what Google can crawl, how it prioritizes, whether Crawl Budget is something you should think about, and of course, questions about how Google renders and indexes things like javascript, content in areas that only load later, as the page scrolls, or as tabs are clicked, and more.

Join us.

http://webpromo.expert/google-qa-crawlingrendering/

Post has attachment
SEO in the year 2020?

A little while back I was interviewed by +David Bain on behalf of +Authoritas as part of a series of such interviews discussing the future direction of SEO and SEM.

https://www.authoritas.com/2016/06/29/ammon-johns-seo-2020-episode-10/

Be warned that the transcript of the 31 minute video gets a few words and phrases wrong, but I think it still communicates all of the salient points clearly enough.

Post has attachment
Back to the SEO Future

...is such a deliciously fun title.  But the article itself is a lot more than just fun.  When +Julie Graff of +Pole Position Marketing got in touch to ask me to answer a few questions, I never imagined that the result would be so interesting.

http://www.polepositionmarketing.com/emp/seo-future-per-pioneers/ 

At this time of year, through into the New Year, these "Expert Prediction Round-Up" posts are everywhere.  Usually overhyped and filled with too much of the exact same answers from too many "Experts".  Huge kudos then to Julie for adding a twist that really did add some value.

This is a smaller group of folks than many, but the answers are more in depth and insightful too.  Julie picked her group of panelists purely from the real long-timers and pioneers of the industry who have kept on rocking from the '90s right up to today.  People who not only therefore know a thing or two about SEO, but have been able to keep up with this ever-changing career since its inception.

Panelists include +Adam Audette+Bruce Clay, +Christine Churchill+Danny Sullivan, +Heather Lloyd-Martin and +Stoney deGeyter 

Well, enough of me gilding the lily, just go read the article for yourself at http://www.polepositionmarketing.com/emp/seo-future-per-pioneers/ 

Post has attachment
This was a fairly deep discussion, with some very tricky issues touched on, so I'll simply break this down by the major question points.

3:00 Ammon asks, I imagine that there must have been many times where something that seemed like it would be a very reliable and informative signal of quality turned out to be much poorer and more noisy than anticipated.  Can you tell us about any one of those?

6:38 Jon asks, We know there is a balance between organic and advertising, but from SEO it seems organic is disappearing, how do you share the landscape?  Is organic disappearing behind advertising?

12:56 Bill asks, In the past you've had quality scores influence paid position, but also many patents reference quality scores.  Can you tell us what webmasters can do to improve quality scores?

17:43 Ammon asks How much does social feed into search?  We know that plusses and likes are not good signals, but what is the benefit to Google of having access to the Twitter firehose?

20:36 Ammon follows-up, Using site:twitter.com shows little results (equal to just a few days worth), is the main value in Twitter in detecting "burstiness" and temporal context change?

23:08 Jon asks, in the past Google often shared what they were targeting with spam, today it seems quieter.  Have you fixed all spam, or just not making as many examples?  What are unsolved issues you are still fighting?

29:50 Bill asks, Can you tell us how important you see the role of schema and structured data to Google?  How much markup should we be using?

32:55 Ammon follows up with, Given that markup for Authorship was only a small proportion of the overall corpus, and abandoned, is structured data markup also going to be a small proportion mainly of value simply to help train machine learning?

35:06 Jon asks, as Mobile increases, and things like Siri and Google Now give one answer, and as you said earlier that the idea of "ten blue links" is in the past, what does 2018 look like?

42:45 Ammon redirects, Sure in the past copyright meant once we published, if others wanted to reuse and repackage our information they paid us.  But answer boxes etc are sometimes seen to be taking away advertising money from publishers ...

45:29 Ammon redirects again, well, Google wants this traffic enough to be devising systems to provide those answers, to capture those users...

48:44 Bill asks, I was watching a workshop from the FTC and cross device tracking.  It opened up a new avenue of analytics and advertising I wasn't so aware of from the SEO side.  You can't turn the mic off on an android device ... you can hear what TV shows I watch and what music I listen to so you could track.  Is that the future of connectedness?

