How To Do SEO free course 2015

season 2015

RoCkhAcKeRzZ
Rockhackerzz.blogspot.com
Rao Sohaib

76 lactures on SEO

%100 granted working

What is SEO?

======================
SEO stands for Search Engine Optimization. SEO is everything that helps
a website generate more revenue by converting traffic from search
engines into leads or purchases.
SEO is traditionally identified with the techniques that help improve the
rankings of web pages on Google – this was just one of the visible effects
of SEO.
The basics of SEO can be applied not only to generic search engines like
Google, Baidu or Yandex but also to vertical ones like Indeed or Yelp,
social networking services like FacRoCkHaCkErZz or LinkedIn and to virtually all
repositories of content with search engine functionalities.
This RoCkHaCkErZz focuses on SEO for Google because most of users have a
strong preference for Google, not only as a generic search engine, but
also as their gateway to the Internet.
E.g. when a user thinks about checking a movie on the Internet Movie
Database website, he or she will often just write “imdb” on Google and
click on the first result rather than directly typing the domain name and
extension “imdb.com” in the address bar of their browser.
Which are the differences between technical and off-page SEO?
Two SEO approaches are required to drive users from search engines to
websites: Rockhackerzz.blogspot.com and off-page SEO.
Rockhackerzz.blogspot.com is everything related to a page and a website that is under
the direct and usually immediate control of web developers and
webmasters. This document is focused on Rockhackerzz.blogspot.com almost
exclusively.
Off-page SEO is everything external to the development of a website like
content marketing, link building and social sharing, which are not under
the direct control of developers and webmasters.
In the early days of Internet, search engines simply did not use links as a
ranking factor. Websites managed to show up on search engines’ results
SEO Rao Sohaib

7 Paradigma Tecnolgico – Rao Sohaib
just by minding basic on-page SEO guidelines, like inserting titles on their
HTML documents.
In addition to on-page and off-page SEO, content marketing is the other
essential component for a sustainable and profitable online business.
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
============================

Why is SEO important?

===============================
Being visible and ranking high on Google results not only in inbound
traffic, but also in trustworthiness, authority or empowerment of
prescription for websites and businesses of all sizes, markets and
languages.
Figure. Screenshot of Google search box for the query “why is SEO”. The search engine
suggests auto-completions of related queries
Washingtonpost.com is a news website. It enjoys the massive awareness
and brand reputation of its offline precursor, The Washington Post. You
might think that the newspaper does not “need” Google for its business.
The website however actively helps Google find all of the content by
posting a comprehensive and updated sitemap.
The file http://www.washingtonpost.com/robots.txt includes the line:
Sitemap: http://www.washingtonpost.com/web-sitemapindex.xml
By pointing the crawling bots to its sitemap, The Washington Post is
investing in their SEO for profit.
SEO Rao Sohaib

8 Paradigma Tecnolgico – Rao Sohaib

================

Is SEO free?

===============================
SEO is definitely not free. It is however hugely cost-effective in
comparison to any other investment in the marketing mix.
In addition to getting good training and reading books like this one, we
generally advise to work with professional SEOs and digital strategists to
save time and resources later in the lifetime your project.
Long term success in search marketing requires best practises,
experience and intuition. SEO has a reputation of being a trade restricted
to a few in the know. This stereotype is rooted in the fact that, so far,
there is no tool, automated method or machine learning approach that
manages to squeeze all the value of all the SEO developments. That is
why you will not find “SEO tricks” on this RoCkHaCkErZz.
Organic traffic flows in when great content meets good SEO. Content is
King - as long as it is fresh, relevant or engaging. SEO just makes it easy
for bots to find, index and rank websites with the right content. However
even the best content needs to be published efficiently so that search
engines find it and deal with it.
The conclusion is that good SEO requires experts and content, neither of
which come cheap, but it generates potentially massive amounts of
traffic with high rates of conversion over the long term.

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
====================================

Google’s official stance on SEO

=============================================
While many affiliates and some SEOs are known for trying to
systematically out-smart search engines with short term tactics, the bestpractice SEO requires patience, experience, good relations within the
search industry, great communication skills and, above all, an avid
curiosity.
Google recently claimed that “Many SEOs and other agencies and
consultants provide useful services for website owners”i.
The relationship between Google and the marketing industry is rich and
complex. Google communicates regularly with the SEO industry and
SEO Rao Sohaib
9 Paradigma Tecnolgico – Rao Sohaib
provide tools, posts on forums by Google employees, videosii, etc to
webmasters and marketers.
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
================

Jargon buster

==================
These terms will help you understand some of the concepts used in this
RoCkHaCkErZz. We list them in alphabetical order and some of them are a bit
abstract but please soldier on:
x AJAX is a number of techniques to create client-side asynchronous
web applications brought together by JavaScript. Rich features
using AJAX are popular because they usually improve the user
experience by efficiently refreshing content. AJAX is however an
issue for Google’s crawler because it cannot read its content.
Google expects that developers carry out some extensive hacking
to deal with AJAX (see section “Make AJAX crawl-able” below).
x Backlinks or inbound links are links from external websites
pointing to another website as opposed to internal links from
pages on one website to other pages on the same site. Backlinks
are not to be confused with the links to search results on Google
SERPs.

x Corpus is a collection of documents in a machine-readable format,
usually text. Examples of corpora (plural of corpus) are dumps of
databases of any nature and format, the scraping of a website or
any number of websites, etc.
x Crawler or web spider is a bot programmed to browse
systematically websites for the purpose of indexing, like Googlebot
x CTR stands for Click Through Rate or clicks on a search result
divided by the number of impressions (or how many times it
showed on any SERP)
x Document is a piece of text or rich media that can be accessed and
stored individually. An example of a document is a webpage or a
downloadable pdf file.
x Graph is the interconnection between documents (vertices) by
edges (links). An example of a graph is the network of websites
linked with each other.
SEO Rao Sohaib

10 Paradigma Tecnolgico – Rao Sohaib
Figure 1 of a graph of 5 million edges and an estimated 50 million hop count by the Opte
project modelling the Internet in 2003. Colours modified from the original representing
different regions of IP addresses like Asia Pacific, Europe, North and South America, etc.
The concept of graph is a key concept in SEO. Many projects of new
websites usually start their life as the output of a number of
functionalities and ad-hoc extensions, rather than a body of
interconnections. Search engines and SEOs however think of websites
and actually the entire Internet as a graph.
x Information Architecture (abbreviated as IA) is the organization of
documents and their connections. Websites are, in terms of IA,
dynamic and connected structures of bot-readable content.
Rockhackerzz.blogspot.com is mostly applied IA for search.
x PageRank is a metric used by Google to determine the importance
of an element (e.g. document, graphs or parts of them). It is one of
more than 200 factors used to determine rankings of search
results on Google. SEOs tend to prefer concepts like link juice or
authority and new metrics to Google’s PageRank.
SEO Rao Sohaib
11 Paradigma Tecnolgico – Rao Sohaib
x SEOs or search marketers are professional practitioners of SEO.
x SERP is the acronym of Search Engine Results Page or the list of
links to results that search engines return in response to a user’s
query, e.g. http://www.google.com/#output=search&q=serp
x Silos are groups of subject-specific content on websites, e.g.
categories separated as tree or sub-categories and detail pages.
Vertical silos are frequent in tree structures where category pages
are linking down to sub-category pages. The webpages under silos
are hardly linked with webpages of other silos, i.e. there are few or
no transversal or cross links across silos.
SEO Rao Sohaib
12 Paradigma Tecnolgico – Rao Sohaib
++++++++++++++++++++++++++++++++++++++++++++++++++++

==============================================

SEO deals with the bottlenecks in the search flow

===============================================
SEO exists because people think and write in a different way from how
search engines work.
There are a number of components in the search process that result in
bottlenecks in the flow of information. The bottlenecks are inefficiencies
that may result in a poor match between the search intent of the user
and the purpose of the author or publisher.
Figure 2 the flow of search is represented on the diagram above from the left (users) to the
right (content). There are a few bottlenecks that affect the efficiency of the search.
SEOs can only try to understand, but not influence, the systemic
bottlenecks: from the true meaning of keywords and search intent to
Google’s limitations.
Web developers and SEOs can optimize or have a direct control over the
rest of bottlenecks down the flow: mostly speed, structure and the
content itself.
SEO Rao Sohaib
13 Paradigma Tecnolgico – Rao Sohaib
+++++++++++++++++++++++++++++++++++++++++++++++

=============================

Bottleneck 1: Limitations of keywords

=====================================

Meaning of keywords
Many words convey different meanings. This is a challenge both for
search engines and for SEOs.
E.g. “Metro” is a word that, when combined with the name of a location
as in “metro {location}”, might mean different searches in Europe,
Canada and US:
- a local railway transport
- a local section of a newspaper or local news paper
- a brand of food stores
Search intent
There are a few possible search intents:
- Informational: wikis, news, blogs and publishing sites
- Navigational: video hosts, social networks
- Commercial: informational search with future transactional
implications, e.g. vertical search engines, classified aggregators,
price comparison sites
- Transactional: retail, e-commerce sites
Google tries to estimate search intent from users’ previous activity and
context but it is a formidable challenge for a generic search engine.
Changes in business models are often followed by SEO adjusting to
different search intent than the one previously targeted.
E.g. When paywalls are introduced in newspapers, their SEO adjusts from
informational intent (advertising intermingled with content) to
transactional intent (content only with restricted access).
Google search features
The average number of words per query by users increased steadily from
2+ to 4+ search terms (excluding stop words) in the last few years. We
users educated ourselves to write longer queries.
SEO Rao Sohaib
14 Paradigma Tecnolgico – Rao Sohaib
Google recently started to alter the search behaviour of users by limiting
the long tail of search queries by tapping into the search data of users:
- Auto-complete search suggestions: Google displays suggestions
that might be related to the one you are typing. This influences
your query by modifying it or by accepting the one Google
suggests instead of the one you originally intended to write.
E.g. when you type just “bank”, Google suggests
Bank of America
Bank
Bank of Scotland
- Google Instant: Google updates its SERP with different results as
you type your query in the search box. This deters many users
from typing queries longer than three or four search terms
because they are presented with results earlier on.
Search engines’ bias
Brands, media titles, universities and institutions are said to be overrepresented on Google SERPs. Google probably deals with entities in
their graph and training dataset that might be equivalent to what we
humans refer to as brands.
Google discretionary classifies reputation and authority of those entities
according to their undisclosed criteria. Google adjusts their criteria
logarithmically or with manual actions.
When Google releases multiple or significant adjustments at the same
time, a significant amount of websites might be affected. These
situations are known as updates. The best known recent updates are
called Google Panda and Penguin.
SEO Rao Sohaib
15 Paradigma Tecnolgico – Rao Sohaib

