Data Driven SEO - Webinar - Webpromo.expert

International Freelance SEO

International Freelance SEO

Brand Ambassador Majestic

Cycling & Skating

Science: Physics in particular

Simplicity is key

Links still tell something about relationships

between pages but also between entities.

Collect as much data as possible. More is not

always better but filtering is easier than a

shortage of data.

Answering and explaining outcomes can be

difficult, so make sure you present data in a

understandable way

What are the pages that are

already ranking well and how

are they performing?

Best linked pages, high

potential pages without links

etcetera.

SEO relevant elements filled

in? Amount of content per

page vs performance

First step to make

Make sure you master the Vlookup function

in Excel: http://www.gcflearnfree.org/excel-

tips/how-to-use-excels-vlookup-function

Lets start with some

Next step is

Link: http://seotoolsforexcel.com/

• =HtmlTitle(string url)

• =HtmlMetaDescription(string url)

• =HtmlH1(string url)

Next step is

Nice post from Martijn Scheijbeler: http://thenextweb.com/insider/2016/03/08/marketing-

tnw-way-10-google-search-console-master-855-properties/

• Segment data for organic (if the website is

international, I suggest you segment per unique

search engine)

• Segment data for mobile and desktop traffic

separately to analyse difference between

Google’s indices

• Download sessions, % new sessions, bounce

rate, session duration, conversion or e-

commerce data to start with.

Next step is

Always think about the real user

Complete tutorial at http://www.stateofdigital.com/data-driven-seo-

getting-detailed-insights-using-free-tools/

Create a list of page URLs that already

established visibility (but don’t have the top

positions yet) in Google but lack internal

links.

Action: Add more internal links to those

pages.

Built a list of pages with a good amount of

organic traffic but with a high bounce rate.

Action: Ask yourself the question: is the

content relevant to the used queries?

Enrich, update or remove content

accordingly.

Find pages that are not ranking well in

Google, but getting valuable, good

converting traffic via other sources.

Action: Content matches the brand or

website obviously so try to get to rank those

pages in the organic search results.

Look for pages that are visible, have great

onpage metrics but no top 3 positions.

Action: get more external or internal links

pointing to that page.

Pages with low amount of internal links, but

a lot of external links and having top 10

positions.

Action: Add internal links, additional content

and move towards to the top 3 positions!

Create an overview of pages that are

ranking well, but not having a great click

through rate.

Action: Optimize your meta data to increase

the CTR. If it doesn’t improve your results,

ask yourself if your websites is expected to

be ranked

Create a list of URLs that are visible in

search results and getting impressions

(GSC data) but are not having any real

content targeting a specific query space.

Action: Optimize the on page content and

your site will benefit from it in most cases.

Now… Let’s be honest

The key of data driven SEO is

• Crawl a website

• Download all internal linking relations

• Use Gephi or NodeXL to visualize the relations

between pages

• Use Gephi to calculate PageRank

Tutorials: • http://www.vpascual.org/blog/visualization/visualize-the-link-structure-of-

your-website-using-free-tools/

• http://www.stateofdigital.com/using-pagerank-internal-link-optimisation/

• http://www.stateofdigital.com/visualize-your-sites-internal-linking-structure-

with-nodexl/

Important page?

Site has 6 languages, what is that 7th

cluster?

Need to get more budget for

Our average website page has only

Similar to high bounce rates, look for pages

with good traffic numbers and bad

conversion rates.

Action: try testing a new layout, different

content or different pricing for example.

Next to conversion rates, have a look at the

average value per user.

Action: Prioritising can be done according to

the best performance page in terms of

revenue or the average number of

conversions per unique user.

Fighting the Panda: does your website has a

lot of pages that don’t receive any organic

traffic at all?

Action: If the majority of your pages is not

receiving any traffic at all, start removing old,

irrelevant pages.

Try clustering your pages in themes. For e-

commerce platform you can sum up metrics for

specific product categories for example.

Are there any specific folders or clusters of

pages that are relevant to each other not receiving

any traffic? This could be a sign of a link penalty or

Google not finding your website relevant enough

for that specific query space.

Watch my feeds (@jbobbink /

notprovided.eu) where I’ll share some

machine learning APIs that can do it for

you

Or check Monkeylearn.com to start already!

Pages that perform well (good conversion

rate) in AdWords but not ranking yet in the

organic results?

Action: Push them harder