ReadingSEO Master Deck - 1st August 2019

ReadingSEO
The only Reading SEO Meet-up
@ReadingSEO
Hosted by Blue Array / Sponsored by SISTRIX
Organisers: Matt Williamson (@MattWSEO) & Nicky Elson (@nickynoises)
Speakers: Mark Osborne (@MarkSEOsborne) / Ric Rodriquez (@RicRodriguez_UK) / Luke Carthy (@MrLukeCarthy)

6:40 pm: Take your seats
6:45 pm: Dicking Around With The DOM and
Chrome Dev Tools
7:05 pm: *Break*
7:15 pm: How Information Will Transform The
Digital Landscape
7:35 pm: *Break*
7:45 pm: How to properly optimise your internal
search for big SEO and revenue returns
8:05 pm: *Questions/Networking/Drinks/Food*
Schedule: Mark Osborne - SEO Manager
(Blue Array)
@MarkSEOsborne
Ric Rodriquez - SEO Consultant
(Yext)
@RicRodriguez_UK
Luke Carthy - Digital Manager -
(Mayflex)
@MrLukeCarthy

Understanding the problems
House Rules
- Please save your questions for the end of all the
talks.
- Must be questions - not comments
- Max 5 questions per speaker
(Feel free to chat to them afterwards)
- Please put your mobile on silent!
- Toilets & Fire Exit (Not the same place)
@ReadingSEO

Our Sponsor:
“We're currently offering a customer training day for free to trial users.
We have spaces available in Cambridge and London and more are
being added. Sign-up for the training, or for notifications, here.
https://www.sistrix.com/seminar/toolbox/ The free, no-commitment
trial is available here. https://www.sistrix.com/pricing/”
Steve Paine:

Reading SEO
Up First: Dicking Around With The DOM
& Chrome Dev Tools
Mark Osborne - SEO Manager
(Blue Array)
@MarkSEOsborne

Dicking Around With The DOM
& Chrome Dev Tools

Who The Hell Am I?
SEO Manager
At Blue Array
@MarkSEOsborne
Real creative twitter handle
right?

What Dodgy Looking People Do I Have To Work With?

So for any one that has
guessed the rather bad
joke in the title of this
talk

Don't worry to much as
there won't be any jokes
about these guys

So back to the
main reason we
all come to these
SEO events...

Some real useful new
SEO insights and free
pizza and beer :)
Thanks Sistrix

And start by
answering a
few key
questions

DOM actually stands for
Document Object Model

Simply it is a code
representation of what
you see on a page as it is
displayed in your
browser

And Is a middle
ground between your
page source code and
what you users see
when the page is fully
loaded

The DOM looks
different to the source
code. As additional
code and scripts are
injected into page after
it loads

Which means
there can be
problems on
your site that are
in the DOM
rather than in the
source code

Javascript makes
this a whole load
more fun, and
gives more issues
within the DOM

As dynamic scripts
load after and screw
things up.

What is chrome dev
tools for those that
don’t know?

Chrome DevTools is a set of
web developer tools built
directly into the Google
Chrome browser.
And can help you edit pages
on-the-fly and diagnose
problems

In short right click on
any page and click
inspect, and Chrome
dev tools opens up
with the DOM

By dicking around
with chrome dev
tools there is some
amazing stuff you
can do

So what can you
do with it I hear
you cry!

Well you can solve your one
device blindness
(this is a new phrase I'm trying to make stick so go
with it)

One device blindness is
- basically only
focusing on the device
you are working on
and not any other
device types

For most us that
is a desktop or a
laptop

In case you
weren’t aware
we live in a
mobile first
world now

In Chrome Dev Tools
you can easily
change your viewport
to different device
types

This is helpful when
checking parity between
mobile and desktop sites

We all know fast sites are
a big SEO win

With Chrome Dev
Tools you can see
how quickly Google is
loading your site in
the DOM

You can see this stage
by stage as its load all
by clicking the
network tab

As well as all the
individual elements
in the sequence as
they’re loading in
the DOM

This is wonderful but
how do I see how
my mobile site
speed is doing?