=================================================

Bottleneck 2: The World Wide Web & Search Engines

==============================================

Bottleneck 2: The World Wide Web & Search Engines
The World Wide Web (WWW) is an unstructured, de-centralized and
ever changing ecosystem. Google tries to make sense of it with different
types of software:
- Crawling software: bots are fast and greedy but blind to whatever
is not text. Search engines try to cope with users creating
incomplete, biased or inadequate connections between pieces of
information. Google in particular is trying to cope with the surge of
social networks.
The quality of links as indicators of relevancy or popularity is also
evolving: it is more convenient to link to content via social shares
than with traditional hyperlinks on web pages.
- Indexing software: the main current challenges are identifying
unique content and attributing authorship. Valuable content is
often very duplicated across the WWW.
- Ranking software: Google claims to fight web spam with training
data built for machine learning algorithms allegedly used for userspecific and session-specific rankings.
++++++++++++++++++++++++++++++++++++++++++++++++++++++
===============================================

Bottleneck 3: Web servers, websites and code

==================================
The performance of websites in terms of access to findable content and
downloading speed depends on two main factors:
x Site architecture: the distribution of the information in a structure
of categories or sections organised by topic and other criteria
x Page speed: from the point of view of the crawlers, the code
engineered for simplicity and speed is key for downloading the
content from the server fast
When the speed of the web servers or the availability of networks is
poor, the flow of information gets disrupted.
++++++++++++++++++++++++++++++++++++++
==============================

Bottleneck 4: The content itself

=================================
The publication of the content can itself be a bottleneck. SEO optimizes
the IA of the content to match the requirements of search engines:
SEO Rao Sohaib
16 Paradigma Tecnolgico – Rao Sohaib

Mark-up: Poor or missing tagging frequently leaves search engines to
classify content on their own without the help from its publishers.
Format: Search engines are good at interpreting text only. They only get
the mark-up of images and videos – if it exists at all. All the client-side
interaction with AJAX, Flash content, etc. are totally lost to search
engines.
Duplication: If there are duplicated versions of the same content in
multiple documents, search engines have trouble identifying the original
document or even the first to be detected.
Attribution: Attribution of authorship is, like uniqueness, a hard issue for
search engines to deal with.
SEO Rao Sohaib
17 Paradigma Tecnolgico – Rao Sohaib
+++++++++++++++++++++++++++++++++++++++++++++
=====================

SEO for new websites

=========================
Web developers that pay attention to on-page SEO in the early stages of
new projects usually save lots time once in production. Critically, a well
thought-through site architecture and prioritization of conversion funnels
scale up beautifully.
We list below those SEO techniques that you need to consider in loose
chronological order:

+++++++++++++++++++++++++
======================================

Why is Rockhackerzz.blogspot.com relevant?

=======================================
Nowadays on-page SEO remains an important technique in the online
marketing toolbox because:
1. Web servers and search engines only have in common the fact that
they are extremely fast but essentially dumb machines. The
Rockhackerzz.blogspot.com helps close the gap between both systems.
2. Search engines fall short of the expectations of users: it is very
hard to determine search intent of a query, never mind matching it
with the purpose of the content.
SEO helps close the gap between software and users and minimize their
limitations.
+++++++++++++++++++++++++++++++++++
===================

Hosting

=========================

++++++++++++++
===========

Host your site on reliable servers with excellent connectivity

================================================================
You need a server uptime of 99.9% or higher over any period of time and
as much bandwidth, memory and processing power as it takes. The good
news is that all of the infrastructure costs keep dropping in price over
time.
Measure the number of hops from your LAN the host of your website. If
you are using a well interconnected local ISP, chances are that the
number of hops that you are measuring is not too different from the
number of hops that sets apart the Googlebot from your server.
The command traceroute (tracert on Windows) displays the path and
transit delays of packets across a given route at a point in time.
SEO Rao Sohaib
18 Paradigma Tecnolgico – Rao Sohaib
E.g.
$ traceroute github.com
Tracing route to github.com [207.97.XXX.XXX] over a maximum of 30
hops:
1 1 ms 1 ms 1 ms XXX.XXX.XX.X
2 2 ms 4 ms 2 ms XX.XX.X.XX [XX.XX.X.XX]
3 1 ms 1 ms 1 ms XX.XX.X.XX
4 1 ms 1 ms 1 ms XX.XX.X.XX
5 3 ms 3 ms 3 ms mad-b1-link.telia.net [213.248.93.21]
6 32 ms 30 ms 30 ms prs-bb1-link.telia.net [80.91.245.58]
7 123 ms 123 ms 112 ms ash-bb3-link.telia.net [80.91.251.98]
8 117 ms 123 ms 111 ms ash-bb1-link.telia.net
[80.91.248.161]
9 217 ms 118 ms 123 ms rackspace-ic-138625-ashbb1.c.telia.net [213.248.98.218]
10 260 ms 213 ms 274 ms vlan905.core5.iad2.rackspace.net
[72.4.122.10]
11 118 ms 118 ms 124 ms aggr301a-1-core5.iad2.rackspace.net
[72.4.122.121]
12 118 ms 118 ms 118 ms github.com [207.97.227.239]
Trace complete.
There were 12 hops between my computer and github.com (the IP
addresses are obfuscated in the example above) at the time of executing
the command.
If you are physically in your target local market (country) and you are
connected to an incumbent telecom operator, it is safe to assume that
most of your visitors will be at roughly the same network distance as you
are from the host.
If you are offered two or more hosting locations by the same host and
your users are evenly distributed worldwide, you may want to choose
the location with the lower average count of hops as measured a number
of times.
++++++++++++++++++++++++++++++++++++

================================================

Check your neighbours in shared hosting environments

===============================================
Use reverse IP services like http://reverseip.domaintools.com to find
which other websites share your hosting server. You do not want to be
SEO Rao Sohaib
19 Paradigma Tecnolgico – Rao Sohaib
on IP addresses blacklisted for email spammi1ng or carrying out illegal
activities.

++++++++++++++++++++++++++++++++++
=====================================

Hosting services with dynamic IP addresss

=========================================
You are usually safe in terms of SEO using cloud hosting, Content Delivery
Networks (CDN) systems, Amazon’s AWS or load balancing services that
change their IP address regularly. When IP addresses change often
Google “may crawl a bit more conservatively than if we've figured out
that your server is sturdy enough to be crawled at higher rates”iii.
Unless you are running a very large website with a high frequency of
publication of new content, the crawling rate should not be a problem in
this instance.
+++++++++++++++++++++
============================

Information Architecture

=======================
=====================

Google PageRank

=============
While dismissed by many SEOs as irrelevant nowadays, it may still be a
useful metric if only at the level of order of magnitude for comparison
and contextual purposes (an order of magnitude greater on a base 10 is
to be 10 times as large).
E.g. the ESA (European Space Agency) and the NASA both enjoy a
PageRank (PR) of 8, while the Indian Space Research Organisation
features a PR of 7 and the China National Space Administration a PR of 5.
Please note that the PR uses a logarithmic scale so a PR of 8 denotes
several times more weighting than a PR of 7.
PageRank is now just one signal in more than 200 metrics that Google
include in their ranking algorithms. The values of PageRank take
nowadays many months to be publicly updated by Google.
+++++++++++++++++++++++++++++++++++++++
========================================

Design a lean site architecture

==========================
Restrict the number of categories to low numbers that make sense to
your users, keep a flat hierarchy and make sure that your conversion
pages (products, SKUs, etc) are as few links away from the home as
possible.
SEO Rao Sohaib
20 Paradigma Tecnolgico – Rao Sohaib
Avoid a tree structure with complex, deep or duplicated branches.
Usually, but not always, the home is the page with the highest PageRank
of a website.
The PageRank is diluted with every level down the structure of categories
and sub-categories. When there is duplication, the PageRank is wasted.
Figure 3 represents part of a (fictional) architecture with 3 levels below the homepage. The
symbols curly brackets indicate variables. The type of page, SERP or detail, is indicated at the
top of the box. The URL is composed by appending sections to the page. For instance the URL
of the bottom left pages is domain/job/offers/{jobname}/in-{region}.
A simpler structure would consist in having only two levels below the
homepage. It would be more effective to index the same inventory of
content on Google. It also helps direct the user to the best converting
webpages, namely detail pages of job vacancies in the example.
It is important to discuss this IA with all the stakeholders of a project in
its very early stages. Business, usability, marketing and technical should
align to the best architecture to fit all their different interests.
SEO Rao Sohaib
21 Paradigma Tecnolgico – Rao Sohaib
Figure 4 represents part of a lean architecture of 2 levels below the homepage. The symbol
“?” indicates sections of the path that are facultative if there is a value for that variable. The
symbol # is a character hash on the URL path.
Many product owners and usability consultants tend to “know” how they
want to present their content to the end-users. Explain them the
concepts above and try to set up an architecture that makes it easy to
expand horizontally and vertically.
A successful architecture is often a trade-off between how a website
should look, work and how search engines deal with the organization of
information.
++++++++++++++++++++++++++++
==============================================

Link your internal pages sensibly

============================================
Densely connected graphs make crawling bots’ job easier in principle
because they can jump from node to node via multiple edges spreading
through the whole network. Following this rationale, the more links you
can insert on one page to the rest of the pages of the website, the better.
Avoid however linking to dozens of internal pages unless they are strictly
relevant and genuinely useful to visitors. Search engines learned to
discern the value of internal links by context, their topology and cooccurrence with other links on the same page. Just like users, search
engines disregard any excess of links on-page (see Topology of on-page
links below).
SEO Rao Sohaib
22 Paradigma Tecnolgico – Rao Sohaib
New websites have a PageRank value of zero at launch. If you already
have a certain amount of PageRank, Google is only willing to crawl so
much from your site. Administer your PageRank budgetiv wisely: promote
with internal links and backlinks only the pages that really matter to the
business. Do not push this to the point of obfuscating links with
JavaScript redirects or frames or of annotating them with nofollow
attributes (those techniques are known as PageRank sculptingv).
Backlinks are, for Google, a stronger signal of authority and popularity
than internal links. Just a reminder: this document is not covering the
techniques of off-page SEO and link building.

