Google has unveiled a new schema markup designed to display data tables directly in search results.
This markup is targeted at data journalists working on news articles, though it can be applied to any dataset irrespective of the publisher.
Google provides examples of what can be considered a dataset, including:
– A table or a CSV file containing data
– An organized collection of tables
– A data-containing file in a proprietary format
– A collection of files forming a coherent dataset
– A structured object with data in a format suitable for specialized processing tools
– Images depicting data
– Files associated with machine learning, such as trained parameters or definitions of neural network structures
– Anything that resembles a dataset to you
Google’s objective is to enhance data discoverability and clarify to searchers when an article or document includes a potentially useful dataset.
Detecting and understanding tables of data is challenging for Google, which is why site owners need to indicate which information is most valuable using the new dataset markup.
“It works like this: news organizations that publish data in table form can add structured data to make the dataset elements of the page easier to identify for relevant Search features,” says Google.
Google has already collaborated with 30 leading data journalists worldwide to integrate this structured data markup.
One participant, ProPublica, describes adding the markup as “trivial” compared to its potential benefits.
For details on implementing dataset structured markup, refer to Google’s developer documentation.
