How Brand Queries and User Signals Drive Organic SEO By Jonah Stein
-
Upload
search-marketing-expo-smx -
Category
Marketing
-
view
404 -
download
3
Transcript of How Brand Queries and User Signals Drive Organic SEO By Jonah Stein
#SMX #23A1 @JonahStein
How Brand Queries & User Signals Drive Organic SEO
Think Like A Search
Engine:
#SMX #23A1 @JonahStein
SEO CAN’T HEAR GOOGLEThis Was The Wrong Guy
#SMX #23A1 @JonahStein
TL;DRMake great content and Google will find your site
and send you traffic.
#SMX #23A1 @JonahStein
90 Second VersionGoogle is a measurement & data mining company
Most trusted ranking signals come from users
500+ Changes win in testing and launch each year.
Google does thousands of tests every years, trying
to figure out what users like
Crucial question to ask:
WHAT ARE THE CONVERSION GOALS
#SMX #23A1 @JonahStein
Google “Update” Process1. Engineer writes some code
2. Runs some tests in sandbox. Sees whether it is
better or worse (compared to reference/training set
established by manual quality rankers.)
3. Runs live test against percentage of users.
4. If better, goes to evaluation committee.
5. Goes to launch committee
4
#SMX #23A1 @JonahStein
POSITIVE
• Navigational/Brand Queries
• Clicking #1 Result
• Fully Met Needs, Useful
• Long Session (weighted)
• Conversion Action
• Sharing
PROBABLE CONVERSION GOALS
NEGATIVE
• Query Refinement
• Subsequent Clicks
• First Click below 1-3
position
• 2nd Results Page
#SMX #23A1 @JonahStein
Capturing User BehaviorSearch Box
SERP
Clickstream
Adsense
Display Network
Conversion Trackers
Analytics
#SMX #23A1 @JonahStein
Chrome DESKTOP Data
https://www.netmarketshare.com/browser-market-share.aspx
December 2015
#SMX #23A1 @JonahStein
Android Army
#SMX #23A1 @JonahStein
Evolution of Ranking Signals
9
#SMX #23A1 @JonahStein
2001 -2008 Ranking Factors
Relevancy
Backlink anchor text + content
Trust
Backlinks
Authority
Citations Backlinks
SEOs successfully gamed signals
#SMX #23A1 @JonahStein
Algorithmic Relevancy FAILS
In 2008, Google’s Eric Schmidt told us
that "the internet is fast becoming a
cesspool where false information thrives.
Brands are how you sort out the
cesspool”
@jonahstein #UnGagged2015
#SMX #23A1 @JonahStein
USER SIGNALS EMERGE
SITELINKS
BRAND QUERY DATA
Machine Learning
1
2
#SMX #23A1 @JonahStein
User Data At Work
#SMX #23A1 @JonahStein
Big Brand Bailout Feb. 2009
#SMX #23A1 @JonahStein
8 Months Of Wonder(ing)Matthew Trewhella at Meet The Search Engines Q&A, A4UExpo
(reported by Patrick Altoft, Branded3 (Formerly blogstorm.co.uk)
The brand update is about Google minimising the number of times people have to
search to find the products or information they are looking for. Every time a user
has to perform a second search Google regards it as their failure for not
bring up the right result the first time.
So what Google is doing is testing which results are going to give the least
number of secondary searches and displaying those. In the past somebody
might have searched for “travel insurance” and found a few good sites before
remembering that the Post Office does travel insurance too and searched for
them to get a comparison. For Google this is regarded as a bit of a failure
because they didn’t bring up the Post Office in the first place.
#SMX #23A1 @JonahStein
Vince: Query Refinement
Query Refinement = Failure
Query Refinement shows user intent
@SEOBook Graph, Microsoft Search Funnel Tool
#SMX #23A1 @JonahStein
Big Brand Bailout Take Aways
• Initially targeted money terms
• Limited to “a few dozen keywords”
• Validated by Multivariate Testing with the
conversion goal of minimizing query
refinement.
• Search was “fixed” until 2010….
