Google Assistant app on iPhone with the results of a “what time does dr. ruhlman office close” query. The results displayed include a card with “8:30AM–5:00PM” and the label, “Dr. Ruhlman Scott MD, Tuesday hours,” as well as links to call the office, search on Google, get directions, and visit a website. Additionally, there are four buttons labeled with the words “directions,” “phone number,” and “address,” and a thumbs-up emoji.

There’s not enough proof within this little sample to encourage a broad claim that calculations have”cognitive” bias, but even if we allow for possibly confounding variables, we could see the compounding issues we hazard by ignoring structured content. “Donlon,” for instance, may well be much more common name than”Donion” and may be easily mistyped onto a QWERTY keyboard. Regardless, the Kaiser Permanente outcome we are given previously for Dr. Donion is to get the wrong physician. What’s more, in the Google Assistant voice search, the interaction format does not verify whether we meant Dr. Donlon; it just provides us with her facility’s contact information. In these scenarios, providing transparent, machine-readable content may work to our advantage.
Structured content is a mainstay of many types of information on the web. Listings, for example, have been predicated on content for ages. While I hunt, by Way of Example,”bouillabaisse recipe” on Google, I am provided with a standard list of links to recipes, as well as an overview of recipe measures, an image, and a set of tags describing one instance recipe:
In order to tailor results to such questions that were more specifically formulated, software agents have started then and inferring intent using the data that was linked at their disposal to build a targeted, concise response.

Google Assistant app on iPhone with the results of a “how do I pay a parking ticket in Boston” query, showing results only weakly related to the intended content.
  • Content Modelling: An Expert Ability ,” Rachel Lovinger
  • Google Structured Data Testing tool showing the markup for a bouillabaisse recipe website on the left half of the screen and the structured data attributes and values for structured content on the right half of the screen.
    The exact same page seen in Google’s Structured Data Testing Tool. The values that are machine-readable are shown by the pane on the right.

    The role of structured articles

  • Designing Connected Content, Carrie Hane and Mike Atherton
  • The City of Seattle website‘s “Pay My Ticket” page, showing four methods to pay a parking ticket in a simple, all-text layout.

    In late 2016, Gartner called that 30 percent of internet browsing sessions would be achieved with no screen by 2020. Earlier the exact same year, Comscore had predicted that half of all searches would be voice hunts by 2020. Though there’s recent signs to imply the 2020 picture might be more complex than those broad-strokes projections imply, we’re already seeing the effect that voice hunt, artificial intelligence, and intelligent software agents like Alexa and Google Assistant are creating about the way information can be found and consumed on the web.

    Google Assistant app on iPhone with the results of a “how do I pay a parking ticket in Seattle” query, showing nearly the same results as on the desktop web page referenced above.

    Say that I want to collect more info about two recommendations I’ve been given for orthopedic surgeons. A hunt for an initial recommendation, Scott Ruhlman, MD, brings up a set of connections in addition to a Knowledge Graph info box containing a photo, location, hours, telephone number, and reviews from the net.
    In their latest publication, Designing Connected Content, Carrie Hane and Mike Atherton define structured content as articles which is”planned, designed, and connected outside an interface so that it’s prepared for any interface.” A structured content layout approach frames articles resources–such as recipes, articles, product descriptions, how-tos, profiles, etc.–maybe not as pages to be found and read, but as packages composed of little chunks of content information that all relate to one another in meaningful ways.
    Semantic HTML is markup that communicates information about the purposeful relationships between document elements, as opposed to just describing how they should look on screen. Semantic elements like heading tags and list tags, for instance, indicate that the text they enclose is a heading (

    ) for the set of record items (

  • ) in the ordered list (
      ) that follows.
      In this instance, Dr. Ruhlman’s profile has been marked up using microdata based on the language. This base that is content that is ordered provides the exact base on which additional content relationships can be constructed. The Knowledge Graph info box, for example, comprises Google testimonials, which aren’t part of Dr. Ruhlman’s profile, but which have been aggregated into this overview. The overview also has an interactive map, made possible because Dr. Ruhlman’s workplace location is machine-readable.
      Not one of the links supplied in this Google Assistant results take me straight to the”How to Pay a Parking Ticket” page, nor do the descriptions clearly let me know I am on the ideal path. (I didn’t ask about asking a hearing) This is due to the fact that the content on the City of Boston parking ticket page is designed to convey content connections visually but is not structured in a way that also conveys those connections to inquisitive algorithms.
      MultiCare Neuroscience Center, you’ll recall, is Dr. Donlon–the neuroscientist Google believes I might be searching for, not the orthopedic surgeon I am really looking for–practices.
      You’ll also observe that while Dr. Stacey Donion is an specific match in all the listed search results–which are many enough to meet with the first results page–we are shown a”did you mean” link to get another physician. Stacy Donlon, MD, is a neurologist who practices. Multicare related profiles that are data-rich for their doctors and does provide semantic.
      Instead of a concentrated selection of focused actions to follow up on my query, I am presented .