Adjust the network
setting to disable cache
and set speed to slow
3G then hit command R
to refresh the page

From here you
can see your sites
load time at its
worst on mobile

You can even find the fixes for
your slow load time in CDT
(i'm sick of keep saying chrome dev tools so got lazy and
referring to it as CDT now :) )

You can
get a full
lighthouse
speed
report in
CDT

Choose audit from the
top navigation and hit
run an you’ll get a full
detailed lighthouse
report within CDT

Playing
hide and
seek in
the DOM

There could be
hidden information
within the CSS so
It’s best to check
that it not being
missed

It’s real easy:
Open up CDT, hit
cmd F to open up
a search bar

From there enter one of the
two common CCS elements
that hide content:
display:none
visibility:hidden

CDT then highlights all of the
occurrences for you and you have
to find the important ones

Checking a
site all
secure and
locked down

In CDT there is a
whole security panel
which unsurprisingly
has all the site's
security information in

Shows you if you
have an SSL and if
there is any security
issues

The cool thing is if
you click links
through the site, in
the site panel in
CDT it show shows
you the security
status for individual
links

A few months
ago Google
updated their
developer guides
with new news

That you are now able
to override your
location with CDT to
any custom location
and see how a page
recacts

Open up command
menu with
cmd+shift+p

Search and select:
show sensors

Choose a
location from the
Geolocation
dropdown

If you can’t find a
location you want from
drop down hit manage
button

Finally add in a name
and the latitude and
longitude for your
custom location

Writing xpath without
the headache

When your scraping
a site for a certain
element

Open up CDT and find
the element you want to
scrape then...

Right click > Copy >
Copy xpath

The xpath is copied
to your clipboard
ready to go in any
scraping tool

Checking if a site
is HTTP2 or not

Yeah sure I
know there
is a bunch of
plugins that
do this as
well

But CDT gives you
more detail and is
only a few more
clicks

Open up network
tab in CDT & hit
refresh

Then right click and
add in protocol
column

Watch the
waterfall load
and see if the
sites is HTTP 2
or not

Why do we want to find
unused code?

Most sites are littered with old unused
JS & CSS that is slowing them down

The coverage
report shows you
where the unused
code is

Open CDT click
the three dots
in the top right
and select
coverage from
drop down

And ta-dah you get all
the unused code loaded

You can change the live
code on any page

And do fun things like
this...

Or more helpful things,
like seeing what a
change to the front end
of the site may look like

-The DOM is the bit between the
source code and what users see
-There is much more to CDT than you
think
-Mark’s puns are terrible

Get you thinking
hat on and start
dicking around
with the DOM and
CDT and see what
it can do

W
Thanks for listening, any questions?

5 Minute Break
Up Next: How Information Will Transform
The Digital Landscape
Ric Rodriguez - SEO Consultant
(Yext)
@RicRodriguez_UK

How Information Will
Transform The Search
Landscape
Ric Rodriguez, SEO Consultant

2019 - SEO Consultant, Yext
2018 - SEO Director, iProspect
2016 - SEO Director, Croud
2016 - SEO Strategist, Three Whiskey
2014 - SEO Analyst, iCrossing
@RicRodriguez_UK
#ReadingSEO

There is a massive paradigm shift
going on in the world of search

No click searches are increasing
Zero-Click
Searches
(49%)
Organic Clicks To
Non-Google Sites
(41%)
Any Click To
Google Sites
(6%)
Paid Clicks To Non-
Google Sites (4%)
Source: https://sparktoro.com/blog/how-much-of-googles-search-traffic-is-left-for-anyone-but-themselves/

The customer journey starts with a
question; consumers expect answers

The role of search is changing
1998 2018
Discovery Prioritisation Answers
# Of Websites In Existence
Source: https://www.internetlivestats.com/total-number-of-websites/
17m 29m 52m 86m
172m 207m
697m
969m
1.04bn
1.6bn

