Webinar Structured Data

The Rise of Structured Data: Why
NOW is the Time to Optimize
ANNABELLE BOUARD
Product Evangelist
Prior to her role at Botify, she was an SEO consultant, team
leader and trainer at iProspect France, where she was
specializing in large and international websites.
At Botify since 2014, Annabelle supports some of the world’s
largest e-commerce, classifieds and publishing corporations
in their SEO.
Follow us: @Botify - #BotifyWebinar

1. Structured Data, What for?
2. How Does It Work?
3. Analyzing Structured Data with Botify
3.1. What information, where to find it?
3.2. How to use it: as filters
3.3. How to use it: in metrics tables and
custom reports
Agenda

Part #1
Structured Data, What For?

Annotations in the page code
around specific pieces of information
to explain what they are.
(so that Search Engines better understand)
What is structured data?

Classifying things is a universal need.
We have a taxonomy to describe all living
things
➔ Things with a name,
➔ with specific characteristics,
➔ and relationships between them.
Sounds obvious doesn't it ?
Organizing information to create meaning
Source: commons.wikimedia.orgFollow us: @Botify - #BotifyWebinar

Things on the web need to be classified too.
This allows to explain what things are and
how they are connected to other things.
They can be described through the following
notions:
• Entities ⇔ type: what is it?
• Attributes ⇔ properties: what’s special about it?
• Interrelations ⇔ links: how does it relate to other entities?
• Classes ⇔ groups of entities: what attributes do they have in common?
Organizing information to create meaning
Source: https://aws.amazon.com/neptune/

This allows to build
a Knowledge Graph
Google is building a knowledge graph.
This allows the search engine to display information
cards for popular “things”.
Of course, Google didn’t wait for structured data to
recognize certain types of information. But the more
information we provide, the better Google will
understand our pages.

And present results
in the most appropriate way
If Google knows what something is, it can adjust
the result presentation to show what’s most
relevant and interesting about that “thing”, through
Rich Snippets
with all the benefits we know (better user
experience, higher CTR…)

Part #2
How Does Structured Data
Work?

First, the W3C created a base for the semantic web with a
classical conceptual linking approaches (triples) :
- RDF (for Resource Description Framework)
Websites could then describe structured data in web
pages using attribute-based syntaxes :
- RDFa Lite (in Attributes) or Microdata
Since Jan.2014, a new preferred format took the lead :
- JSON-LD (JS Object Notation for Linking Data)
Collaborative work started by Google and other leading
search engines resulted in schema.org.
History and Standards

● Schema.org is a vocabulary: It indicates how the information should be
structured to be understood by all systems : it defines what to say about what.
➢ Type of object ⇔ itemtype ← what it is
➢ Property of an object ⇔ itemprop ← what is specific to it
● Several formats can be used to provide that formatted information:
● RDFa / Microdata are the historical methods of providing semantic annotations in HTML docs
● JSON-LD (latest one) was strongly promoted by Google and as of Sept.2017, recommended
● JSON-LD (JavaScript Object Notation for Linked Data) is also appreciated by developers,
because it is close to the way websites are conceived, is very convenient.
The Schema.org initiative

What Schema.org defines is huge:
close to 600 “types”
Ranging from very generic:
A “thing” at the highest level, generic
types such as creative work (article,
book, movie, music recording….),
Down to very specific types, like a
car, an address or a recipe.
Schema.org hierarchy

Botify automatically retrieves:
● A number of commonly used Objects (product, review, ...)
● Using commonly used Formats (JSON-LD and Microdata)
Botify also retrieves Facebook’s Open Graph markup – this is not
schema.org, it may be present in pages in addition to schema.org markup.
“Facebook Open Graph serves its purpose well, but it doesn't provide the detailed information search
engines need to improve the user experience.” - See http://schema.org/docs/faq.html#4 for more.
What does Botify report on?

Analyzing Structured Data
With Botify
Part #3.1
What
information?
Where to find it?