54:29 Ammon says, I've been joking in the side chat, but the amount of data Google have, they have made themselves the biggest, fattest target for the NSA, CIA, MI6, and every spy network in the world.  Its not that I don't trust Google as much as that I don't trust organisations to not put undue pressure on Google.  Until Google can float off into International waters in Carriers filled with datacenters...

56:00 Bill brings us the Google Aware project that was cancelled for being "too invasive".

56:25 Jon asks, bring it back to data, Google Analytics is so widely used, I'm surprised Google say they don't use this data for search quality.  Why?  Do you just not need it?

1:00:40 Bill asks, I've seen references to a "media consumption history" and how that could influence search results so what videos I'd watched might affect my search results...

Post has attachment
0:00 Introduction of the session and guests.

0:50 Why we have divided the topic into three sessions: what engines want from search, what users want from search, and what business wants from search. 

1:20 Where did it begin?  A brief overview of search, from where it started just looking at file names and title, to full text indexing, and beyond into semantic and contextual search.

5:05 Search is like a referral - you ask if they know something and they give you a referral.

6:06 Jon Kleinberg, Hubs and Authorities, and mathematically modelling networks

7:34 False simplicity, the Pareto Principle, and Power Laws.

8:50 Link building, what started it?  PageRank doesn't rank pages, and the issues of canonicals.

10:30 Is link building dead or is link building still vital?  Yandex vs Google, and depths of link analysis.

11:48 Links are still vital, but link building is dead.  A quick explanation of weighted links and authority.

14:10 How does Google or any other engine determine Authority, Expertise, and Reputation?

16:39 Where have search engines failed at relevance and how have they evolved for this?

16:58 The problem of mathematical models (the basis of algorithms)

19:17 Many people still say "Google is a machine, I can always outsmart a machine", Ammon shows the flaw in this logic.

20:20 Short term exploits vs long term SEO strategy.

20:40 Commercially motivated search is the biggest part of SEO, but a very small proportion of actual search use (5% at best), which creates a bias of perception.

22:50 Audience question: +Phil Aston asked if SEO is still the right term in 2015 and what the move of companies like Moz away from branding just as SEO means today.

25:01 SEO is evolving because where you rank in search is only valuable if it is the right search, made by the right people, and you are able to convert against what competitors also offer.

25:34 Vin talks about how the face-to-face or telephone sales process matches the search-based sales process.

26:10 Semantic search and intention-based search results

27:33 Google came late to the party, how it gathered employees from Altavista, and the shortage of experts in IR (Information Retrieval).

29:48 How does paid search such as adwords create conflict with the prime content of organic search?

33:47 What single thing influences the search engine the most?

34:01 Ye venerable olde "200 Ranking Factors" - a decade out of date almost.

34:53 Bayesian Probability in spam, or search ranking.

36:59 Search signals are complex and contextual, not on simple set of universal rules.  The thing that would get you banned in one sector may be perfectly okay, even a positive, in other sectors.  Links from porn sites may be bad for one site, yet are fully expected for a porn site.

40:19 Audience question: +Roger Mudd asks whether 200 ranking factors stil make a good point of foundation even if long out of date?

41:05 The importance of testing and study.

42:58 The three ways that good SEOs get their knowledge and data (Testing, Research, and Trend Prediction).

46:42 There are too many SEOs chasing the algorithm, so they do great work that 3 months later stops working and was worth nothing in the long term.

46:51 A good time to dive deeper into semantic search and personalization - are keywords Dead?  Where do search engines want to go with semantic search? 

48:44 When using language (and keywords) you need to consider varieties of context in the broad range of search intents.

51:53 The big take-away from the session: Google and other engines don't make up arbitrary rules for algorithms that are plucked out of the air.  The algorithms are based on mathematical models of how people search, and what they hope to find

54:40 Vin and Audience question from +Peter Lunn - How does personalised, contextual search results matter to traditional ideas of ranking, when possibly no two users get the same results and rankings?

Post has attachment
The most important discussion I've ever recorded

That's how I have to describe any talk about the fundamentals of Marketing.  Marketing is the most misunderstood word in the English language.  Because we have all heard the word, often, we think we've figured it out.  But the word doesn't really mean what it seems to.  Watch just the first 5 minutes to see what I mean.
Wait while more posts are being loaded