Accessibility
+
Relevance
+
Authority
Traditional search framework
Intent
Driven
Queries
Relevant
& Useful
Results

In the age of information, this
is a one dimensional strategy

Providing useful results to users
Source: https://www.searchenginejournal.com/how-google-search-ranking-works/307591/
Factor 2 Factor 3 Factor 4Factor 1 Factor 5
Website 1 4 6 31 9
Total
648
The Jason Barnard Theory

Website 2 7 8 22 4 896
Website 3 9 2 35 1 270
Website 4 1 3 96 2 324
Factor 2 Factor 3 Factor 4Factor 1 Factor 5
Website 1 4 6 31 9
Total
648

Website BidPosition
Website 2 8961
Website 1 6482
Website 4 3243
Website 3 2704

Website First Bid
Primary
Position
Website 2 8961
Website 1 6482
Website 4 3243
Website 3 2704
Second Bid
627
192
519
297
Final
Position
1
4
2
3

ImagePosition
5831
4682
2353
1724
Snippet
787
551
397
257
News
438
275
195
164
Web
627
519
297
192

This tells us about “how” an input is
assessed, but not “what” the input is

Search is more than your website
what is a horse
Direct
Answers
Websites
Meta
Search /
Platforms
Voice /
Chat
AR / VR Apps
Search
Results

Search is more than your website
what is a horse
Direct
Answers
Websites
Meta
Search /
Platforms
Voice /
Chat
AR / VR Apps
Search
Results
Knowledge + Artificial Intelligence

An algorithm is the set of specific
instructions that facilitates action

How a network works – your brain
Neurons (brain
cells) connect to
create a network
The connections
(“synapses”) talk via
chemical signals

When a response
is required a cell
is “activated”
It passes a positive
charge to other cells
who may do the same

Some cells need
a certain level of
charge to activate
When enough cells
activate together,
something happens

A network is only as strong as
the connections that make it up

A simple artificial neural network
(39)
1
2
3
4
5
6
7
Hidden Layer
Output Layer
Input Layer
The number “4”
Multilayer
Perceptron
Source: https://www.youtube.com/watch?v=aircAruvnKk – check out the channel!

1
2
3
4
5
6
7
Activation is either ”on” or “off”
Nodes
Edges
Multilayer
Perceptron
(39)

Weighted edges
help to determine
the outcome
1
2
3
4
5
6
7
Different layers establish different
components – e.g., “a loop”
(39)

1
2
3
4
5
6
7
Using examples, neural networks
can derive an “error score”
By adjusting the weighting, the model
can make more accurate predictions
(39) (39)

We can use networks to predict
intent and serve relevant results

A possible flow for learning intent
Query
- Result Type 1
- Result Type 1
- Result Type 2
- Result Type 2
- Result Type 3
- Result Type 3
Local / Map Pack
Ecommerce Pages
Article Content
1st Time Results

Query
- Result Type 1
- Result Type 2
- Result Type 3
Clicks* Bounces* B / C*
5201 256 5%
4329 297 7%
1063 491 46%
Average based on similar
“high confidence” pages 3956 324 8%
Intent Scoring
*Metrics chosen to illustrate the process, not representative.

Query
- Result Type 1
- Result Type 1
- Result Type 1
- Result Type 2
- Result Type 2
- Result Type 2
Local / Map Pack
Ecommerce Pages2nd Time Results
Sequence iterates until the search results yield a high confidence outcome

The information that provides intent
comes from the web as a whole

Understanding words on a page
Term
Frequency (TF)
Inverse Document
Frequency (IDF)/
(Disclaimer: Google uses more complex methods,
but this is useful to understanding the concept)

Term
Frequency (TF)
Inverse Document
Frequency (IDF)/
The number of times a word
appears in a piece of text

Term
Frequency (TF)
Inverse Document
Frequency (IDF)/
The importance of that word
in the overall document

