Richard Wallis – Structured Data: It’s All about the Graph, Benchmark 2019 Talk Review

Dec 6th, 2019

With more than 20 years working in and advising on information architecture, Wallis is a leading voice in what has become one of the most important web developments to have emerged in the last decade – structured data


My obsession with structured data has developed to the point where it is now a running joke within the marketing department – so when I gave my talk at Benchmark this year, I did so with an extra helping of nerves. The reason being, I was touching on schema and structured data and sat in the front row, directly in my line of sight, was Richard Wallis – one of the original Schema.org community and a consultant to Google amongst others on the topic of information architecture.

The argument Wallis’ talk makes for schema is that you can view your content as a piece of the jigsaw that is the wider ontology of your industry. To properly fit, and to be correctly contextualised within that ontology, it needs to connect with as much of that external meaning as possible – and this can be done using markup.

He presents an example – using travel blogging as an example, and laying out the various ways in which a piece of content – and the domain it’s posted on – can be connected to the wider ‘knowledge graph’. In a fictitious post on Paris as a tourist destination (which we can see as one entity), for example, he used schema to refer to the entity ‘Paris’, but also contextualises the author (another entity) – using schema to reference expertise. Using layers of schema, Wallis presents the basic methodology used to contextualise information for the knowledge graphs.



In a slide titled ‘Google Knowledge Graph – How do I get my stuff into it?’ Wallis gives two steps that can facilitate the process:

Step 1 (SEO)

  • Get your pages crawled and indexed
  • If Googlebot doesn’t find your pages useful, it’s unlikely data will be loaded

Step 2 (Structured Data)

  • Describe the things the pages are about – as against the pages themselves
  • Use Schema.org
  • Relate them to other things – the author is a Person, that needs describing. Don’t just provide a name and hope Google works out who it is!
  • Markup all your entities including your Organization and link them together
  • Don’t worry about lists – marking up the individual things is more important
  • Preferably use JSON-LD – validate with tools, Google & Schema.org docs.


Overall, while we’ve been promoting the use of structured data for some time (one of the reasons we were delighted Wallis could speak), the talk did a fantastic job of clearly describing not just the concepts involved but a method of execution that attendees could genuinely go back to the office and implement.

Dev time can often be difficult to obtain for SEO in many organisations – especially for something like structured data which is difficult to communicate effectively to stakeholders – but this talk provides a ready to go business case for anyone needing one, and – with a section at the end covering the future usefulness of markup, it really was a talk that SEOs and digital marketers in general should definitely spend 25 mins watching.


Make sure you don’t miss out on the remaining Benchmark talk reviews, as well as keeping up to date with news, how tos and interviews, subscribe to our mailing list! Otherwise, why not contact us today to see what we can do for your brand?

Facebook Twitter Instagram Linkedin Youtube