Large websites with a long tail of pages do not usually get backlinks for
each of them. Internal links are the next best thing you can do to position
all those pages. Ideally your internal links should be organised
consistently with the IA of the site. Decide the architecture first and then
the internal links.
You want to link internally in a relevant way for both the users and
Googlebot. It is OK to theme your website by topic organized in
directories but beware that some techniques that concentrate links to
and in between the pages of each directory are best left to Rockhackerzz.blogspot.com
professionals who have experience with silos and their risks.
Topology of on-page links
In the early days of the WWW it was assumed that documents are
essentially lists of hyperlinks connecting information. One of the
proposals to determine the relative value of each document was
published as the PageRank (for Larry “Page”, not for “webpage”).
The PageRank in its primordial version assigned each value to each link
regardless of its position on the document.
SEO Rao Sohaib

23 Paradigma Tecnolgico – Rao Sohaib
Figure 5 representing the flow of PageRank through documents via outgoing links in a
“random surfer” model. See how the document A passes equal value of PageRank, 0.2, from
its two links regardless of their position on the document. Source: Method for node ranking
in a linked databasevi
Google moved from a paradigm of random surfer to reasonable surfer.
Figure 6 representing a flow of PageRank through documents via outgoing links in a
“reasonable surfer” model. See how the document A passes more PageRank, 0.6, from the
hyperlink in the middle of the document than the PageRank passed by the hyperlinks
regardless of their position on the document. Source: Ranking documents based on user
behaviour and/or feature datavii
SEO Rao Sohaib
24 Paradigma Tecnolgico – Rao Sohaib

Place the most relevant content and links at the top of your page. Use
meaningful anchor texts.
If your layout includes many navigational links, like breadcrumbs, menus
or facets with nested menus try not to place them all on top of the page.
Unfortunately for us users many websites still list dozens of deep links at
the bottom of the page. Those stacks of hyperlinks are illegible at best
and a poor SEO practise in all cases.
Responsive websites
Browsers build the DOM of a web document, parse its tree to render its
HTML and then interpret the CSS.
Figure: Rendering engine basic flow. Source: HTML5rocks.com
The front-end development of responsive websites almost always
requires to displace content with CSS to adapt it to screens of many
geometries and interactions.
Try to follow the reasonable surfer model in placing code and rendered
content in the same order. Do not hide links, content or menus on
different versions of the responsive design: just position them sensibly.
You are on the safe side in terms of SEO as long as the bot parses the
content as you placed it on the HTML source code of the page - even if it
is not exactly in the order we humans read it on all devices.
Number of internal links
Help users find their way around with navigation facets, breadcrumbs,
menus and relevant pages. Restrict the number of internal links on each
page to a reasonable number.
SEO Rao Sohaib
25 Paradigma Tecnolgico – Rao Sohaib
There is not any clear-cut maximum and minimum limits of links; they
depend on the nature of the business or on the value proposition of the
website.
It is commonly accepted that 100 links or more on a page (either internal
links or outgoing ones) are, as a rule-of-thumb, an excessive number of
links.
E.g. enter this on your bash terminal after you have w3m installed
$ w3m "http://www.king.com"
By pressing “L” on w3m you will get a list of all links, anchors and images
of the current page. At the time of writing this guide we counted 373
links on a page. That looks like a few hundred too many! Are they all
relevant? The homepage of King.com is designed as a catalogue so such a
high number of links are to be expected.
The homepages of some e-commerce sites may need very few internal
links pointing to what really matters: trial, support and checkout.
E.g. enter this on your bash terminal and press “L”
$ w3m "http://www.perspectivemockups.com/"
We counted just 3 links on that page. Those are probably are all the links
that the website needed for business.
Alternatively, you may try to get a full list of links of a small website – do
not try the following one-liner on a large website:
$ MYSITE='http://www.website.com';wget -nv -r --spider
$MYSITE 2>&1 | egrep ' URL:' | awk '{print $3}' | sed
"s@URL:${MYSITE}@@g"
The shell will go quiet for a while until it lists all the links it found onpage.
SEO Rao Sohaib
26 Paradigma Tecnolgico – Rao Sohaib
Prevent broken links before going to production
Identify broken links, internal and outgoing with regular crawls and fix
them. It is easy to do it with just by fetching documents from the net
directly from a shell.
E.g. the Library for WWW in Perl has a bash command, lwp-request. If
you are using Perl on cygwin for Windows, you may have to install HTMLTree first so enter this on the shell:
$ perl -MCPAN -e shell
cpan[1]> install HTML::TreeBuilder
Then enter this one-liner back at the shell promptviii:
lwp-request -o links http://www.bigdataspain.org | perl -
pe 'chomp; $_ =~ s{ ^ \w* \s* ( [^#]+ ) .* $ }{$1}x; undef
$_ unless m/^http/; $_ = qq{\"$_\"\n} if $_' | sort | uniq
| perl -ne 'chomp; print $_ . qq{\t} . qx{lwp-request -ds
$_}'
The (long) one-liner results in this output:
"http://bigdataspain.us5.listmanage.com/subscribe/post?u=d86f829047abbeaff0d30973f&id=8
832bb34a7" 200 OK
"http://fonts.googleapis.com/css?family=Days+One"
200 OK
"http://microformats.org/profile/hcalendar" 200 OK
"http://www.bigdataspain.org/" 200 OK
"http://www.bigdataspain.org/2012/" 200 OK
"http://www.bigdataspain.org/2012/Terms-and-ConditionsBig-Data-Spain-conference.pdf" 200 OK
[…]
"http://www.bigdataspain.org/en/infographic-conference-
2012" 200 OK
"http://www.bigdataspain.org/jobs/en/" 200 OK
"http://www.flickr.com/photos/87479166@N04/" 200 OK
There are no broken links to report in this example. There are plenty of
online checkers that are fast and reliable also.
SEO Rao Sohaib
27 Paradigma Tecnolgico – Rao Sohaib
Link attribute rel nofollow
The nofollow value of the rel attribute is great to control the crawling of
specific pagesix.
The nofollow value helps Googlebot understand that it should not crawl
the linked document. In fact, the bot does crawl as much as it can for the
given PageRank of a website, including nofollow-tagged pages.
Google claims that nofollow links do not pass PageRank and that their
anchor text is skipped altogether.
This is useful for IA purposes when you try to avoid indexing duplicate
content on Google or to indicate Google that a link with rel=”nofollow” is
a paid link or an external document that you do not necessarily endorse.
This attribute was popular for PageRank sculpting X and waterproofing
silos. Those practices are not recommended.
+++++++++++++++++++++++++++++++++++++++++++

====================================================

Configuration of mobile rendering

===================================================
There are basically three approaches to SEO for mobile web (as in
webpages for interactive screens of smartphones and tablets):
1. Sites that serve all devices on the same set of URLs, with each URL
serving the same HTML to all devices. CSS just changes how the
page is rendered on the device. If all Googlebot user agents are
allowed to crawl the page assets (CSS, javascript, and images),
Google’s algorithms automatically detect responsive designs.
2. Sites that dynamically serve all devices on the same set of URLs,
with each URL serving different HTML (and CSS) depending on
whether the user agent is a desktop or a mobile device.
3. Sites that have separate mobile and desktop URLs ej.
http://m.website.com and http://www.website.com respectively
SEO Rao Sohaib
28 Paradigma Tecnolgico – Rao Sohaib
The first approach – the responsive web design – saves resources and it is
the one preferred by Google x
++++++++++++++++++++++++++++++++++++++++
==========================================

Uniform Resource Identifier URL

=====================================

URI Syntax

========================
URI stands for Uniform Resource Identifier. URLS Or Uniform Resource
Locators are just a subset of URIs describing the primary access
mechanism.
Eg. URLs http://www.example.org:80 or mailto:me@example.org
URI = scheme ":" hier-part [ "?" query ] [ "#" fragment ]
Example:
foo://example.com:8042/over/there?name=ferret#nose
\_/ \______________/\_________/ \_________/ \__/
| | | | |
scheme authority path query fragment
| _____________________|__
/ \ / \
urn:example:animal:ferret:nose
++++++++++++++++++++++++++++++++++++++++++
========================================
Compose a simple URL path
==========================
Web browsers request webpages from servers by using Uniform
Resource Locators (URLs). Write URLs paths as similar to natural
language as possible. Try to use keywords as a user would read them.
E.g. Majestic is a wine retailer in the UK. In 2013 the URLs of their SERPs
were for instance
http://www.majestic.co.uk/find/category-is-Wine/category-is-Spain
From the point of view of a user the URL reads unnatural. In terms of
SEO, the strings of text “category-is” in the URL path are probably
redundant.
In 2014 their equivalent URL is http://www.majestic.co.uk/spanish-wine
which is more legible and relevant.
SEO Rao Sohaib
29 Paradigma Tecnolgico – Rao Sohaib
Less is more about URL paths: try to limit the number of folders with
categories and subcategories to a maximum of three separated by
forward slash. Separate keywords by hyphens instead of underscores or
any other symbol.
URLs can theoretically be hundreds or even thousands of characters long
(depending on the browser) and still be operational but optimal ones
should of course be much shorter.
++++++++++++++++++++++++++++
===============================

URL encoding

==========================
A URI is a sequence of characters from a very limited set: the letters of
the basic Latin alphabet, digits, and a few special characters.
URLs can only be sent over the Internet using the ASCII set of printable
characters. Eg. in German /München/ is best written as /muenchen/ or
even /munchen/. Avoid punctuation symbols and blanks to compose
URLs. They encode in an ugly, unreadable strings of text if encoded with
percent-encoding in an URL.
E.g. The sentence in Spanish “La eñe es una letra del alfabeto español,
bretón y guaraní entre otros” looks terrible on a URL with encoding:
http://www.domain.com/La%20e%C3%B1e%20es%20una%20letra%20d
el%20alfabeto%20espa%C3%B1ol%2C%20bret%C3%B3n%20y%20guaran
%C3%AD%20entre%20otros
You want something like this, free from triplets including “%”:
http://www.domain.com/la-ene-es-una-letra-del-alfabetoespanol-breton-y-guarani-entre-otros
All programming languages have URL encode and decode functions to
normalize characters in the URI.
+++++++++++++++++++
=====================

Friendly URLs

====================
SEO Rao Sohaib
30 Paradigma Tecnolgico – Rao Sohaib
Pay particular attention to reserved charactersxi . Reserved characters
provide a set of delimiters to distinguish data within a URI.
reserved = gen-delims / sub-delims
gen-delims = ":" / "/" / "?" / "#" / "[" / "]" / "@"
sub-delims = "!" / "$" / "&" / "'" / "(" / ")"
/ "*" / "+" / "," / ";" / "="

Restrict the delimitation with sub-delims and gen-delims as much as you
can. You want to keep the number of parameters in queries to no more
than 2. This is a ballpark numbers suggested by Google long time ago. It
is a rule particularly applicable to sessions with logged in users or at
ecommerce checkouts with multiple fields in their forms. Those are
typically identified with symbols in URLs composed dynamically such as
question marks “?” and ampersands “&” combined with equal to “=”.
E.g. Zara is a fashion retailer whose website used to composed URLs that
contained too many folders and parameters in the path:
https://www.zara.com/webapp/wcs/stores/servlet/ShopCartPag
e?calculationUsageId=-
1&updatePrices=1&catalogId=24052&orderId=.&langId=-
1&storeId=10701&URL=ShopCartPage
E.g. pingg.com is an online e-cards retailer with shorter, more intuitive
URLs on its website. Their URLs are OK although they can be improved
http://www.celebrations.com/public_event/dw72fy6hz2sxyby8d
Client-side modifications of the content of a webpage that don’t reach
the server use fragment identifier preceded by a hash mark # (see
section “Make AJAX crawlable”).

http://www.domain.com/section1/subsection2#fragment
Google standardizes URLs by removing any fragments from the URL, i.e.
does not reach the optional last part of a URL for a document from the
hash onwards. See the section “Making AJAX crawlable”.
SEO Rao Sohaib
31 Paradigma Tecnolgico – Rao Sohaib
Prepare to scale up the URL paths. Internet allows for economies of scale
so it is only natural that new projects should prepare to scale across
languages, countries, categories and verticals.
There are no defined rules to deal with scaling. Check the Appendix
Domain names for some considerations about scaling vertically and
horizontally.
+++++++++++++++++++++++++++++++++++++++++++++++++++
==================================================================================