• Product ⇔ (name, id, brand, type, categories…)
• Offers ⇔ (name, price, currency, availability, condition)
• Ratings ⇔ (number of reviews, ratings average…)
• News ⇔ (author, publication, sections…)
• Dates ⇔ (date published, date modified)
• Recipe ⇔ (name, prep time, cooking time…)
• Restaurant ⇔ (name, category...)
• Job Posting ⇔ (title, job type, industry)
• PostalAddress ⇔ (street, locality, post code, region, country…)
• Breadcrumbs (available soon!)
Types of information retrieved by Botify
Examples:
• Product page:
○ product info,
○ ratings, offers
• Job ad page:
○ job posting,
○ postal address
• Article page:
○ news,
○ date

● Is there structured data in this page ?
→ Notion of main object found
● Is this structured data valid ?
● What ItemTypes are found in this page ?
● What ItemProp are found in this page ?
● What Values do they have ?
What info do we get in Botify?

Number of structured data objects found:
- by type of object (product name, brand, date, …)
- by segment (volume)
Overview + HTML tags section:
Structured Data Inventory

Key information: No. of URLs
or No. of Keywords (imported from GSC)
or No. of Visits (imported from Web analytics)
or URLs crawled by Google (imported from Botify Log Analyzer)
for each
structured data
type
by structured
data value
HTML Tags section: Top indicators by data type

Pages by segment, showing:
- Structured data found & no errors ⇔ green
- Structured data found with errors ⇔ orange
- Structured data not found ⇔ grey
HTML Tags section: Presence and Errors

Report filters and URL Explorer: only types
found on the site appear in the list

The HTML Extract feature retrieves any information from the page code,
to create a custom data field.
It was already possible to extract structured data using the HTML Extract feature, but:
- The number of custom fields is limited.
- These custom fields need to be set up.
Now the Structured Data feature retrieves this information automatically.
No settings needed.
How does the Structured Data feature
compare to the HTML Extract feature?

Part #3.2
Use as Filters
Structured data:
Highly relevant
site-specific information

How To Use: as Filters (check implementation)

How To Use: as Filters (check the competition)

How To Use: Metric Tables (to compare)
Your Competitor :
(same schema data / similar segments to compare)

Analyze the competition: quick methodology
1) Crawl your website
• Define your segmentation ⇔ what to focus on for comparison with competitor
• Assess available Schema ⇔ into your templates
• Complement missing Schema ⇔ define HTML extract (eg. crossed out price)
2) Crawl your competitor's website
• Speed: no more than 3 URLs / sec allowed - Net etiquette
• Similar segmentation
• Assess available Schema ⇔ complement missing Schema with HTML extract)
3) Compare
• Create metric tables (by brand name, by price range, etc.)
• Create custom charts (by schema type, by botify metrics, etc.)

Part #3.3
Use in
Metric Tables
&
Custom
Reports
Structured data:
Highly relevant
site-specific information

How To Use: With Metrics Tables
Publishing website:
- Number of pages,
visits and Google
crawl volume
- by year of publication.
→ Should SEO actions
focus on older articles?
1) Group URLs based on structured data (lines) and display any other indicator (columns)

E-commerce website:
- Linking
- By product price range.
→ Should linking be tweaked,
based on the website
strategy?

2) Group URLs on any criteria and display structured data metrics in columns
Publishing website
with recipes:
- Reviews and
ratings
- By page
segment and
page depth.
→ Some pages in
the top ratings are
very deep and would
deserve to be linked
higher.

How To Use: With Custom Reports
Select any chart….

….and apply filters based
on structured data to
compare sets of pages.

Jobs classifieds website:
- Showing active pages by
depth
- Comparing ads in some
of the top industries
according to jobs
structured data.
→ How could sales jobs pages
be optimized?
Also evaluate their linking,
anchor texts, HTML Tags...
Construction Sales
Hospitality Chefs

Thank you for
your attention!
Q&A
BotifyCONNECT
GET READY FOR
THE MOBILE
FIRST INDEX!
March 2018
Webinar on March 15th

Any questions?
Get in touch!
marketing@botify.com

Webinar Structured Data

Related slideshows

More Related Content

Webinar Structured Data