A bank generally refers to the land
alongside a body of water. Different
structures are referred to as banks in
different fields of geography, as follows.
In limnology, a stream bank or river bank
is the terrain alongside the bed of a river,
creek or stream.
A bank is a financial institution that
accepts deposits from the public and
creates credit. Lending activities can be
performed directly or indirectly through
capital markets. Due to their importance
in the financial stability of a country,
banks are highly regulated.
Source: https://en.wikipedia.org/wiki/Bank. & https://en.wikipedia.org/wiki/Bank_(geography)

A` bank generally refers to the land
alongside a body of water. Different
structures are referred to as banks in
different fields of geography, as follows.
In limnology, a stream bank or river bank
is the terrain alongside the bed of a river,
creek or stream.
A bank is a financial institution that
accepts deposits from the public and
creates credit. Lending activities can be
performed directly or indirectly through
capital markets. Due to their importance
in the financial stability of a country,
banks are highly regulated.

Bank
body
water
structures
fields
geography
limnology
stream
river
terrain
creek
financial institution
deposits
public
credit.
lending
capital markets
importance
stability
country
regulated

Bank
River / Stream Financial Institution

Search engines seek to understand
the human world through the web

Entity-based understanding
Horse Jockey
What is a horse?

Horse
Jockey
What is a horse?

Horse
Jockey Carpenter
What is a horse?

Horse
Jockey Carpenter Gymnast
What is a horse?

What is a horse?
Horse
Horse Horse Vector
Domain

?
Horse
What is a horse?
?

Sports
Horse
What is a horse?
?

Horse
What is a horse?
Gymnastics
Sports

Horse
What is a horse?
Gymnastics Deriving context from search history
Sports

StablesRaces
Horse
Jockeys
Riding
The knowledge graph is a network of connected entities and topics

StablesRaces
Horse
Jockeys
Riding
? ?
? ? ? ?
?

StablesRaces
Horse
Jockeys
Riding
Riding
Jockey Stables
Race Race Horse Horse

Riding
Jockey Stables
Race Race Horse Horse
“Topic
Relevance”
“Entity
Relevance”
Using hierarchy to position us as a
subject matter expert (i.e., “trustworthy”)

Knowing the ”what”, let’s apply the
“how” and consider what this means

Building a search results page
What is a horse?

Building a search results page
#6#5#4#3#2#1
What is a horse?

As a result, the idea of “keyword
optimisation” may be unhelpful

Information
A new way to approach search
Platform + Presence
Intent
Driven
Queries
Intent
Driven
Answers

Platform
Create user-centric experiences
that inspire, engage and convert

Build a footprint that positions
you as a trusted online source
Presence

Information
Control the data that powers the
knowledge behind every result

This framework aligns to
the past pillars of search

Information
Search experience marketing
Relevance Platform + Presence
Content

Information
Accessibility Platform + Presence
Technical

Information
Authority Platform + Presence
Outreach

Information powers knowledge;
this is now the basis for search

10 Minute Break
Up Next: How to properly optimise your
internal search for big SEO and revenue
returns
(Mayflex)
@MrLukeCarthy

Inspirational Design192
How to optimise your
internal search
@MrLukeCarthy #ReadingSEO

Inspirational Design
We supply the folks that install cameras and
Internet stuff in to buildings
100 years old in 2017
Previously worked for…

So, why internal
search SEO?

4m+ search
URLs removed
(and counting)
30% increase in
MoM organic
traffic
=

800k+ search
URLs removed
20% increase in
YoY organic
traffic
=

4m+ search
URLs removed
18% increase in
YoY organic
traffic
=

Ok cool, anything else?

Exhausting crawl
quota
Poor crawl efficiency Unnecessary server
loads
Duplicate content Dilution of link equity Poor crawl efficiency
Poorly optimised
pages
Typically high
bounce rates
False positives on
search reporting
Can generate
countless new URLs
Keyword
cannibalisation
Impacts conversion /
UX
Slower to index new
content
Poor rankings
signals
Can be impossible
to maintain

Can generate
countless new
URLs
Keyword
cannibalisation
Impacts
conversion / UX

A real-world example…

115 filter URLs on
this page alone

115 filter URLs on
this page alone
65,000 possible filter
URLs in just ONE
parent category!