      Along with the indexing purpose that search engines function, AI-powered search calculations and smart agents are bringing into the mainstream of accessing advice: aggregation and 34, additional modes. As a result, design efforts that focus on creating pages that are visually effective are not enough to guarantee the integrity or accuracy of content. Instead, by focusing on providing access to information in a structured, systematic manner that’s legible to both humans and machines, content publishers may make sure that their content is equally accurate and accessible in those new contexts, whether or not they’re creating chatbots or tapping into AI directly. In this article, we’ll consider the forms and effect of material that is structured, and we’ll close with a set of tools which can help you get started with a content that is structured approach to information design.
      In addition to finding and excerpting info, such as parking ticket payment options or recipe measures, software and search representative algorithms now today aggregate content from several sources by using linked data.
      In this instance, we can see that Google is able to find lots of links to Dr. Donion in its own standard index outcomes, but it is not able to”know” the information about those sources well enough to demonstrate an aggregated outcome. In cases like this, that the Knowledge Graph knows Dr. Donion is a Kaiser Permanente doctor, but it pulls in the wrong location and the wrong doctor’s name in its endeavor to construct a Knowledge Graph display.

      Google search results page for Dr. Donion, showing a list of standard links for Dr. Donion, and a 'Did you mean: Dr Stacy Donlon MD' link at the top. There is a Google info box, as with the previous search results page example. But in this case the box does not display information about the doctor we searched for, Dr. Donion, but rather for 'Kaiser Permanente Orthopedics: Morris Joseph MD.'

      The equivalent Google Assistant search, however, offers a much more useful result than we see with Boston. In cases like this, the Google Assistant result links right to the”Pay My Ticket” page and also lists a number of ways I can pay my ticket: online, by mail, and also in person.
      Despite the visual simplicity of the Town of Seattle parking ticket page, it ensures the integrity of its material across contexts because it’s composed of content that is marked up. “Pay My Ticket” is a level-one going (

      ), and also every one of the options below it are level-two key words (

      ), which indicate that they are subordinate to the level-one component.

      Voice inquiries and content inference

      These components, when designed convey connections and data hierarchy visually to readers, and semantically to algorithms.

      In a structured content design process, the relationships between content chunks are defined and clarified. This makes the connections between them as well as the content chunks legible into calculations. Algorithms can then translate a content package as the”webpage” I’m looking for–or remix and accommodate that same content to give me a list of instructions, the number of celebrities on a review, the period of time left before an office shuts, and some number of other concise answers to specific questions.
      Design practices that build bridges between user requirements and technology requirements to fulfill company goals are critical to making this vision a reality. Content strategists, information architects, developers, and expertise designers all have a part to play in providing and designing content options.
      These outcomes aren’t just aggregated from sources, but are interpreted and remixed to provide a customized response to my particular question. Getting instructions, placing a phone call, and accessing Dr. Ruhlman’s profile page on are at the tips of my hands.
      If we conduct Dr. Ruhlman’s Swedish Hospital profile page through Google’s Structured Data Testing Tool, we can see that articles about him is organised as small, different elements, each of which is marked up using descriptive types and characteristics that communicate both the meaning of those attributes’ values and the way they fit together as a whole–all in a machine-readable arrangement.
      This announcement was intended to aid strategists, designers, and companies get ready for the rise of mobile. With the prevalence of supporters and voice-based inquiries, a company’s website is less and less likely to become a customer’s first experience with rich content. Such as ratings, hours, phone numbers, and finding location information –this engagement may be a user interaction with a data source.

      Because people understand what lists and headings look like and mean HTML structured in this way is both semantic and presentational, and calculations can comprehend them as components with interpretable relationships.

      The City of Boston website's “How to Pay a Parking Ticket” page, showing a tabbed view of ways to pay and instructions for the first of those ways, paying online.

      Program broker search and semantic HTML

      HTML markup which concentrates just on the presentational aspects of a”webpage” may seem perfectly fine to an individual reader but be completely illegible into an algorithm. If I Need to find information about how to pay a parking ticket, a link in the home page takes me directly to the”How to Pay a Parking Ticket” display (scrolled to show detail):

      The business case for structured content layout

    1. Content Everywhere, Sara Wachter-Boettcher
    2. Google Structured Data Testing tool, showing the markup for Dr. Ruhlman's profile page on the left half of the screen, and the structured data attributes and values for the structured content on that page on the right half of the screen.

      It is possible that during repeated exposure to the search phrase”Dr. Stacey Donion,” Google Assistant fine-tuned the answers it provided. The very first result, however, indicates that smart agents might be at least partially susceptible to the same accessibility heuristic that affects individuals, wherein the information that is easiest to recall frequently seems the most correct.
      In its most basic form, connected data is”a set of best practices for linking structured data on the web.” Linked data extends the fundamental capacities of semantic HTML by describing not just what kind of thing a page element is (“Pay My Ticket” is an

      ), but also the real world concept that item represents: that

      signifies a”cover action,” which inherits the structural attributes of”trade actions” (the market of goods and services for money) and”activities” (actions carried out by a broker upon an item ). Data creates a more nuanced description of the relationship between page components, and it supplies the conceptual and structural advice that calculations need to bring data together.

