onsite seo in 2015: an elegant weapon for a more civilized marketer
TRANSCRIPT
By 2007, Link Spam Was Ubiquitous
This paper/presentation from Yahoo’s spam team in 2007 predicted a lot of what Google would launch in Penguin Oct, 2012 (including machine learning)
Even in 2012, It Felt Like Google Was Making Liars Out
of the White Hat SEO World
Via Wil Reynolds
They Finally Launched Effective Algorithms to Fight
Manipulative Links & Content
Via Google
And They Leveraged Fear + Uncertainty of
Penalization to Keep Sites Inline
Via Moz Q+A
These Advancements Brought Google (mostly) Back
in Line w/ Its Public Statements
Via Google
Early On, Google Rejected Machine Learning in the
Organic Ranking Algo
Via Datawocky, 2008
Amit Singhal Shared Norvig’s Concerns About ML
Via Quora
In 2012, Google Published a Paper About How
they Use ML to Predict Ad CTRs:
Via Google
Susan Wojcicki, Google SVP, at All Things Digital, 2012
“Our SmartASS system is a
machine learning system. It
learns whether our users are
interested in that ad, and
whether users are going to click
on them.”
By 2013, It Was
Something Google’s
Search Folks Talked
About Publicly
Via SELand
As ML Takes Over More of Google’s Algo, the
Underpinnings of the Rankings Change
Via Colossal
Google is Public About How They Use ML in Image
Recognition & Classification
Potential ID Factors(e.g. color, shapes, gradients,
perspective, interlacing, alt tags,
surrounding text, etc)
Training Data(i.e. human-labeled images)
Learning
Process
Best
Match
Algo
Google is Public About How They Use ML in Image
Recognition & Classification
Via Jeff Dean’s Slides on Deep Learning; a Must Read for SEOs
Machine Learning in Search Could Work Like This:
Potential Ranking
Factors(e.g. PageRank, TF*IDF,
Topic Modeling, QDF, Clicks,
Entity Association, etc.)
Training Data(i.e. good & bad search results)
Learning
Process
Best Fit
Algo
Training Data(e.g. good search results)
This is a good SERP – searchers
rarely bounce, rarely short-click,
and rarely need to enter other
queries or go to page 2.
Training Data(e.g. bad search results!)
This is a bad SERP – searchers
bounce often, click other results,
rarely long-click, and try other
queries. They’re definitely not
happy.
The Machines Learn to Emulate the Good Results & Try to Fix
or Tweak the Bad Results
Potential Ranking
Factors(e.g. PageRank, TF*IDF,
Topic Modeling, QDF, Clicks,
Entity Association, etc.)
Training Data(i.e. good & bad search results)
Learning
Process
Best Fit
Algo
Deep Learning is Even More Advanced:
Dean says by using deep learning,
they don’t have to tell the system
what a cat is, the machines learn,
unsupervised, for themselves…
Googlers Don’t Feed in Ranking Factors… The Machines
Determine Those Themselves.
Potential Ranking
Factors(e.g. PageRank, TF*IDF,
Topic Modeling, QDF, Clicks,
Entity Association, etc.)
Training Data(i.e. good search results)
Learning
Process
Best Fit
Algo
No wonder these guys are stressed about Google
unleashing the Terminators Via CNET & Washington Post
Googlers Won’t Know Why Something Ranks or
Whether a Variable’s in the Algo
He means other Googlers.
I’m Jeff Dean. I’ll know.
The Query Success Metrics Will Be All That
Matters to the Machines
Long to Short Click Ratio Relative CTR vs. Other Results
Rate of Searchers Conducting
Additional, Related Searches
Metrics of User Engagement
on the Page
Metrics of User Engagement
Across the Domain
Sharing/Amplifcation Rate
vs. Other Results
The Query Success Metrics Will Be All That
Matters to the Machines
Long to Short Click Ratio Relative CTR vs. Other Results
Rate of Searchers Conducting
Additional, Related Searches
Metrics of User Engagement
on the Page
Metrics of User Engagement
Across the Domain
Sharing/Amplifcation Rate
vs. Other Results
If lots of results on a SERP do
these well, and higher results
outperform lower results, our
deep learning algo will consider
it a success.
We’ll Be Optimizing Less
for Ranking Inputs
Unique Linking Domains
Keywords in Title
Anchor Text
Content Uniqueness
Page Load Speed
And Optimizing More for Searcher Outputs
High CTR for this position?
Good engagement?
High amplification rate?
Low bounce rate?
Strong pages/visit afterlanding on this URL?These are likely to be the criteria of
on-site SEO’s future… People return to the siteafter an initial search visit
Remember Our Queries & Clicks Test from 2014?
Via Rand’s Blog
Case closed! Google says they don’t use clicks in the rankings.
Via Linkarati’s Coverage of SMX Advanced
40 Minutes & ~400
Interactions Later
Moved up 2 positions after 2+ weeks
of the top 5 staying static.
Via Google Trends, we can see the relative impact
of the test on query volume
~5-10X normal volume over
3-4 hours
BTW – This is hard to replicate. 600+
real searchers using a variety of
devices, browsers, accounts, geos, etc.
will not look the same to Google as a
Fiverr buy, a clickfarm, or a bot. And
note how G penalized the page after the
test… They might not put it back if they
thought the site itself was to blame for
the click manipulation.
Today, I Think We Know,
Better Than Ever, Where That Is
Welcome to your new home, the User/Usage Signals + ML Model Cabin
Classic On-Site(ranking inputs)
New On-Site(searcher outputs)
Keyword Targeting Relative CTR
Short vs. Long-Click
Content Gap Fulfillment
Amplification & Loyalty
Task Completion Success
Quality & Uniqueness
Crawl/Bot Friendly
Snippet Optimization
UX / Multi-Device
Optimizing the Title, Meta Description, & URL
a Little for KWs, but a Lot for Clicks
If you rank #3, but have a higher-than-
average CTR for that position, you might
get moved up.
Via Philip Petrescu on Moz
Every Element Counts Does the title match what
searchers want?
Does the URL seem
compelling?
Do searchers recognize
& want to click your
domain?
Is your result fresh? Do
searchers want a newer
result?
Does the description
create curiousity &
entice a click?
Do you get the brand
dropdown?
Given Google Often Tests New Results Briefly on Page One…
It May Be Worth Repeated Publication on a Topic to Earn that High CTR
Shoot! My post only made it to #15…
Perhaps I’ll try again in a few months.
With Google Trends’
new, more accurate,
more customizable
ranges, you can
actually watch the
effects of events and
ads on search query
volume
Together, Pogo-Sticking & Long Clicks Might
Determine a Lot of Where You Rank (and for how
long)
Via Bill Slawski on Moz
Speed, Speed, and More Speed
Delivers the Best UX on Every Browser
Compels Visitors to Go Deeper Into Your Site
Avoids Features that Annoy or Dissuade Visitors
Content that Fulfills the Searcher’s Conscious &
Unconscious Needs
An SEO’s Checklist for Better Engagement:
Via NY Times
e.g. this interactive graph that asks visitors to draw
their best guess likely gets remarkable engagement
e.g. Poor Norbert does a terrible job at SEO, but the simplicity compels visitors to go deeper and to return time and again
Via VoilaNorbert
e.g. Nomadlist’s superb, filterable database of
cities and community for remote workers.
Via Nomadlist
Google’s looking for content signals that a
page will fulfill ALL of a searcher’s needs.
I think I know a few
ways to figure that
out.
ML models may note that the presence of certain
words, phrases, & topics predict more successful
searches
e.g. a page about New York that doesn’t mention Brooklyn or Long Island may not be very comprehensive
If Your Content Doesn’t Fill the Gaps in Searcher’s Needs…
e.g. for this query, Google might seek content that includes topics
like “text classification,” “tokenization,” “parsing,” and
“question answering”
Those Rankings Go to Pages/Sites That Do.
