SlideShare a Scribd company logo
SearchLeeds 2017 - Head of Global Marketing - Olga Andrienko - SEMrush - SEMrush ranking factors study
Full research available at:
semrush.com/ranking-factors
Web changes
FAST (almost in
real-time)
10
More content,
more data, more
devices
More impatience,
less attention
Google SERPs are constantly changing
Average volatility score is 4.9 out of 10 (SEMrush Sensor data for Google US)
We face the challenge of
giving clients relevant data
on the market that is
constantly changing.
@olgandrienko #SearchLeeds
Change = disaster
@olgandrienko #SearchLeeds
If you fail to adjust
If you’re ready for it
Change = opportunity
Why ranking
factors
study?
SEMrush data helps search
marketers prioritize their
tasks based on what’s
important and do first
things first.
@olgandrienko #SearchLeeds
Agenda
10
Methodology
Key Takeaways
Opportunities for improved
rankings
@olgandrienko #SearchLeeds
Data Top 100 positions for 600,000
keywords in Google US
@olgandrienko #SearchLeeds
Data
On-page
factors
Traffic
data
Referring
domains
Top 100 positions for 600,000
keywords in Google US
@olgandrienko #SearchLeeds
10
Keywords are grouped by
search volume as the
competition level per group
is different.
Keyword
grouping
@olgandrienko #SearchLeeds
Search
Volume
Groups
@olgandrienko #SearchLeeds
VERY HIGH
HIGH
MEDIUM
LOW
10,001 and up
1,001 – 10,000
101 – 1,000
1 - 100
Keyword
segmentation
per group
Long
tail
Top 3 search volume groups
were further segmented by:
Short
head
@olgandrienko #SearchLeeds
Approaches
10
Correlation analysis
Data Mining
Resources/tools:
R Project, MySQL, Excel
@olgandrienko #SearchLeeds
KEY TAKEAWAYS
Sneak peek at the SEMrush Ranking factors study
Website security (HTTPS)
How website security correlates with positions in search
65% of domains ranking in TOP-3 for high volume
keywords are already secure
@olgandrienko #SearchLeeds
HTTPS Adoption rate of HTTPS is very
impressive. Over 50% domains in
TOP-3 are secure (for any search
volume group)
@olgandrienko #SearchLeeds
Percentage of secure websites: short-head vs. long-tail
@olgandrienko #SearchLeeds
High volume keywords segmented by number of words: up to 3 words in keyword vs. 3+ words
There is a huge increase in
websites adopting HTTPS.
You can’t compete for high
volume keywords without
HTTPS now.
@olgandrienko #SearchLeeds
HTTPS adoption rate in less
competitive niches is 10%
lower which is a great
opportunity to outrun
competition.
@olgandrienko #SearchLeeds
Referring domains
How the number of referring domains correlates with positions in search
The 1st position has almost twice as much ref domains
as pages on 10th position (~10K difference)
@olgandrienko #SearchLeeds
Referring
domains
Higher positions correlate with
more ref domains
@olgandrienko #SearchLeeds
The number of ref domains is
bigger for higher search volumes
For high volumes the curve is a
lot rougher
Backlinks still matter for SEO.
High search volume group is
highly competitive (difference
is 10k referring domains!)
@olgandrienko #SearchLeeds
For lower search volumes
competition is not as strong
and proper link building can
lead to outstanding results.
@olgandrienko #SearchLeeds
Content Length
How content length correlates with positions in search
Higher rankings correlate with longer content
@olgandrienko #SearchLeeds
Content length within TOP-3 is 45% longer than content length for 20th position in ALL search volume groups
Content for long-tail keywords is 20% longer than
for short-head
@olgandrienko #SearchLeeds
Content
Length
The bigger the search volume, the
longer the content.
@olgandrienko #SearchLeeds
Pages ranking for short head and
long tail keywords are NOT the
same ones.
Keyword in title
How keyword presence in title correlates with positions in search
Does keyword in title matter less than we think it does?
@olgandrienko #SearchLeeds
No significant impact was identified for any search volume groups
Bounce rate
How bounce rate correlates with positions in search
Bounce rate increases after TOP-3
@olgandrienko #SearchLeeds
Bounce rate on websites that are ranking for high and low search volume keywords
Bounce rate Bounce rate is lower for top
positions in Google.