      The City of Seattle website’s 'Pay My Ticket' page, with two HTML heading elements outlined and labeled for illustration, and an open inspector panel, where we can see that the headings look the same to viewers but are marked up differently in the code.

      These kinds of interactions, but are only one small part of a larger problem: linked data is increasingly key to maintaining the integrity of content online. The associations I have used as examples, such as the hospitals, government agencies, and colleges I have consulted for years, do not measure the achievement of the communications efforts in ad clicks or page views. Success for them means linking community members, components, and patients with precise information regarding the business that information may be found and solutions. This communication-based definition of success easily applies to practically any type of organization working to further its business goals on the web.

      Google Assistant app on iPhone with the results of a “what time does Doc Dr Stacey donion office close” query. The results displayed include a card with “8AM–5PM” and the label “MulitCare Neuroscience Center, Monday hours,” as well as links to call the office, search on Google, get directions, or visit a website.

      By conveying clearly in a digital context that includes inference and aggregation, organizations are able to speak to their users where users actually are, be it on a web site, an internet search engine results page, or even even a digital helper. They’re also able to maintain control over the accuracy of their messages by ensuring that the proper content are available and hauled across contexts.

      I readily understand what my choices are for paying as a human reading this page I can pay online, in person, on the telephone, or by email. If I ask Google Assistant to pay a parking ticket in Boston, but things get a little confusing:
      The City of Seattle’s”Pay My Ticket” page, though it lacks the glistening visual design of Boston’s site, also communicates parking ticket options obviously to individual visitors:
      While this usage of semantic HTML offers distinct advantages over the”page screen” styling we saw on the Town of Boston’s site, the Seattle page also shows a weakness that’s typical of manual approaches to semantic HTML. You will observe that, in the Google Assistant results, the”Pay by Phone” option we found on the webpage was not listed. If we examine the markup of the webpage, we can see that while the three choices found by Google Assistant are wrapped in both and

      tags,”Pay by Phone” is only marked up using an

      . This irregularity in semantic structure could be what’s causing this option to be omitted by Google Assistant .

      The model of constructing pages and then expecting users parse and to detect those pages to answer queries , though time-tested in the age, is quickly becoming inadequate for successful communication. From engaging in emerging patterns of information discovery and seeking, it precludes organizations. And it might lead software representatives to make inferences based on information that is erroneous or insufficient , possibly routing customers to rivals who communicate effectively.
      An enormous difference is read by the machine Though every one of those elements would seem the exact same into a human creating this page. They fall prey into the idiosyncrasies of even the most well-intentioned content writers Even though WYSIWYG text entry fields can encourage HTML, in training. By making content structure a core part of a website’s content management system, organizations can create semantically correct HTML every moment, for each element. This is also the base which makes it possible to capitalize on the rich relationship descriptions given by linked data.

      Getting started: that and how

      The City of Seattle website’s “Pay My Ticket” page, with the HTML heading elements outlined and labeled for illustration.

      The hunt for a recommendation, Stacey Donion, MD, provides a different experience. Like the City of Boston site above, Dr. Donion’s profile on the Kaiser Permanente website is perfectly intelligible to some sighted human reader. But because its markup is presentational, its material is invisible to software agents.

        Google search results page for Scott Ruhlman, MD, showing a list of standard links and an info box with an image, a map, ratings, an address, and reviews information.
      • Content Strategy for Mobile, Karen McGrane
      • Linked content and data aggregation

        This”featured snippet” perspective is possible because the content writer,, has broken this recipe into the smallest meaningful chunks appropriate with this subject matter and audience, then expressed advice about these chunks as well as the connections between them at a machine-readable manner. In this instance, has used both semantic HTML and connected data to make this content not only a webpage, but also legible, accessible data which may be correctly interpreted, accommodated, and remixed by algorithms and smart agents. Let’s look at each one of those components in turn to learn how they work together across inference contexts, and indexing, aggregation.

      A combined HTML code editor and preview window showing markup and results for heading, ordered list, and list item HTML tags.

      On creating material systems which work for algorithms and humans alike practitioners from across the design community have shared a wealth of resources in recent years. To learn more about executing a content that is structured strategy for your organization, these books and articles are a great place to start:

      Google search results info box for Dr. Ruhlman, showing an photo; a map; ratings; an address; reviews; buttons to ask a question, leave a review, and add a photo; and other people searched for.

      The prevalence of voice as a mode of access to data makes supplying structured content more significant. Software agents that are intelligent and voice are not solely preventing users they’re changing user behaviour. Based on LSA Insider, there are many important differences between voice queries and typed queries. Voice queries often be: