The Why and How of Structured Data Markup Superstars (#smx #22C)
Pardon me while I put on my big girl pants for this black diamond structured data whirlwind of awesome that is about to happen.
I love structured data. Seriously. It might be my favorite SEM topic and I am beyond excited to be in a room full of Schema superstars.
The session is moderated by Elisabeth Osmeloski (@elisabethos, Director of Audience Development, Third Door Media, Inc.) and there are three panelists who will be presenting.
The agenda promises examples of how each has “effectively used structured data to improve search visibility for articles, movies, products, local business information, and more.”
Ok – introductions aside, let’s get to it.
Evan Sandhaus, Lead Architect, Semantic Platforms, NY Times (@kansandhaus) is up first to talk about Embedded Data and rNews, a form of sematic markup he created for the News Industry. He is the Lead Architect for Semantic Platforms (NY Times Company; part of the research and development team). Woah. Now that is a title…
A Semantic Markup background story from Evan:
It all started with a problem: People weren’t calling the same things the same things, and without consensus on what to call things languages aren’t all that useful. SO his company started looking at a way to formalize a language database – his markup language was called rNews. At the same Google was launching Schema.org; today Schema.org and rNews work together.
Even will cover today:
Why we need Semantic and what Semantic is
The Structured Data standard
Benefits of the SD standard
Why is it important to embed structured HTML into documents
The answer, as with all important things, starts with Justin Bieber.
Evan shows us a news article about Justin Bieber and points out what makes sense to human viewers (image with caption, body text, headline, etc.) and how we, humans, are able to parse meaning because we are – well, human. Then he shows up how spiders/technology see the page and how they are not able to parse meaning/implication just from intuitively reading/scanning the page:
HTML is a standard for expressing the ways things should look; not necessarily what things mean. One underlying result of this is that your SERP results don’t look “as pretty” as those with markup that allow SEs to draw conclusions about meaning and implication.
4 Semantic Markup Standard Types
These are just tools that let you say “there is a thing in this thing called this.” IE: This is a headline. It doesn’t explain What a headline is; what the implication of a headline is – what “headline” means; why it matters.
What is rNews??
rNews is a data model. Developed for machine-readable publishing metadata into web documents (IE: headline, byline, tags, creator).
It’s a set of suggested implementations.
rNEWS expresses the syntactic features of a news document, but also expresses the underlying details of a news article. Like all semantic markup, it helps Search Engines and social networks better understand the meaning of content.
rNews scope: It was first proposed in 2010, then implemented on nytimes.com in September 2011.
Why Semantic Markup?
1) Superior algorithmically – helps SEs and social networks understand your content to make assumptions about your content and then generated links that look pretty in SERPs and social feeds.
2) Superior Tool Support (vertical search, commenting, platforms, rights management—they can all make use of this semantic data)
3) Better Analytics – here are all the page views you got on Semantic markup embedded in a page allows your search team to look at all pages on your site holistically.
Schema.Org and rNews are Friends
Schema was introduced about the same time as rNews – rather than trying to compete with Schema, rNews works with and collaborated with rNews.
Warren Lee, SEO Manager, Adobe (@seocubed) is up next to talk about Schema at Adobe and strategic consideration in regards to which schema you should go after/work on.
About Warren: Joined Adobe as an in-house SEO; helped take Adobe from all external to all-internal SEO team. 11 years in search and 15 years in Digital.. Adobe approaches Schema with a “How do we make, manage, optimize and monetize our product with Schema” mindset.
Warren is going to offer three case study examples of how they (Adobe) use Schema (but he also reiterates that which Schema you go with will depend 100% on your business, your goals, and what you want to accomplish).
Semantic Markup Tools
Structured Data Markup Helper and Google Webmaster Tools Data Highlighter Tool
Three Structured Data Types and What To Use Them For
- Snippets v site links
- Not meant for advertising
- Can include event title event date and venue
- Increase CTR
- With reviews, the voice of the customer is a great way to move customers through the buying cycle funnel
- Test including actual reviews in meta description v. using a controlled meta description
- Consider regional impact (EG: Reviews in Japan can be used more as a complaint platform than a praise platform)
- Consider impact by Industry
- Use the Structured Data testing tool (always TEST!!)
- Hcard is nice to have but…. Authorship is more important than you may think (it has value beyond CTR)
- If you’re not paying for the product then “you are the product”
- Author rank is a ranking factor
Warren definitely encourages you to use re=author (don’t fluff past this opportunity!)
How are they going to improve the relevancy of their current ranking algorithms – Warren thinks of G+ as an engine that helps Google evaluate your author authority. He thinks there is a very good chance it is going to become more of a ranking factor down the line.
Chris Silver Smith, President, Argent Media (@si1very) kicks off his presentation with three reasons why you should use Structured Data:
Three Reasons to Use Structured Data?
If you need to make a case to your CEO, this is your section!
1) Enables search engines to more accurately interpret site content. Makes the message/meaning you are trying to convey crystal clear.
2) Can enhance your SERP listings, make them more attractive, and make them take up more real estate. CTR may improve 15-50% with Rich Snippets.
3) Can increase the number of links per SERP. Logically having more links per SERP = increased odds of clicks. Google actually calls this Schema implementation a “Bonus” for you; if they call is a “bonus” you should probably capitalize on it.
What Spiders are Looking For
1) Schema.org Markup
2) OpenGraph Markup
3) Title and Meta Description
4) If Google doesn’t find any Semantic Markup it will just crawl your page and try to deduce its own description… which is not ideal.
Use both OpenGraph and Schema.org markup simultaneously!
Real World Example: Local Organization Combined with Review Schema:
Always use the structured data testing tool! Check your assumptions. It will show you if you have your author markup implemented correctly, and it will check for any other schema markup you may have, too.
Biggest Tip of the Year for SEO: The Author Tag!
To implement you need to have a G+ page associated with your website and then you need to use the rel=author author tag:
Tables or lists can get Schema, too.
Great example of a page with a comparison chart on it, and how that chart can show up in SERPs with Schemac
Twitter’s Semantic Markup
Twitter Cards – dev.Twitter.com/cards
Add Twitter Card Meta Data to pages >> validate your cards >> then it will show up within a day or two (pretty fast)
Final thoughts on tools from the panel:
If you can’t do the programming use the Data Highlighter tool. If you CAN do the markup – do the markup! The highlighter tool is brittle and
Google Webmaster Tools Structured Data Reports
(This is in bold because it was mentioned with emphasis by all panelists)