Possible reasons: more trust
towards domains on first
positions, content is more
relevant, site loads faster.
@olgandrienko #SearchLeeds
User signals might be too
noisy for Google, but high
bounce rate indicates that
users are not engaging with
your site as they should be!
@olgandrienko #SearchLeeds
Time on site
How time on site correlates with positions in search
Time on site increases for top ranking domains
@olgandrienko #SearchLeeds
Time on site is almost the same for TOP-4, and then starts to decline
Time on site for domains
ranking for low volume
keywords is 40 seconds less
than for high volume.
@olgandrienko #SearchLeeds
Pages per session
How the number of pages per session correlates with positions in search
Pages per session: high volume vs. low volume keywords
@olgandrienko #SearchLeeds
The number of pages per session is similar among TOP-4, and then starts to decline
Visitors coming to the website
from search visit 3 – 3.5 pages
per one session.
@olgandrienko #SearchLeeds
Pages per
session
Higher number of pages per
visit for TOP-4 might be a result
of bigger trust towards domains
on higher positions, content
relevance and higher website
load speed.
@olgandrienko #SearchLeeds
Visits (total and unique)
How the number of visitors and traffic correlate with positions in search
Number of visits matters for high volume keywords
@olgandrienko #SearchLeeds
For lower volume keyword groups the impact is insignificant
Total visits (not unique) including search traffic
@olgandrienko #SearchLeeds
Traffic impacts the rankings of high volume group, for lower volume groups the correlation is way lower
Page
visits
@olgandrienko #SearchLeeds
Search traffic has no
significant impact on rankings.
48
Page
visits
Pages ranking for high
volume keywords have 3
times more visits than lower
search volume groups.
@olgandrienko #SearchLeeds
User behavior signals for TOP-3
Average scores for pages ranking on first 3 positions in high and medium volume groups
User signals
(Average for TOP-3,
high volume)
@olgandrienko #SMX #21A2
All visits
Unique visitors
Direct visits
Pages per visit
97,000
65,000
29,000
3.4
49%Bounce rate
Time on site 4.1 minutes
User signals
(Average for TOP-3,
medium volume)
@olgandrienko #SMX #21A2
All visits
Unique visitors
Direct visits
Pages per visit
63,000
42,000
19,000
3
49%Bounce rate
Time on site 4 minutes
NOW WHAT?
Opportunities for improved rankings
#1
@olgandrienko #SearchLeeds
MAKE SURE YOUR CONTENT LENGTH IS
SUFFICIENT. MAKE SURE IT’S RELEVANT
AND MATCHES USER INTENT.
#2
@olgandrienko #SearchLeeds
CHECK YOUR COMPETITION AND MAKE
SURE THEY DON’T OUTPERFORM IN
TERMS OF LENGTH / AMOUNT OF
WORDS – BUT KEEP IT RELEVANT.
#3
@olgandrienko #SearchLeeds
PROPER HIGH QUALITY LINK BUILDING WILL
HELP YOU RANK HIGHER FOR KEYWORDS
WITH VOLUME OF LESS THAN 10,000 QUERIES
#4
MOVE TO HTTPS. EVEN IF IT’S NOT A
MASSIVE FACTOR, IT WILL HELP WITH
CONVERSIONS & BUILD TRUST.
@olgandrienko #SearchLeeds
#5
CHECK PAGES WITH HIGHER BOUNCE RATE
AND KEYWORDS THESE PAGES RANK FOR.
MAKE THE CONTENT MORE RELEVANT TO
REDUCE THE BOUNCE RATE.
@olgandrienko #SearchLeeds
#6
OPTIMIZE YOUR PAGE LOAD TIMES
TO <3 SECONDS. THIS WILL REDUCE
BOUNCE RATE AND INCREASE SERP CTR.
@olgandrienko #SearchLeeds
Thank you!
OLGA ANDRIENKO
HEAD OF GLOBAL MARKETING
@OLGANDRIENKO
O.ANDRIENKO@SEMRUSH.COM
Dive into the study => semrush.com/ranking-factors

More Related Content

SearchLeeds 2017 - Head of Global Marketing - Olga Andrienko - SEMrush - SEMrush ranking factors study