Is your website invisible to AI? Find out if your site has the technical infrastructure to serve content to ChatGPT, Gemini & Google's AI bots - free AI Visibility Scan.
Book a Free Demo
Log in

Dataset Rich Results

This markup is intended to create ease for finding datasets relating to civic and government data, life sciences, social sciences, and more within Google’s Dataset Search Tool. It should describe the datasets information and represent the contents of the dataset.

Below are Google’ examples of datasets that qualify for this structured data markup:

  • A table or a CSV file with some data
  • An organized collection of tables
  • A file in a proprietary format that contains data
  • A collection of files that together constitute some meaningful dataset
  • A structured object with data in some other format that you might want to load into a special tool for processing
  • Images capturing data
  • Files relating to machine learning, such as trained parameters or neural network structure definitions
  • Anything that looks like a dataset to you
dataset.png

What does your website need?

  • Description: text that describes in 50 – 5000 characters a summary of the dataset
  • Name: a descriptive name of the dataset

There are many other properties that can be added to the dataset markup and are recommended by Google if they are applicable. Below is a list of additional properties that will assist in the accessibility of the dataset.

  • alternateName: another name that the dataset may be referred to as
  • Creator: the author of the dataset
  • Citation: articles in which the dataset is supplementary for
  • isPartOf: if the dataset is the larger set of smaller datasets this property will denote the relationship between multiple datasets
  • Identifier: a DOI or Compact Identifier
  • Keywords: focus keywords of the dataset
  • License: which the dataset is distributed under
  • spatialCoverage: the area in which the data was collected
  • temporalCoverage: the time period in which the data was collected
  • variableMeasured: the particular variable, such as temperature, that was measured
  • Version: the number of the dataset’s version
  • Url: the page in which the dataset is described
  • DataCatalog: the data catalog/repository that the dataset belongs to
  • DataDownload: signifies that the data is downloadable and a link to the download
Frequently Asked Questions

  • Dataset Structured Data markup is intended to make datasets related to civic and government data, life sciences, social sciences, and more discoverable within Google's Dataset Search Tool. It helps describe the datasets' information and represent their contents effectively.

  • Various types of datasets qualify for this structured data markup, including tables or CSV files with data, organized collections of tables, files in proprietary formats containing data, meaningful datasets composed of multiple files, structured objects with data in different formats, images capturing data, and files related to machine learning, such as trained parameters or neural network structures.

  • A website needs to provide a Description, which is a summary of the dataset in 50 to 5000 characters, and a Name that serves as a descriptive title for the dataset.

  • Yes, Google recommends adding several additional properties to enhance dataset accessibility. These include alternateName (another name the dataset may be referred to as), Creator (the author of the dataset), Citation (articles where the dataset is supplementary), isPartOf (denoting relationships between datasets), Identifier (a DOI or Compact Identifier), Keywords (focus keywords of the dataset), License (distribution license of the dataset), spatialCoverage (area where data was collected), temporalCoverage (time period of data collection), variableMeasured (particular variables measured, e.g., temperature), Version (dataset's version number), Url (page where the dataset is described), DataCatalog (data catalog/repository where the dataset belongs), and DataDownload (indicating data is downloadable with a link).

  • Any data that resembles a dataset can be considered for inclusion as structured data. It can include various formats like tables, CSV files, organized collections, proprietary files, meaningful datasets, structured objects, data-related images, and more.

  • Yes, the Dataset Structured Data markup is versatile and can be used for datasets in different domains, such as civic, government, life sciences, and social sciences, among others. It allows datasets from diverse fields to be effectively represented and discovered through Google's Dataset Search Tool.