This is a summary of the talk that Alexis Sanders gave at Brighton SEO 2018.
What is structured data?
Structured data refers to information with a high degree of organisation, using some sort of microdata which structures said data for bots to access and better understand.
However, it’s very easy to make small errors in structured data so it’s essential to always validate.
Initially, Google started recognising Microdata only but now has started evaluating JSON-LD as well as Bing. Microdata is embedded in html whereas JSON-LD is in javascript meaning it is a little bit easier to understand and copy in.
Why we love structured data
Structured data presents a great opportunity to feature in the enhanced components of the SERPs, by increasing SERP ‘real estate’.
A study that Alexis conducted on 7 ecommerce sites found that those sites that implemented structured data such as review and price had a greater click-through rate almost 20% over those sites that didn’t.
Smart home devices such as Amazon Echo or Google Home pull their information from featured snippets, which are generated from relevant structured data. Alexis conducted another study testing 100,000 queries, checked 25,000 of those and found that Microdata was present for 75% of the results.
The top 5 most popular schema are webpage, sitenavigarion, organisation, person and localbusiness.
A new and interesting schema to look into is speakablespecification schema markup, which provides better results with voice search, CNN is an example of a big brand with this already implemented.
Common Pitfalls
Syntax: a difference between curly quotation marks and brackets can cause a lot of issues if incorrect
Vocabulary: use specific vocabulary to give Google the best chance to understand your structured data
Policy Violations
Nesting Errors: make sure to attribute the correct product to the correct category type
Schema Prediction
Alexis thinks there will be big developments in the future for schema that supports voice, as well as specifically for health both proactive and reactively.
This post may have been written by a past member of the Impression team, or be a collaboration of us all. We hope you like it!