Automate the generation of URLs with intuitive rules

================================
Some frameworks escalate the generation of URLs easily, e.g. the URL
mappings plug-in on Grailsxii
URL re-writing is a technique to reverse engineer the links from the URL
mappings. Given a mapping:
static mappings = {
"/$blog/$year?/$month?/$day?/$id?"(controller:"blog",
action:"show")
}
If you tag as follows:
<g:link controller="blog" action="show"
params="[blog:'fred', year:2007]">
My Blog
</g:link>
<g:link controller="blog" action="show"
params="[blog:'fred', year:2007, month:10]">
My Blog - October 2007 Posts
</g:link>
Grails will automatically re-write the URL in the correct format on HTML:
<a href="/fred/2007">My Blog</a>
<a href="/fred/2007/10">My Blog - October 2007 Posts</a>
SEO Rao Sohaib
32 Paradigma Tecnolgico – Rao Sohaib
+++++++++++++++++++
-=================================

Mark-up your content

=========================
Insert tags into the appropriate HTML elements to define textual content
within a page.
Each webpage needs to be relevant to its content. After backlinks and
relevant anchor texts, marking up your documents is the most effective
technique to help Google correlate your documents with their content
purpose.
The importance of copywriting relevant and informative titles is often
underestimated. Good titles are important for the CTR of the links to
your results on Google’s SERPs. In some cases, titles help to associate
your brand to a category or niche if you include your brand name.
+++++++++++++
===================

Title Tag

=====================
Write descriptive titles less than 70 characters long in the head container
of your HTML document.
<head>
<title>Not too many keywords here</title>
</head>
Unless you have a reason not to, include your brand name in the title.
<head>
<title>Keyword/s Category | Your brand</title>
</head>
++++++++++++++++++++++
=====================

Meta elements

====================
It is usually trivial to generate, from a coding perspective, meta tags for
webpages. In real life however you will code webpages with missing,
duplicate, too long, too short or non-informative HTML meta elements so
reserve some time to check everything before beta launches.
Content is the attribute of the meta elements that help SEO the most.
The <meta> tags always go inside the <head> element.
SEO Rao Sohaib
33 Paradigma Tecnolgico – Rao Sohaib
Description attribute
Think of it as a tweet-long summary of the most relevant content on your
page. It should help you improve the Click Through Rate of your results
on Google’s SERPs.
<meta name="description" content="This attribute is a
great opportunity to improve your CTR if Google picks it
up and inserts it as a snippet on its SERP" >
Do not exceed 170 characters when composing the text on the tag.
Robots attribute
The values {index,noindex} and {follow,nofollow} are directives that help
you control the content that you want to index at page-level. They can be
search engine-specific. These meta tags are placed in the HTML header.
They can be effective also in combination with the robots.txt protocol so
you retain control on your content on Google.
<meta name="googlebot" content="noindex">
Language attribute

This helps search engine match the language of search queries and
locales with your content.
<meta http-equiv="content-language" content="pt-br" />
Keywords attribute
Common wisdom is that stuffing keywords on this meta tag is “SEO”. This
practise was so abused that Google gave up using the keywords attribute
altogether a few years ago.
You may still want to use it for your own purposes, like semantic tagging,
etc but definitely not for SEO.
SEO Rao Sohaib
34 Paradigma Tecnolgico – Rao Sohaib
++++++++++++++++++++++
======================

Headings

================
Organize the distribution of your content on-page with hierarchical
criteria. Place the most important content on the top and the least
relevant on the bottom of the webpage as the crawler (or a text-only
browser) reads it.
<h1> to <h6> tags are used to define HTML headings.
<h1>Most Relevant Keywords</h1>
<h2>Second Most Relevant Keywords</h2>
<h3>Specific sub-sections Most Important</h3>
Tagging with headings does not need to be strictly nested if it takes too
much of your time or the editors of the content cannot be briefed.
You can use more than one h1 tag on the same page in HTML5. HTML5 is
the latest standard for HTML.
++++++++++++++++
===============================

Main and aside tags

========================
HTML5 enables tagging content-specific elements like <section>, <main>
and <aside>. Its impact on SEO is probably null still in 2014.
++++++++++++++++
====================================

Rich media (images, videos)

================================
Mark up your videos following schema.org’s indications and provide alt
descriptive and accurate attributes on images.
Not good
<img src="conference.jpg" alt=""/>
Acceptable
<img src="conference.jpg" alt="red dot on white
background"/>
Best
<img src="conferences.jpg" alt="the flat of Japan">
SEO Rao Sohaib
35 Paradigma Tecnolgico – Rao Sohaib
++++++++++
========================
Canonicalization
=======================
Duplication of content often occurs when sites render the same content
on different URLs. The software of popular Content Management
Systems (CMS) like WordPress and Drupal tends to create duplicated
content by way of categorisation (archive, pages, categories, tags) or
formatting (web, printable, feed).
E-commerce sites may incur in duplication by indexing URLs with
multiple parameters (session IDs, shopping cart and checkout
parameters, etc).
Offer a unique version of your content: avoid content showing up in
more than one page, and if it has to, use the metatag canonical to help
Google find with is the important page for your business.
Declare the link attributes rel="next" and rel="prev" and rel="canonical"
to help Google understand what are the pages that really matter to you
and the user xiii.
<link rel="canonical"
href="http://www.example.com/article?story=abc&page=2"/>
<link rel="prev"
href="http://www.example.com/article?story=abc&page=1&sess
ionid=123" />
<link rel="next"
href="http://www.example.com/article?story=abc&page=3&sess
ionid=123" />
You can also help Google handle duplicate content across domains xiv
with the canonical tag. This tag is just an indication to Google, and it
should be used if there is not any other, more efficient, ways to tackle
duplication.
+++++++++++++++++++++++++
==============================

Anchor text

======================
Use short, descriptive and relevant anchor texts on hyperlinks. Try not to
stuff the anchor text with keywords. Anchor text used to be very
important for ranking purposes but context and other techniques like
Latent Dirichlet allocation (LDA) xv are gaining in weight.
SEO Rao Sohaib
36 Paradigma Tecnolgico – Rao Sohaib
If you link more than once to the same destination URL on the same
webpage, only the anchor text of the first link is the one picked up by
Google for correlation statistics.
+++++++++++++++++++
==============================

Structured Data

===================
Consider marking up individual pieces of text that can be itemised such
as details of people, events, places, resumes for job boards, listings of
classifieds, etc. Tagging microdata helps Google identify what is your
content is about at very deep minutiae. There are 3 standards of
structured data markup that you can follow with similar effects:
x Microdata Schema.org - a well-organized and documented
HTML5 microdata
x Microformats http://microformats.org/ - simpler and earlier
standard
x RDFa http://rdfa.info/
The tangible SEO benefits of tagging structured data are:
- you can get larger snippets on Google SERPs if you use smart
microformats. Eg. hListing for classifieds sites, hResume for job
boards, hReview
- you can rank higher on Google if you are organising an event with
hCalendar
E.g. case of use of the Open Graph protocol to enable webpages to
become a rich object in a social graph
<meta name="og:title" content="Big Data Spain 2012
conference - Madrid Nov 16th"/>
<meta name="og.summary" content="Timetable of the schedule
of the sessions of the Big Data Spain 2012 conference"/>
<meta name="og:type" content="website"/>
<meta name="og:url"
content="http://www.bigdataspain.org/en-
2012/program.php"/>
<meta name="og:site_name" content="Big Data Spain"/>
<meta name="og:latitude" content="40.452537"/>
<meta name="og:longitude" content="-3.726412"/>
SEO Rao Sohaib
37 Paradigma Tecnolgico – Rao Sohaib
<meta name="og:street-address" content="ETSI
Telecomunicaciones, Avenida Complutense nº 30, Ciudad
Universitaria"/>
<meta name="og:locality" content="Madrid"/>
<meta name="og:region" content="Madrid"/>
<meta name="og:postal-code" content="28040"/>
<meta name="og:country-name" content="Spain"/>
<meta property="og:locale" content="en_GB"/>
<meta property="og:locale:alternate" content="es_ES"/>
Figure 7 Snapshot of the key information that users would see right in Google’s SERPs thanks
to tagging content with the standards of hCalendar
+_++++++++++++++++++++
=================================

Authorship

=========================
=========================
Google requests the help from webmasters to match content with its
authorxvi. This can be done by marking up links with rel=”me” and
rel=”author” bi-directionally between the authors’ profile page on
Google+ and the webpage hosting the authors’ content.
Google allows rel=”author” markup through a <link> element in the
head:
<link rel="author"
href="http://plus.google.com/117486480899235875959/"/>
Identified authors get an avatar picture displayed on Google SERPs,
which is great for higher CTR and possibly rankings for some queries:
SEO Rao Sohaib
38 Paradigma Tecnolgico – Rao Sohaib
Figure 8 screenshot of the profile of the page of Google+ of an identified author
++++++++++++++++++++++++
===================================

Robots.txt protocol

========================
The robots.txt standard is quite popular among webmasters. It is only an
advisory protocol to allow or disallow pages or sections to Googlebot.
Google complies with it for but it may still fetch the disallowed content.
Depending on a number of circumstances it may display the URLs and
titles of disallowed content on its SERPs. The difference between the
results of disallowed pages on Google SERPs and allowed ones is that
snippets are missing on the former.
If you site architecture is lean and flat, you might not need to make an
intensive use of robots.txt. There are of course exceptions of large,
complex or legacy websites. E.g. enter this on your bash terminal
$ w3m -dump "http://www.bbc.co.uk/robots.txt" | wc –l
User-agent: Googlebot
Disallow: /iplayer/episode/*?from=r*
Disallow: /iplayer/cy/episode/*?from=r*
Disallow: /iplayer/gd/episode/*?from=r*
Sitemap: http://www.bbc.co.uk/news_sitemap.xml
Sitemap: http://www.bbc.co.uk/video_sitemap.xml
Sitemap: http://www.bbc.co.uk/sitemap.xml
Disallow: /_programmes
Disallow: /606/
Disallow: /apps/cbbc
Disallow: /apps/flash
[…]
At the time of writing this, the output of the entire robots.txt was
unusually long even for a website as large of the BBC: a count of 362
SEO Rao Sohaib
39 Paradigma Tecnolgico – Rao Sohaib
lines! Those many lines are probably the result of legacy directories. You
have a chance at the start of each new project to have a simple
robots.txt with a proper IA.
You can use regular expressions but test the syntax thoroughly on Google
Webmasters Tools first if you do not want to inadvertently disallow or
allow entire sections or website.
+++++++++++++++++++++++++
=====================================

Monitor your site for hacked content

==========================

xvii
Keep an eye on injected, added or hidden content and malware scripts
and remove tjhe as soon as it appears. Google Webmasters Tools
includes a section “Malware” to help deal with situations.
++++++
====================================