#SMX #23A1 @JonahStein
2010 Content Farms Break Search
“Google has become a jungle: a tropical paradise for spammers and
marketers. Almost every search takes you to websites that want you to click
on links that make them money, or to sponsored sites that make Google
money.”http://techcrunch.com/2011/01/01/why-we-desperately-need-a-new-and-better-google-2/
#SMX #23A1 @JonahStein
Google Search Is Broken
http://www.techradar.com/us/news/internet/google-search-is-broken-928718
https://moz.com/blog/how-organized-crime-is-taking-control-of-googles-search-results
@Jonahstein #UnGagged2015
#SMX #23A1 @JonahStein
Panda Unleashed
#SMX #23A1 @JonahStein
What Is PandaPanda is named after Navneet Panda, a Google
Search Quality Engineer.
Panda is a quality score with its own patent
Panda is the big brand bailout…
On Balco steroids
On a mission to punish sites that have
“benefited unfairly by cheating”
With the nuance and compassion of a religious
fanatic.
#SMX #23A1 @JonahStein
Guidance On Building High Quality Sites
Advice is all very good
Advice is mostly about “articles” and
PUBLISHER focused.
Advice never mentions anything important to
the Panda update.http://googlewebmastercentral.blogspot.com/2011/05/more-guidance-on-building-
high-quality.html
#SMX #23A1 @JonahStein
Navneet Panda PatentFrom 2012 Patent:
The score is determined from quantities indicating user
actions of seeking out and preferring particular sites and the
resources found in particular sites. A site quality score for a
particular site can be determined by computing a ratio of a
numerator that represents user interest in the site as
reflected in user queries directed to the site and a
denominator that represents user interest in the resources
found in the site as responses to queries of all kinds
#SMX #23A1 @JonahStein
Direct + Branded + NavigationalDirect = Type-in, Book Mark & Auto Complete Traffic
Branded Search = Queries for your company, products or content
Navigational Search = queries directed to a specific site
For example, if the search system has data indicating that the
terms “example sf” and “esf” are commonly used by users to refer
to a site “sf.example.com,” queries that contain the terms “example
sf” or “esf”, e.g., the queries “example sf news” and “esf restaurant
reviews,” can be counted as queries that refer to the site
“sf.example.com.”
DBN Queries Fully Meets searchers intent.
#SMX #23A1 @JonahStein
DBN is NumeratorSEO Traffic is Denominator• Panda Site Score is the
quotient of
DBN/Search Traffic
• Traffic you
“Earned”/“Given”
#SMX #23A1 @JonahStein
Machine Learning If RankBrain sees a word or phrase it isn’t familiar with, the
machine can make a guess as to what words or phrases
might have a similar meaning and filter the result
accordingly, making it more effective at handling never-
before-seen search queries.
Google has trouble with queries it has never seen before– So
Relevance is driven by USER BEHAVIOR ON QUERIES
#SMX #23A1 @JonahStein
Machine Learning Driven by User Testing
Google's Andrey Lipattsev, transcription by Barry Schwartz
Once you don't know precisely how something works, is that ok?
is it ok to be comfortable enough to know the basic principles and
where it all started? I think what helps us a lot is, what helps me
to be comfortable with the way things are developing, is that we
rely a lot on data about the results of what is going on. we
continuously keep checking how whatever is happening, how is
it effecting our users. are they happy with it, or are they not
happy with it? https://www.seroundtable.com/google-why-it-is-so-hard-to-confirm-all-algorithm-updates-21510.html
2
7
#SMX #23A1 @JonahStein
Think Like GoogleKnowledge team doesn’t think about
revenue
Websites are “publishers”
@AmitSinghal, SMX 2014 Keynote
Said “users” 23 times; “users”, in
almost every answer
We test constantly
Decisions are driven by “what
users want”
#SMX #23A1 @JonahStein
What Does The Hive ThinkThey are building a great PRODUCT
The mission is “knowledge”, not search
The world is changing very fast
Mobile is the most disruptive thing in Google history, driving
most of the change.
Threats are from game changers
They don’t feel our pain!