Moz’s Data Science Team
is Working on Something to
Help With This
The (alpha) tool extracts likely focal topics from a given page, which can then be
compared vs. an engines top 10 results
In the meantime, check
out
Alchemy API
Or MonkeyLearn
Even for insanely competitive
keywords, we see this type of
behavior when a URL gets
authentically “hot” in the
social world.
I suspect Google doesn’t
use raw social shares as
a ranking input, because
we share a lot of content
with which we don’t
engage:
Via Chartbeat
Google Could Be Using a Lot of Other Metrics/Sources to Get
Data That Mimics Social Shares:
Clickstream (from Chrome/Android)
Engagement (from Chrome/Android)
Branded Queries (from Search)
Navigational Queries (from Search)
Rate of Link Growth (from Crawl)
BTW – Google Almost Certainly Classifies SERPs
Differently & Optimizes to Different Goals
These URLs have loads of shares & may have high loyalty, but for medical queries, Google has different priorities
But If the Competition Naturally Earns
Them Faster, You’re Outta Luck
4 new shares/day
2 new shares/day
3 new shares/day
10 new shares/day
New KPI #1: Shares & Links Per 1,000 Visits
Unique Visits
÷
Shares + Links
Via Moz’s 1Metric
Knowing What Makes Our Audience (and their
influencers) Share is Essential
From an analysis of the 10,000 pieces of content receiving the most social shares on the web by Buzzsumo.
We Don’t Need “Better” Content… We Need “10X” Content.
Via Whiteboard Friday
Wrong Question:“How do we make something as
good as this?”
Right Question:“How do we make something 10X
better than any of these?”
10X Content is the Future, Because It’s the Only Way to Stand
Out from the Increasingly-Noisy Crowd
http://www.simplereach.com/blog/facebook-continues-to-be-the-biggest-driver-of-social-traffic/
The top 10% of content gets all the social shares
and traffic.
Old School On-Site Old School Off-Site
Keyword Targeting Link Diversity
Anchor Text
Brand Mentions
3rd Party Reviews
Reputation Management
Quality & Uniqueness
Crawl/Bot Friendly
Snippet Optimization
UX / Multi-Device
None of our old school tactics will get this done.
Broad search Narrower search
Even narrower search
Website visit
Website visit Brand search
Social validation Highly-specific search
Type-in/direct visit Completion of Task
Google Wants to Get Searchers Accomplishing
Their Tasks Faster
Broad search
All the sites (or answers) you probably would have visited/sought along that path
Completion of Task
This is Their Ultimate Goal:
Eventually end their queries on the topic after visiting Ramen
Rater…
The Ramen Rater
They might use the clickstream data to help
rank that site higher, even if it doesn’t have
traditional ranking signals
A Page That Answers the Searcher’s Initial Query
May Not Be Enough
Searchers performing this query are likely to have the goal of
completing a transaction
Google Wants to Send Searchers
to Websites that Resolve their
Mission
This is the only site where you can reliably find the
back issues and collector covers
Keyword Targeting Relative CTR
Short vs. Long-Click
Content Gap Fulfillment
Amplify & Return Rates
Task Completion Success
Quality & Uniqueness
Crawl/Bot Friendly
Snippet Optimization
UX / Multi-Device
Engines People
I’ve Been Curating a List of “10X” Content Over
the Last 100 Days… It’s All Yours:#1:
bit.ly/10Xcontent
FYI that’s a capital “X”
MonkeyLearn created a tool just for Mozcon to help w/ topic
modeling, keyword extraction, & comparison#2:
Bit.ly/monkeylearnseo
Rand Fishkin, Wizard of Moz | @randfish | [email protected]
bit.ly/onsiteseo2015