HTML, JavaScript, AJAX and CSS

=========================

Code for speed

============
As a general rule, you want to minimize the traffic between your server
and any client browser. Analyse your webpages and diagnose their
performance as you code.
Use third party tools like www.webpagetest.org or YSlow. YSlow's web
page analysis is based on 23 rules that affect web page performance and
are testable:
x Minimize HTTP Requests
x Use a Content Delivery Network
x Avoid empty src or href
x Add an expire or a cache-control Header
x Add gzip components
x Put stylesheets at the top
x Put scripts at the Bottom
x Avoid CSS expressions
x Make JavaScript and CSS external
x Reduce DNS lookups
x Minify JavaScript and CSS
x Avoid redirects
SEO Rao Sohaib
40 Paradigma Tecnolgico – Rao Sohaib

x Make AJAX cacheable
x Use GET for AJAX requests
x Reduce the number of DOM Elements
x Reduce cookie size
x Use cookie-free domains
x Do not scale images in HTML
Some programming languages allow you to minify HTML too. E.g. Twig is
a template engine for PHP that features spaceless capabilities.
++++++++++++++++++++++++++++++++++++++++++++
==========================================

Debug for crawlers

========================
Avoid crawling pitfalls in forms and redirect chains and loops, e.g. test
thoroughly regular expressions in JavaScript.
Dynamic URLs are OK but avoid session IDs and tracking arguments on
URL paths for webpages that do not require authentication because they
interfere with how crawlers access websites and de-duplicate content.
+++++++++++++++++++++
=================================

Avoid cloaking

=====================
Cloakingxviii consists in serving different content to human users and to
search engines. Technically it starts by identifying user agents by the IP
address and then serving up different content on the server side.
Legitimate coding tactics might inadvertently result in cloaking.
Examples of cloaking:
x Serving text to search engines while showing images or Flash to
users for the same page or section
x Inserting text or keywords into a page only when the User-agent
requesting the page is a search engine, not a human visitor
+++++++++++++++++++++++++++++++
==============================

Make AJAX content crawlable

===============================
The asynchronous capabilities of AJAX allow to update part of the
content of a webpage without having to reload all of it client-side. Since
Google does not interpret JavaScript to get content on the fly the way a
browser can, i.e. Google misses all the dynamic content loaded via AJAX.
SEO Rao Sohaib
41 Paradigma Tecnolgico – Rao Sohaib
Google are now asking webmasters to make their AJAX-based websites
crawl-able xix. As we will see below, their requirements are a tall order for
most webmasters – you need to invest some significant amount of
resources to deal with them.
Google have a method to allow for dynamically created content to be
visible to crawlersxx. The method is convoluted and intricate. Its number
one requirement is to present AJAX URLs, that is, URLs containing a #!
hash fragment. This requirement is contentious issue because
exclamation marks happen to be illegal in HTML, XHTML, and XML URL
identifiers and they make pages inaccessible to non-JavaScript-enabled
browsers.
If the URLs of your AJAX-based webpages are
www.domain.com/ajax-based.html#key=value
you want to turn them to
www.domain.com/ajax-based.html#!key=value
so the crawler can modify the URL to temporarily be
www.domain.com/ajaxbased.html?_escaped_fragment_=key=value
The rest of the requirements are based on what Google calls the HTML
snapshot mechanism.
SEO Rao Sohaib
42 Paradigma Tecnolgico – Rao Sohaib
Figure 9 Google’s HTML snapshot method to duplicate AJAX dynamic webpages to botreadable content
Since many of your URLs might be just rendered as static HTML whose
paths don’t contain hash-identified fragments, you will have to include a
new special meta tag in the head:
<meta name="fragment" content="!">
++++++++++++++++++++++++++++++++++++++++++++
==========================================

<noscript> tag for content on JavaScript

=========================================
Use a <noscript> tag to ensure that the content contained in JavaScript
can be read by Google. The text has to be exactly the same as what’s
contained in the JavaScript.
Using <noscript> this way is best practice although Google will not pass
PageRank on pages linked from within the <noscript> tag or treat
<noscript> content with the same value as the rest of normal text onpage.
<script type="text/javascript">
document.write("Out of sight, out of mind")
</script>
<noscript>Out of sight, out of mind</noscript>
SEO Rao Sohaib
43 Paradigma Tecnolgico – Rao Sohaib
Disable JavaScript on your browser to check the result.
+++++++++++++++++++++++++++
===================================

Avoid frames and Flash

===========================
Search engines cannot read Flash and they often skip frames. Images,
videos are challenging at best for search engines. Google has trouble
dealing with dynamically generated DOMs whose URLs contain session
IDs in the path.
Try to use readable text rendered on HTML as your preferred format. You
can also get pdf, doc and txt files indexed on Google if you link to them.
Use a text browser like w3m or lynx to get a closer picture to what search
engines “get” your content in the sequential order they crawl your
webpages.
E.g. install w3m, a text-based browser that cannot handle JavaScript. You
can find at Sourceforge. Enter this on your bash terminal
$ w3m -dump "http://www.ft.com/" | less
Inspect the output and compare it with the output of the parsed HTML
on your browser. Is the text sorted as you can see it on a browser?
+++++++++++++++++++++++++++++++++++++
========================================

Avoid using CSS to hide text

===============================
CSS Image Replacement places some text off-screen when calling a logo
image or similar from a CSS stylesheet. E.g. a CSS class with the text
moved way off any screen size using the -9999px hack.
h1.mylogo {
width: 200px; height: 60px;
background: url(/images/logo.jpg);
text-indent: -dpx;
}
Invoking the class “hides” any piece of text, in the example the one
wrapped by h1:
SEO Rao Sohaib
44 Paradigma Tecnolgico – Rao Sohaib
<h1 class="mylogo">The name of my company</h1>
There are legitimate reasons to use the technique, accessibility being one
of them e.g. text to speech and browsers with images or JavaScript
disabled for instance. The use of the alt label may be more appropriate in
those instances.
There are other shortcuts to hide text, such as matching the colour of the
font and the background or setting the font size to zero. Google learned
to deal with them. Specifically CSS image replacement is a contentious
issue when it comes to SEO. Check with an expert before you implement
it.
=====================================

Generate sitemaps

===========================
Help Google find your content when you launch a new website or when
you implement significant changes to it by supplying comprehensive and
updated HTML and XML sitemaps. Set up and follow a simple policy to let
Google know when you renew your content.
HTML sitemaps are best for the SEO of small websites. They are
impractical for large websites: users and bots are not expected to go
through long lists of backlinks on-page. XML sitemaps are effective for
large inventories of content that rotates regularly.
+++++++++++++
==============

HTML sitemaps

=================
HTML sitemaps are more important for users than for SEO. An HTML
sitemap should link only to the most relevant pages of a website. It is
tempting to link to all your inventory of documents from a HTML
sitemap. This usually results in pyramidal architectures.
SEO Rao Sohaib
45 Paradigma Tecnolgico – Rao Sohaib
Figure 10 diagram of pyramidal structure of a HTML sitemap listing links to deer pages. Each
webpage contains 3 links. The results in 39 pages linked from the HTML home page.
This practice is detrimental to the user experience and ineffective with
search engines.
E.g. Indeed.com links to all their individual profile pages from a wide and
deep tree of linked pages
Home of the HTML sitemap
http://www.indeed.com/resumes/directory/
Top category page
http://www.indeed.com/resumes/directory/1
Sub-category page
http://www.indeed.com/resumes/directory/1-1
Detail page
http://www.indeed.com/r/2e24d4a2aba4f0f1
The category and sub-category page are meaningless to end-users. In
order to find out how many of these pages are indexed on Google you
may use the advanced operator “site:”
http://www.google.com/search?q=site%3Aindeed.com%2Fdirecto
ry
Indeed’s scheme is linking to around 85,000 pages. Most of those pages
are just listings of hyperlinks with terrible, if any, usability. In addition to
that, it does not work in terms of SEO because Google counts about
SEO Rao Sohaib
46 Paradigma Tecnolgico – Rao Sohaib
1,980 indexed pages or 2.3% of all the ones linked to from the HTML
sitemap.
+++++++
=============

XML sitemaps

==============
XML sitemaps are a good indicator of the inventory of the content that
the webmaster intends to promote on Google.
Upload specific XML sitemaps by categories or types of pages of large
websites xxi and for imagesxxii. Images may drive in sizeable organic traffic
from the Google Images index if properly tagged.
<?xml version="1.0" encoding="UTF-8"?>
<urlset
xmlns="http://www.sitemaps.org/schemas/sitemap/0.9"
xmlns:image="http://www.google.com/schemas/sitemapimage/1.1">
<url>
<loc>http://example.com/sample.html</loc>
<image:image>
<image:loc>http://example.com/image.jpg</image:loc>
</image:image>
<image:image>
<image:loc>http://example.com/photo.jpg</image:loc>
</image:image>
</url>
</urlset>
If you have a mobile version of your site that you want to be specifically
indexed on Google, supply a mobile sitemap
<?xml version="1.0" encoding="UTF-8" ?>
<urlset
xmlns="http://www.sitemaps.org/schemas/sitemap/0.9"
xmlns:mobile="http://www.google.com/schemas/sitemapmobile/1.0">
<url>
<loc>http://mobile.example.com/article100.html</loc>
<mobile:mobile/>
</url>

</urlset>

SEO Rao Sohaib
47 Paradigma Tecnolgico – Rao Sohaib
Update your sitemaps as often as you add new pages of large websites to
help Google find them.
++++++++++++++++++++++
===========================================

If-Modified-Since HTTP header

===============================
Make sure your web server supports the If-Modified-Since HTTP header.
This feature allows your web server to tell Google whether your content
has changed since the last crawl.

