Search engine optimization (SEO) is the practice of strategically optimizing web content to make it easier for search engines to find and display your content. Adding metadata or structured markup to web pages is an important component to SEO. While search engines continually change their practices and algorithms, below are some general guidelines.
Key Factors for Web SEO
This checklist notes key factors and considerations that impact SEO:
- Aim for an established, long-standing website with stable urls, incoming links, and regular crawling/indexing by Google.
- Provide good Titles and Descriptions (note character limits – see below). See Google's documentation for more information.
- Use keywords within actual content/content relevancy (see below).
- Use the H1 tag selectively – provie a version of your title in an <h1> tag (Heading 1).
- Provide readable page URLs (words vs id's/numbers; separate words using hyphens)
- Register for a Google webmaster account, and also create a Google Sitemap – Sitemaps helps Google crawl. Webmaster Tools monitor when Google crawls your site and allows you to request a recrawl (helpful for site launches, or if you move your site). Google Analytics can be configured which allow you to see search activity within your site.
- Consider adding Dublin Core metadata (especially if it can be automated) in case the content may be utilized by a library or repository system.
- Consider adding custom metadata standards for social media: specific metadata options can be configured for Facebook, Twitter, and Pinterest.
- Use an online SEO testing tool: woorank.com can run an automated scan of your page to check for good SEO practices in your markup.
Metadata and Markup for Websites
HTML documents can contain metadata stored in the <head> of a document, such as the <title> tag and <meta> tags. Other types of structured metadata can be added to the <body> of the document using markup techniques such as microdata; Schema.org is one example of adding semantic markup to HTML content instead of using <meta> tags.
In HTML 5, <meta> tags are more strictly validated than in previous versions of HTML and XHTML. The The Web Hypertext Application Technology Working Group (WHATWG) maintains a wiki of accepted and proposed values for the meta tag's name attribute.
Be sure to provide unique, meaningful Title entries for all pages and assets (avoid generic or default titles). The <title> tag appears at the top of the browser window, and also as the first line in Google search results. See the Metadata Working Group's Core Title element guidelines for additional suggestions.
Note: Google search results typically display only the first 50-70 characters of your title (as stored in an HTML <title> tag). This has implications for how you format your titles.
Using your Website Name as a Prefix or Suffix
Because Google’s search results limit how much of your title is displayed, you’ll want to consider if you want to emphasize your site name vs. the actual article title. It is common Emory website practice to append the site name after the unique page title. Below is an example showing a site name as a suffix:
Metadata 101 | Emory University
Content-Level Page Title
In the body/content of your document, it is also recommended to have your page title formatted as a <h1> (Heading 1) tag, regardless of its visual styling. It’s recommended that this be the only <h1> used on your page structure, so that it gets greater prominence by the search engine.
You can add keywords to a dedicated <meta> keywords tag, or, make sure you include significant keywords in your page content and within your <meta> description tag. Note that Google and other search engines typically ignore <meta> keywords, but Emory’s local Google Search Appliance (search.emory.edu) does read them, and can utilize them in custom queries. See the Metadata Working Group's Core Subject - Keyword element guidelines for additional suggestions.
The page description displays beneath the title and URL in a search results display in Google.
If you do not supply a dedicated description element, the search engine will often “scrape” the contents of the first paragraph of the page. As noted above, it may be helpful to include significant keywords in the text of your description. Google will display up to the first ~150 characters of your description in the search result listing (but has recently been experimenting with shortening the text it displays). See the Metadata Working Group's Core Description element guidelines for additional suggestions.
Content Management Systems
Standard and customized <meta> tags can be generated via Cascade Server, leveraging metadata entered during content creation. Emory's Standard Template v.2 offers options for configuring some SEO settings and also automatically generates structured schema.org markup in its Bio Pages, to enable richer search result options in Google.
WordPress, which is used in Emory's ScholarBlogs service, is a generally SEO-friendly publishing platform, but may generate little metadata by default. Third-party plugins can be added which will generate richer metadata, inlcuding <meta> tags compatible with Dublin Core and social media environments such as Facebook, Twitter, and Pinterest.