Super-juicy example
of bad internal search SEO

Case study

4.5m total pages
in Google’s index

85k products
in Google’s index

101k int. search URLs
in Google’s index

Let’s compare search to
categories…

Cats have
75% less
URLs than
search
Yields almost
3x more
revenue than
search

Let’s take a closer look…

2, 370 URLs for
the keyword
‘Smart TV’

But the landing
page & SERP is
poor...
Argos is the 1st
result…

The landing page
should be like
this…

Instead of like
this...

Case study

67K total pages
in Google’s index

16K products*
in Google’s index
(approx. wildcard search)

39k search URLs
in Google’s index

Keyword Cannibalisation!

Almost 50% of their site’s google
index is search URLs

Yet search URLs yield less than
5% sales and barely any traffic

Fixes duplicate
content
Reduces crawl
resources
Redistribute link
equity
Passes equity
from external
links
Allow internal link
equity to flow
Noindex, follow YES NO NO YES YES
Canonical tags YES NO YES YES YES
Robots.txt YES YES NO NO NO
Nofollow,
noindex
YES YES NO NO NO
301 redirect
search urls
w/ links
YES YES (if done right) YES YES YES
So what options do you have?

Ok, so what’s the best
solution dammit?!

Well, it depends…

But to find your best solution…

Size up your search URLs issue

Seek out backlinks pointing to search URLs

301 search URLs with worthy backlinks

Identify search URLs with organic traffic / sales

Prepare to de-index all other search URLs
<meta name=“robots”
content=“nofollow,noindex”>

Don’t forget about search filter URLs too

Change out internal links pointing to search
URLs

Fixes duplicate
content
Reduces crawl
resources
Redistribute link
equity
Passes equity
from external
links
Allow internal link
equity to flow
Noindex, follow YES NO NO YES YES
Canonical tags YES NO YES YES YES
Robots.txt YES YES NO NO NO
Nofollow,
noindex
YES YES NO NO NO
301 redirect
search urls
w/ links
YES YES (if done right) YES YES YES
So what options do you have?

Monitor…
Keep tabs on organic traffic

Monitor…
Keep tabs on number of URLs in index

Monitor…
Keep an eye on redirect counters and
server logs

Monitor…
Keep a close eye on your core KPIs

Measure…
Keep a close eye on your core KPIs

Tips…
If an alternate landing page doesn’t exist,
make one!

Tips…
Refrain from adding in a robots.txt
condition

Finally…
Monitor future search backlinks and action
accordingly

Thanks!
Blog > lukecarthy.com

Final Questions?
Mark Osborne - Dicking Around With The DOM
& Chrome Dev Tools
Ric Rodriquez - How Information Will Transform
The Digital Landscape
Luke Carthy - How to properly optimise your
internal search for big SEO and revenue returns
Mark Osborne - SEO Manager
(Blue Array)
@MarkSEOsborne
Ric Rodriguez - SEO Consultant
(Yext)
@RicRodriguez_UK
(Mayflex)
@MrLukeCarthy

Interested In Speaking?
@MattWSEO
Matt Williamson
@nickynoises
Nicky Elson
@ReadingSEO
Website Coming Soon!

Hosted by Blue Array / Sponsored by SISTRIX
Organisers: Matt Williamson (@mattgwseo) & Nicky Elson (@nickynoises)
Speakers: Mark Osborne (@MarkSEOsborne) / Ric Rodriquez (@RicRodriguez_UK) / Luke Carthy (@MrLukeCarthy)

ReadingSEO Master Deck - 1st August 2019

Recommended

Recommended

More Related Content

What's hot

What's hot (20)

Similar to ReadingSEO Master Deck - 1st August 2019

Similar to ReadingSEO Master Deck - 1st August 2019 (20)

More from Matt Williamson

More from Matt Williamson (6)

Recently uploaded

Recently uploaded (20)

ReadingSEO Master Deck - 1st August 2019