======================================

Set the crawling rate of Googlebot

==============================
If your server/s is/are too loaded or if you are releasing lots of new
documents and pages at launch, you want to have a smooth take up on
Google index.
You can set the speed of the crawler of Google from a minimum of 0.002
to a maximum of 2 requests per second and from a pace of 500 to 0.5
seconds between requests – that is a factor of 1,000! In other words, you
have three orders of magnitude of gearing of the crawling speed and
pace of the Googlebot at your server!
Figure 11 Screenshot: menu Settings of Crawl rate of the section of Configuration on the
interface of Google Webmasters Tools
SEO Rao Sohaib
48 Paradigma Tecnolgico – Rao Sohaib
+++++++++++++++++++++++++++++++++++++++
=============================

SEO for established websites

============================
The same way that we would only trust a licenced doctor to diagnose a
health issue, we strongly advise to get professional help when auditing
the SEO potential and status of a given website. Interpreting and
contextualizing SEO information takes a lot of experience and
specialization.
SEO tries to turn the available resources of a website (content and
PageRank) into output (traffic, conversions and engagement). The output
should be in direct proportion to the input over time.
IN OUT
- Content: quantity, quality and
freshness
- PageRank xxiii
- Traffic levels
- Conversions and Engagement
Google measures more than 200 signals and updates their algorithms
several times per week on its ranking software. Even if we knew which
ones they are, it would not possible to measure all of them, reverseengineer Google’s algorithms and last but not least, re-build their
training data in order to run models of their rankings.
Auditing the degree and potential of the SEO of websites can fortunately
be economical for the trained eye. You want to carry out highly accurate
reports with the minimum precision.
You can get a faithful snapshot of the SEO of a website with few
measurements: backlinks, target keywords, content inventory and site
architecture.
Variable Measurement
Backlinks Quantity, quality, growth rate of backlinks, Google PageRank xxiv
Targeted
keywords Monthly searches and competition
Content
inventory
Quality and quantity of content, frequency, duplication, keyword
stuffing, available pages, pages indexed by Google
Site
architecture Organization of the website, URL structure, internal links
SEO Rao Sohaib
49 Paradigma Tecnolgico – Rao Sohaib
+++++++++++++++++++++++++++++

Off-page SEO

=====================

Backlinks

=================
Backlinks are part of off-page SEO. We cover them here for their
considerable weighting in the ranking software of search engines. There
are quick ways to assess the importance of backlinks for any given
website. It is highly recommended that you work first hand with a
professional SEO if or when considering the improvement or increase in
number of your backlinks.
++++++++++++++++
===========================

Quantity of backlinks

========================
Popular websites tend to get many backlinks on an ongoing basis.
E.g. Wired.com is a technology news website with a reputation for
publishing original and high quality content. Ahrefs.com is a tool for
checking backlinks and the social signals of individual websites.
Figure 12 screenshot of a report by ahrefs.com for the domain wired.com with a graph
plotting the evolution of the count of backlinks over time.
SEO Rao Sohaib
50 Paradigma Tecnolgico – Rao Sohaib
In this example, the count of backlinks of 18 Million is a remarkably high
number from a high number of referring domains: 351,000
You can also compare side by side competing websites with
OpenSiteExplorer.org
Figure 13 screenshot of Opensiteexplorer.org comparing the linking profiles of the websites
of popular text editors: sublimetext.org, notepad-plus-plus.org, jedit.org and
bluefish.openoffice.nl. Notepad++ is the best websites for most metrics of quantity and
quality of backlinks
The competition for backlinks depends on categories, locales and
languages. English-driven websites may potentially enjoy more backlinks
from many more websites than other languages on the WWW although
competition is theoretically more intense for the same reason.
SEO Rao Sohaib
51 Paradigma Tecnolgico – Rao Sohaib
These are the estimates of the percentages of Web sites using various
content languages as of 2014xxv:
Rank Language Percentage
1 English 56.2%
2 Russian 5.9%
3 German 5.9%
4 Japanese 4.9%
5 Spanish 4.5%

6 French 3.9%
7 Chinese 3.2%
Some online backlinks checker tools do a great job at helping to
understand just how many links any website gets. Services like
MajesticSEO, Ahrefs and SEOMoz crawl the WWW at a rate of billions of
URLs a day. The size of their databases is of hundreds of billions of
Unique URLs, which makes them effective for short term information and
alerts.
++++++++++++++++++++++++++

Quality of backlinks

====================
Relevant links on sites with authority and good reputation are beneficial
for SEO. The anchor text and the context of links are also signals used by
Google to determine the relevancy of backlinks.
Paid links do not comply with the guidelines of Google web search.
Finding paid links to third party sites takes often some research with a
high degree of manual inspection and knowledge of a vertical and a local
market. Knowing that a webmaster is buying sponsored links and that
those links do not have the nofollow attribute is a sign of poor off-page
SEO practice.
If you are in charge of the SEO of an established website, watch the
nature of the backlinks you are obtaining. Stay away from low quality,
spam sites linking to you. Paid links or advertorials are legitimate as long
as the publishers flag them as such so that Google understands that they
are sponsored links.
E.g. Interflora operates an e-commerce website selling the delivery of
flowers. Its UK operations incurred in the purchasing of mass
SEO Rao Sohaib
52 Paradigma Tecnolgico – Rao Sohaib
advertorials. Google penalised Interflora.co.uk manually in early 2013.
The results on Google of Interflora dropped from 1st and 2nd positions to
the 49th place overnight for many converting keywords like “St.
Valentine’s Day flowers” in the UK.
Worse come to worse, if you suspect that your site's ranking is being
harmed by low-quality links, you can ask Google not to take them into
account when assessing your site.
- Decommission all the unnatural links pointing to your site
- Report them via the disavow links tool on Google Webmasters
Tools
++++++++++++
==========================

Growth rate of backlinks

===========================

Natural links build up over time. Google is sensitive to sudden bursts of
new links. It learned to interpret the sudden peaks as spam behaviour.
Inversely, you want to keep a natural, steady rate of new links without
dramatic troughs or peaks.
SEO Rao Sohaib
53 Paradigma Tecnolgico – Rao Sohaib
+++++++++++++++++++++
=======================

Content inventory

========================

Internal duplication

==========================
Avoid duplication of your content among your own pages. A first step to
diagnose duplication consists in estimating the percentage of matched
words among pages of the same website. Siteliner is an online tool that
counts the co-occurrence of the same words on pages of the same
website.
E.g. the tool highlights the words on a page that are matched with other
pages, e.g. a section of The Guardian
http://www.siteliner.com/www.guardian.co.uk/technology?siteliner=du
plicate
39 matches found covering 974 of 1,658 words – 58% of the page
++++++++++++++++++++++++++++++
===========================

Plagiarism

==================
Use an online plagiarism detection service like Copyscape to find out
copies of any given webpage.
E.g. the text of the patent of Google of the PageRank, when checked on
the online tool, lists the other pages on the Internet with duplicate
snippets of text.
http://copyscape.com/?q=google.com/patents/US6285999
If your content is systematically copied and published somewhere else
without your authorisation or citing you as the source, and if this scraped
content outranks you on Google, you may consider reporting it on their
Scraper Reportxxvi.
++++++++++++++++++++++++++++
===========================

Count of indexed pages

=======================
The search operator “site:” can be used to get a rough estimate of the
number of pages indexed by Google of a given website
E.g. comparison of websites competing for the same audience, for
instance, B2C telecom operators with data provided by the count of
indexed pages by Google for queries with the “site” operator:
SEO Rao Sohaib
54 Paradigma Tecnolgico – Rao Sohaib
http://www.google.es/search?q=site%3Awww.movistar.es
Subdomain # pages indexed by Google
www.vodafone.es 107,000
www.movistar.es 24,500
www.orange.es 6,680
www.yoigo.com 2,790
www.simyo.es 369
The differences of several orders of magnitude in the count indexed
pages on Google by different operators may point to vast differences in
their SEO or in their content marketing strategy.
++++++++++++++
===================

HTTP Status Codes

====================

SEOs hate webpages with 404 HTTP standard response code from the
server. The number 404 denotes a “Not Found error message”, that is,
the server has nothing to return for its URL.
Reduce the generation of 404 errors to an absolute minimum. When 404
are inevitable, provide a useful 404 page that let users and bots find the
next best content that you can offer.
Even in an ideal world you should have a “404 page” ready to help the
user find his/her way around your site. It is important that your 404 page
does return a 404 HTTP response code, instead of a 200 one. Web
developers tend to oversee this since the server does have something to
respond with.
SEO Rao Sohaib
55 Paradigma Tecnolgico – Rao Sohaib
Figure 14 Source: http://css-tricks.com/
Ideally you should the 410 HTTP status code in some cases for obsolete
pages. The code indicates that search engines should remove the
document from their indices because is no longer available and will not
be available again.
Unfortunately the URL whose return code is 410 will still show up on the
reports of Site Errors on Google Webmasters Tools.
++++++++++
================================

Server-side redirects

=====================
The HTTP response status code 301 status code means that a page has
permanently moved to a new location. The 302 ones means temporary
redirection. SEOs usually prefer to use the 301 code to the 302 one
because the first passes some of the link value, but the second does not.
SEOs usually enter rules which webpages redirect to others on the
server's .htaccess file for websites hosted on Apache servers. Mind the
regular expressions that your write on the file and test extensively before
going to production.
SEO Rao Sohaib
56 Paradigma Tecnolgico – Rao Sohaib
+++++++++++++++++++++
=======================================================

Migration from older versions or consolidated properties

=================================================
When a medium-sized website launches a new version, almost inevitable
some or many URLs will be obsolete and return error codes. This is also
the case when domains or subdomains move in under the main www.
Subdomain.
Tackle the issue by fixing broken links on your own site, updating your
sitemaps and typically, adding a 301 redirect to get users to the “real”
page.
An interesting feature of Google Webmasters tools is that the errors are
ranked by priority. Google determine the priority with factors such as the
presence of the URL in a sitemap, number of backlinks and internal links
and, interestingly, whether the URL has gotten any traffic recently from
searchxxvii.
+++++++++++++++++++++++++++++++++++++
==================================

Manage the rotation of content

===================================
Decide what to do with pages or sections or pages when the content
becomes obsolete, redundant or when it is no longer available. Write a
content management protocol to deal with obsolete content. If your
server supports .htaccess (Apache does), customise the error responses
for all the foreseeable situations of missing webpages, obsolete or
deprecated ones.
Even when you try to cater for all contingencies, you still may need to
remove URLs on Google. While cumbersome and impractical for very
large sites, in some cases, submitting individual URLs from Google’s index
may save your day.
Learn the differences between 301 and 302 redirects and use preferably
301 codes to redirect webpages on the same domain or different
domains /obsolete-page.html to /new-page.html
Google Webmasters Tools lets you remove an entire site off Google or
redirect it to a new domain.
SEO Rao Sohaib
57 Paradigma Tecnolgico – Rao Sohaib
+++++++++++++++++++++++++++++++
============================

Site Architecture

=============================
This section will be illustrated step by step with an example. We analyse
the website of an event, www.bigdataspain.org
++++++++++++++++++++
=======================

First step – Crawl a website

=============================
Download Xenu’s Link Sleuth (for Windows). It is highly recommended
that you download the latest 1.3.9 or higher versions.
Run a crawl (menu File > “Check URL”) for the URL www.bigdataspain.org
Once the crawl is finish, export it (Menu File > Export Page Map to a TAB
separated file)
At the window “Export Pagemap”, save the file with the name crawl.txt
++++++++++++++++++++++++++++++++
=========================================================

Second step - Filter the pages with internal links only

============================================
Inspect the fields of the exported file with bash commands or Perl on a
bash terminal
$ head crawl.txt
Remove the redundant text and most of the internal links to accessory
files and external links
$ cut -f1,2 crawl.txt | sed -e
's/http\:\/\/www\.{domain}\.{tld}//g' -e 's/\t/,/g' |grep
-v "\.jpg\|http\:|\.css\|\.js" > filtered.csv
$ head -5 filtered.csv
This results in this output with the comma-separated fields where a
forward slash only represents the homepage. E.g. the page program.php
links up to the home.
OriginPage,LinkToPage
/2012/conference/business-big-data/jon-bruner,/
/2012/program.php,/
/2012/speakers.php,/
SEO Rao Sohaib
58 Paradigma Tecnolgico – Rao Sohaib
/en-2012/speakers.php,/
+++++++++++++++++++++++++++++++++
===============================================