#SMX #23A1 @JonahStein
“A Rising Tide Lifts All Boats”
#SMX #23A1 @JonahStein
POSITIVE
• Navigational/Brand Queries
• Clicking #1 Result
• Fully Met Needs, Useful
• Long Session (weighted)
• Conversion Action
• Sharing
PROBABLE CONVERSION GOALS
NEGATIVE
• Query Refinement
• Subsequent Clicks
• First Click below 1-3
position
• 2nd Results Page
#SMX #23A1 @JonahStein
Google Quality Raters GuidelinesExternal team EVALUATING SEARCH RESULTS
Your site in the context of the query
“ratings from evaluators do not determine
individual site rankings, but are used help us
understand our experiments.”
“The evaluators base their ratings on guidelines
we give them; the guidelines reflect what Google
thinks search users want”.
#SMX #23A1 @JonahStein
EAT
Expertise
Authoritativeness
Trustworthiness
#SMX #23A1 @JonahStein
PREY
Adwords
Your Money or Your Life
Payday
Copy Right
#SMX #23A1 @JonahStein
LOVEUsefulness
Guidelines focus on inferring intent
“Accomplish their task very quickly”
“mobile phones can be difficult to use”
“Help user immediately get information or
content”
#SMX #23A1 @JonahStein
Needs Met“Need Met” as Google Conversion Goal: how
many visitors did I help:
#SMX #23A1 @JonahStein
Fully Met Reserved For
Fully met should be reserved for results
that are the complete and perfect
response or answer’ so that no other
results are necessary for all or almost
all users to be fully satisfied.”
Hierarchy of Needs Met
Navigation/Brand
Knowledge Graph
3-pack
#SMX #23A1 @JonahStein
Highly MeetsHighly Met
“Good fit” for query, lower confidence
– vague query (Walmart)
Highly Meets & Disambiguation
Product/Category pages matching query but not
ONLY matching query (“kids backpacks”)
#SMX #23A1 @JonahStein
Moderately Meets• Moderately Meets refers to content that would be
“helpful for many users or very helpful for some users.” So these are generally still quality results, but don’t quite meet the needs for the searcher.
• Moderately Meets results could be less up-to-date, less comprehensive or simply from sites that aren’t really an authority in that space SOURCE: THE SEM POST
#SMX #23A1 @JonahStein
Slightly Meets
• Google judges content that gets a rating of Slightly Meets as results that are low quality, outdated, clearly neglected, or far too broad or specific. These are results that are helpful to only some or few searchers.
• It also covers things that are minor interpretations, where there is an alternative meaning or interpretation but one that doesn’t see a lot of interest.
• One interesting note is that Google considers Wikipedia an example of Slightly Meets in this section, as it returns a result that is too broad to easily answer the searcher’s query.
SOURCE: THE SEM POST
#SMX #23A1 @JonahStein
Fails To Meet• This is the lowest rating… but there is one major
thing all webmasters and SEOs should know – ALL sites that are NOT mobile friendly will be rated as “Fails to Meet”.
• Other than non-mobile friendly, fails to meet would include the usual crap content suspects like scraper sites, pages ranking for terms the page has nothing to do with as well as really outdated sites.
• This rating is also given to sites that are helpful to no users or very few of them. SOURCE: THE SEM POST
#SMX #23A1 @JonahStein
SoIf Google is a giant Measuring and Data Mining company
If Google has Petabytes of data about user behavior
If Google is trying to figure out what users like
If user behavior can change search results
#SMX #23A1 @JonahStein
Build Useful Site
• Optimize Content for USERS• Optimize user behavior to reduce non-engagement.• Information scent is key to engagement (3-5 seconds)• Segmentation not selling• Answer questions & concerns instead of selling products
#SMX #23A1 @JonahStein
Driving Brand Queries
Content = Assets, not landing pages
Market your content; owned, earned & paid
Brand your experience & value proposition
Create proprietary metrics
Rotten Tomatoes
PageRank, Moz Rank, TrustFlow
Focus on growing your audience
Don’t let CRO get the best of you
#SMX #23A1 @JonahStein
Driving Brand Queries IIVideo
Youtube
Facebook paid
Offline Advertising with CTA
Easter Egg Hunts
Social Media
Online Advertising
Barnacle, Guest Post, & Syndication
#SMX #23A1 @JonahStein
SEE YOU AT THE NEXT #SMX
THANK YOU!