Third and last step - Visualize the network and analyze it

=================================
Gephixxviii is a network analysis and visualization software application.
Download it from Gephi.org and install it.
Menu “New Project” > File “Open” filtered.csv > Window “Import
report”. Options “Graph Type” select the “Directed” value from the
menu.
# of Nodes 55
# of Edges 416
Choose a layout ForceAtlas, for instance, from the Window > Layout
menu and click on “Run”
Iteratively adjust the settings of the algorithm to plot the heavily linked
nodes in a visually effective way.
Figure 15 representing the graph of nodes and edges of www.bigdataspain.org. The size and
colour of some nodes have been adjusting for illustration purposes only.
SEO Rao Sohaib
59 Paradigma Tecnolgico – Rao Sohaib
The node program.php is more important than the venue.php one. This relative importance
on the graph is well aligned with the goals of the website. The graph allows detecting at first
sight many issues that might otherwise escape the attention of the webmaster.
SEO Rao Sohaib
60 Paradigma Tecnolgico – Rao Sohaib
Note statistics such as average path length (2.559) in the example if you
are evolving the website over time so you can compare them.
Examine the graph and find topological information such as:
x Colour of nodes and edges represent different languages on the
same website. Red and orange for the English version and grey for
the Spanish version
x The number of inbound links that they get from the rest of the
pages of the site should be proportional to the value of the node
for the site
Some interesting metrics are derived from the mathematics of graphs.
Menu Window Statistics
Graph Density Directed, (simulated) 0.143
PageRank Directed distributionxxix: how often a user following links will
non-randomly reach the node “page”.
Figure 16 PageRank distribution for a small graph with the parameters Epsilon = 0.001 and
Probability = 0.85 Algorithm Sergey Brin, Lawrence Page, The Anatomy of a Large-Scale
SEO Rao Sohaib
61 Paradigma Tecnolgico – Rao Sohaib
Hypertextual Web Search Engine, in Proceedings of the seventh International Conference on
the World Wide Web (WWW1998):107-117
The internal PageRank of a given graph is not to be confused with Google
Toolbar’s PageRank. The PageRank of a local network is a metric of
internal authority of each node in a network. In other words, you may
assess with webpages enjoy the highest PageRank and thus deduce how
important is that webpage for the webmaster or the SEO of that website.
Repeat this exercise of analysis of site architectures with other websites
and compare these values over time.
+++++++++++++++++++++++++
===========================

Watch the health of your site

===========================
If you are proficient in Perl, a quick script and some bash magic can do
wonders to analyse your access logs every now and then. For deeper
analysis you may consider Apache Log Viewer or tools to look for issues
with the Googlebot. Fluentd is an open source alternative to Splunk for
websites with heavy traffic loads consists in combining ElasticSearch, and
Kibana.
+++++++++++++++++
===================================

Crawling by Google

=========================
Use Google Webmasters Tools before, during your coding project and
after launch. Google Webmasters Tools helps to find out about
Googlebot’s crawling stats and identify crawl errors. The section “Index
Status” is a great indication of just how deep and wide the Googlebots
crawls your site over time.
E.g. If Google is indexing everything that it crawls since you launched
your site you did a great job.
SEO Rao Sohaib
62 Paradigma Tecnolgico – Rao Sohaib
Figure 17 of Index Status of the section Health of Google Webmasters Tools. The take up by
Google of virtually all the content that it crawls on a new website is good news yet quite
unusual
++++++++++++
===============

Server logs

===============
Google used to include almost all the original search queries of its users
in the referrer. It started to omit the query for those sessions of logged-in
users, i.e. those using Google under the secured protocol https instead of
http. Depending on a number of factors, the percentage of visits on https
is 30-40% of the total. Nevertheless extracting search queries from the
logs remains a basic tool in the SEO trade.
- Setup your servers to archive logs automatically.
- Check regularly the IP range of Googlebot instances and the most
important search engines
- Report the frequency of crawling of the top pages, identify all the
redirects and all the error return codes
- Extract queries from the referrer. Google does not pass on the
queries of sessions of logged-in users using the securitised
protocol https. Varying over time and depending on the websites,
the percentage of missing queries on referrers from Google is
around 30% of the total
SEO Rao Sohaib
63 Paradigma Tecnolgico – Rao Sohaib
- You can count the frequency and rankings of the captured
keywords on Google on a per session basis. Use the statistics
package of the R project to understand the number of visits by
query xxx and by ranking of your webpages on Google SERPs.
++++++++++
============================================

Health check of indexed URLs

================================
Obsolete or deprecated pages result in 404 error response codes. Google
interprets these as a signal of poor quality of a given website. It is
important to keep these errors under control.
Put all the URLs in a txt file on your bash shell, e.g. urls.txt
http://www.website.com/j2ee-java-developer
http://www.website.com/java-developer-j2e3-spring-eclipse
http://www.website.com/software-consultant-java
http://www.website.com/technical-lead-java
http://www.website.com/engineer-full-stack-javascriptpython-java
Then check them with a simple shell script as long as they are only a few
hundred so you do not overload the server with your loop.
$ while read i; do echo $i && curl -s -w "%{http_code}\\n"
$i -o /dev/null; done < urls.txt
This is the output you should expect:
http://www.website.com/j2ee-java-developer
404
http://www.website.com/head-developer-j2e3-spring-eclipse
410
http://www.website.com/software-consultant-java
200
http://www.website.com/technical-lead-java
301
http://www.website.com/engineer-full-stack-javascriptpython-java
404
Obviously you are interested in the URLs that present server errors, i.e.
those with 404 for instance.
SEO Rao Sohaib
64 Paradigma Tecnolgico – Rao Sohaib
++++++++++++++++++++++++++++
==================================

Log file parsing

============================
LogFormat "%h %l %u %t \"%r\" %>s %b" common
%r "GET /apache_pb.gif HTTP/1.0" (\"%r\"). The request line from
the client is given in double quotes
%>s status code returned by the host
Extract individual queries and the positions of your pages on Google SERP
with a few lines of code on R xxxi
log <-
getURL("sftp://user:password@host:/path/to/apache/accesslo
g.log")
++++++++++++++++++++++++
================================

Block bots other than search engines

=======================
Block –or try to- bad bots and rippers that may weight down on your
server with .htaccess or via an algorithmic approach.
SEO Rao Sohaib
65 Paradigma Tecnolgico – Rao Sohaib
++++++++++++++++++++++++
=============================================

Tools and references

============================
The tools are as good as the questions you ask their data. We list a few of
them here for you to start getting a feeling of what to ask an SEO about.
Disclaimer: the author is not affiliated with any of the vendors whose
tools or names are provided below.
Measurement Tools
Backlinks ahrefs, OpenSiteExplorer, MajesticSEO
PageRank Google Toolbar (only available for Internet Explorer)
Frequency &
competition of
keywords
Adwords’ Keyword Planner
Rankings Google Webmasters Tools, Advanced Web Rankings
Content inventory Xenu (desktop app Windows), Screaming Frog (SaaS)
Duplicate content Copyscape, Siteliner
Pages indexed by
Google Operator “site:” in web searchxxxii
Site architecture Gephi
Server logs Apache Log Viewer, Splunk, R-project
Crawling Google Webmasters Tools, w3m, LWP, Live HTTP Headers
add-on for Firefox
Engagement R-project web metrics SaaS like Google Analytics
SEO Rao Sohaib
66 Paradigma Tecnolgico – Rao Sohaib
++++++++++++++++++++++++++++++++++
====================================

What now?

========================
Experiment with SEO as much as your business and your resources can
afford. Ask expert SEOs and read regularly blogs about search and IA.
+|
===========

Epilogue

=============
Was this RoCkHaCkErZz useful to you? If you like it, pass it on to your colleagues
and friends!
We wrote this RoCkHaCkErZz to help people but not everyone reads English, so
you can contribute with translating this handbook to other languages as
long as you let us know, provide attribution and generally abide by its
licensing terms.
For questions, reports of errors or suggestions or just to say hello by
dropping a message on Paradigma’s website, mine or via:
x Google+ Rao Sohaib
x Twitter @RubenMartinezS
x LinkedIn rubenm
x Quora Ruben Martinez
SEO Rao Sohaib
67 Paradigma Tecnolgico – Rao Sohaib
+++++++++++
===========================

Appendix – Google Updates

========================

Google updates are to marketing what Black Fridays are to the stock
market. Some SEO veterans still remember the Google “dances” which
basically made or break many businesses online.
More recently Google Caffeine changed the way Google deals with its
index. Google could, by the first time, update its index incrementally
when it was released in June 2010. “When a new page is crawled, Google
can update its index with the necessarily changes rather than rebuilding
the whole thing” - Eisar Lipkovitz, a senior director of engineering at
Google in 2010xxxiii.
Following the big leap in technological Google Panda, Venice, Penguin
and Hummingbird are other important updates that changed
You may find a historical review of Google updates at Moz’s updates
calendar. You can also check if your Google Analytics’ reported visits
match with known Google updates. You can give Barracuda, an online
marketing agency, access to your Google Analytics account so they can
plot the occurrence of updates.
+++++++++++++++++++++++++++++
================================

SERP volatility

=======================
Search engines like Google update their datasets and adjust their
algorithms more than 400 times every year. SEO professionals jump on
SEO Rao Sohaib
68 Paradigma Tecnolgico – Rao Sohaib
their seats when those updates change significantly the rankings of the
best converting queries.
Some tools helps understand the changes in positions on SERPs. One of
them is Serpmetrics.com. Their charts plot a volatility index. It is
composed by comparing the results on Google regularly. A scoring
system applies weighting to the top results and calculates the turbulence
over time.
Figure 18 Graph of the flux of changes on Google's SERPs over time. Some turbulence is
clearly identifiable before and after August 20th
SEO Rao Sohaib
69 Paradigma Tecnolgico – Rao Sohaib
++++++++++++++++++++++++++++++++
==============================================

Appendix – Target keywords

=====================================
Target keywords are those that are used in titles, URLs, anchor texts of
links and other elements of the HTML. Think of the mark up of targeted
keywords as important assets in your SEO. Many webmasters ignore
which targeted keywords they are really promoting on their websites.
The stage of keyword research is not strictly speaking Rockhackerzz.blogspot.com. We
mention it on this document because it is crucial for the success of your
mark-up.
The keywords that define your business and your target audience are
often overlooked at the beginning of web development projects.
E.g. look at the top 5 sites ranking for the query “Terms and Conditions”,
searched for on average 135,000 times in the United Kingdom monthly.
http://www.google.co.uk/#hl=en&output=search&q=terms+and+c
onditions
At the time of writing this document, Tesco.com, a grocery retailer and
O2.co.uk, a mobile network operator, rank 3rd and 4th respectively. Do
they really want to rank for those search terms? The query “Terms and
Conditions” has not transactional or commercial search intent among the
customers of Tesco or O2. The rankings of their websites are not helping
to sell more groceries or phone contracts.
Find if your audience is increasingly interested in your targeted keywords
or not. A way of finding our quickly is by checking the normalized
frequency of monthly searches by Google Trends.
E.g. Arduino and Raspberry Pi are open-source hardware devices.
Arduino became popular before Raspberry Pi was launched.
http://www.google.com/trends/explore#q=arduino%2C%20raspberry%20pi&cm
pt=q
SEO Rao Sohaib
70 Paradigma Tecnolgico – Rao Sohaib
Figure 19 The interest in the Raspberry Pi project (red line on the graph) has recently
overtaken the popularity of the Arduino project (red line) only to drop both in the last few
weeks.
Find out the competition for your targeted keywords and whether they
are head or long tail keywords.
E.g. in English and locale United Kingdom
flights Majorca
flights Menorca
flights Ibiza
flights Tenerife
flights Las Palmas
Google’s Keyword Planner helps you find the popularity of your
keywords in number of monthly searches and their relative “difficulty” to
rank for them.

Keyword Competition Global Monthly
Searches
Local Monthly
Searches
flights Majorca High 49,500 1,000
flights Menorca High 22,200 320
flights Ibiza Medium 74,000 2,400
flights Tenerife Medium 135,000 2,400
flights Las
Palmas
Medium 5,400 390
SEO Rao Sohaib
71 Paradigma Tecnolgico – Rao Sohaib
+_+++++++++++++++++++++++++++
=======================================================

Appendix – Domain names

==============================
The projects of new websites often require the registration of a new
domain name. While the topic is related to off-page SEO, we mention it
here because we usually are asked about it by developers.
Top level domains (TLDs) typically have the .com name for commercial
and .org for organizations. You can register a domain with your brand
name or including keywords.
Google has a reputation of being biased towards big and established
brands. Investing in new, niche or second tier branded domains is a
strategic decision that makes all the sense in terms of marketing. In
terms of SEO however it may take time to pay off even if Google picks up
the correlation between your brand and a category or niche of keywords.
Domain names with keywords on the other hand tend to rank high on
Google’s SERPs and definitely on Bing’s. Exact Match Domain (EMD)
names contain the highest converting keywords of the business.
E.g. Forextrading.com ranks top for the query “forex trading”. The owner,
Saxo Bank, opted to use an EMD for their business than to promote a
brand like Oanda.com do for the same segment of the market.
Google claims to reduce the low-quality “exact-match” domains in search
results but in many countries and languages their SERPs are still plagued
with them.
++++++++++++++++++++++++++
================================

Internationalization of domains

================================
Country-code top-level domains (ccTLDs) are indicated to target local
markets like domain.ca in Canada and domain.es in Spain if you can
afford the complexity in their administration.
In some cases, if you do not want or you cannot localize your domains
with ccTLDs, you can setup folders by language, e.g. both Zara and H&M
organize their local websites with folders
www.zara.com/jp/ and www.hm.com/jp/ both for Japan.
SEO Rao Sohaib
72 Paradigma Tecnolgico – Rao Sohaib
++++++++++++++++++++++++++++++++++++++++++++++
===============================================

Subdomains and subfolders

============================
Sub-domains let you organize your domains sensibly by markets, target
audience and products and services. As a rule of thumb, create subdomains for unrelated content.
E.g. the revenue of Suzuki Motor Corporation in Europe comes mainly
from 2 consumer categories: cars and motorcycles. It markets each
category with a different sub-domain: http://auto.suzuki.es/,
http://moto.suzuki.es/ respectively.
Sub-domains are usually ideal for horizontal scaling of categories of
products or services but they do not benefit from the PageRank of their
root domain. On the other hand, Google Webmasters Tools will not take
sub-domains as addresses when you want to move your site to a new
domain and notify Google.
SEO Rao Sohaib
73 Paradigma Tecnolgico – Rao Sohaib
+++++++++++++++
===============================================

Appendix - Google Analytics

=================================
Many developers have questions about Google Analytics - mostly about
how to implement it and where. The tool is a powerful SaaS suite that is
popular partly for being free for sites under 10M hits processed per
month, i.e. for the vast majority of websites.
Implement the tracker on your website before or at launch of a new
website- You need to get Google to verify your ownership or authority
over a website with a few steps:
1. On the Webmaster Tools Home page, click the Manage Site button
next to the site you want, and then click Verify this site.
2. In the Recommended method tab, follow the steps on your screen
3. Click Verify. Removing the code from your page can cause your
site to become unverified, and you will need to go through the
verification process again.
4. Copy and paste your tracking code on every page you want to
track
Alternatively, create a PHP file named "analyticstracking.php" with the
code above and include it on each PHP template page. Then, add the
following line to each template page immediately after the opening
<body> tag or use a common include or template to paste the code
above instead of manually adding it to every page:
<?php include_once("analyticstracking.php") ?>
Once you set up Google Analytics tracking, you want to track Traffic
Sources – Advanced Segments – Non-paid Search Traffic and Conversion
metrics among many other metrics and reports.
Take Google Analytics’ data with a pinch of salt for a number of reasons:
x asynchronous technologies have limitations of their own
x individual visits are not reported – all the data are aggregated
x visits under the secured protocol https are aggregated under the
“not provided” category for search queries, etc.
A reference guide of Rockhackerzz.blogspot.com
for software developers
Rao Sohaib
Rockhackerzz.blogspot.com
Rao Sohaib
SEO Rao Sohaib
74 Paradigma Tecnolgico – Rao Sohaib
The best way forward is to combine metrics from your server logs and
from Google Analytics or other metrics tools.
++++++++++++++++++++++
==================================

Engagement

================================
Your content needs to engage with your audience. Usage signals – like
clicks on tagged elements, conversions (goals), time per page,
pages/session, social signals, etc. are now part of SEO.
Detect losses in the navigation flow of your website by drilling Google
Analytics and correct them appropriately.
Figure 20 with a snapshot of the downstream interactions of users by geographic origin on
the jobs section of www.bigdataspain.org. Losses are represented by red flows pouring from
the green boxes with counts of visits.
SEO Rao Sohaib
75 Paradigma Tecnolgico – Rao Sohaib
+++++++++++++++++++++
================================

Split or A/B tests

=======================
Split tests are formidable tools to optimize conversion ratios. You can use
them for SEO also. You should test every component of your technical
SEO, from the mark-up of your webpages to the architecture of your
website. You may want to verify for instance the effect of the reasonable
surfer paradigm by testing different distributions of links on-page.
Just about every aspect of a website can be split-tested. Be prepared to
accept insignificant differences and counter-intuitive results. In order to
get significant differences you may have to try to test radically different
things.
SEO Rao Sohaib
76 Paradigma Tecnolgico – Rao Sohaib
++++++++++
=========================

Licensing

===================
This work is licensed under the Creative Commons AttributionNonCommercial-NoDerivs 3.0 Unported License xxxiv.
If you want to translate this work to other languages, the resulting work
is licensed under the same license with attributions to this copy and
version, the author and Paradigma.
SEO Rao Sohaib
77 Paradigma Tecnolgico – Rao Sohaib
+++++++++++++++++++
===================

References

==================

i Search Engine Optimization (SEO)
http://support.google.com/webmasters/bin/answer.py?hl=en&answer=35291
ii Google for Webmasters
http://developers.google.com/webmasters/googleforwebmasters/?hl=en
iii Dynamic IP addresses on cloud or load balancers
http://www.seroundtable.com/seo-amazon-elastic-load-balancing-14123.html
v PageRank sculpting
http://www.mattcutts.com/blog/pagerank-sculpting/
vi Method for node ranking in a linked database, invented by Page, Lawrence, US
Patent 6285999, Filing date: Jan 9, 1998
vii Ranking documents based on user behavior and/or feature data, Invented by
Jeffrey A. Dean, Corin Anderson and Alexis Battle, Assigned to Google Inc.
US Patent 7,716,225, Filing date: June 17, 2004
viii Shell one-liner for check for broken links
http://snipplr.com/view/18344/
ix Nofollow attribute
http://support.google.com/webmasters/bin/answer.py?hl=en&answer=96569
x Building Smartphone-Optimized Websites
https://developers.google.com/webmasters/smartphone-sites/details
xi URI Generic Syntax, Berners-Lee et al, Jan 2005
http://www.ietf.org/rfc/rfc3986.txt
xii URL mappings plug-in on Grails
http://grails.org/doc/2.2.x/ref/Plug-ins/URL%20mappings.html
xiii Pagination
http://support.google.com/webmasters/bin/answer.py?hl=en&answer=1663744
xiv Handling legitimate cross-domain content duplication
http://googlewebmastercentral.blogspot.com.es/2009/12/handling-legitimate-crossdomain.html
xv Latent Dirichlet allocation LDA
http://en.wikipedia.org/wiki/Latent_Dirichlet_allocation
SEO Rao Sohaib
78 Paradigma Tecnolgico – Rao Sohaib
xvi Authorship on Google Webmasters Tools
http://support.google.com/webmasters/bin/answer.py?hl=en&answer=1229920
xvii Hacked content
http://support.google.com/webmasters/bin/answer.py?hl=en&answer=2721435
xviii Cloaking
http://support.google.com/webmasters/bin/answer.py?hl=en&answer=66355
xx Making AJAX Applications Crawlable
https://developers.google.com/webmasters/ajax-crawling/
xxi Multiple XML Sitemaps: Increased Indexation and Traffic
http://www.seomoz.org/blog/multiple-xml-sitemaps-increased-indexation-andtraffic
xxii Image Sitemaps
http://support.google.com/webmasters/bin/answer.py?hl=en&answer=178636
xxiii Webpages with a higher PageRank are more likely to appear at the top of Google
search results
http://support.google.com/toolbar/bin/answer.py?hl=en&answer=79837
xxiv Crawl budget
http://www.stonetemple.com/articles/interview-matt-cutts-012510.shtml
xxv Usage of content languages for websites
http://w3techs.com/technologies/overview/content_language/all
xxvi Scraper report form
https://docs.google.com/a/google.com/forms/d/1Pw1KVOVRyr4a7ezj_6SHghnX1Y6b
p1SOVmy60QjkF0Y/viewform
xxvii Crawl errors reported by Google Webmasters Tools
http://googlewebmastercentral.blogspot.com.es/2012/03/crawl-errors-nextgeneration.html
xxviii Gephi

https://gephi.org/
xxix PageRank on Gephi http://wiki.gephi.org/index.php/PageRank
xxx Una aplicación SEO con R
http://www.datanalytics.com/blog/2013/01/10/una-aplicacion-seo-con-r/
SEO Rao Sohaib
79 Paradigma Tecnolgico – Rao Sohaib
xxxii Advanced Operators for Web Search
http://sites.google.com/site/gwebsearcheducation/advanced-operators
xxxiii Google Caffeine explained
http://www.theregister.co.uk/2010/09/09/google_caffeine_explained/
xxxiv http://creativecommons.org/licenses/by-nc-nd/3.0/

Rock Hackerzz

Tuesday, 28 July 2015

How To Do SEO free course 2015