NIH Web Authors Group (WAG)
A Guide for Web Managers at the National Institutes of Health
Use the following tips to improve and protect the findability of your website, furthering NIH efforts to support biomedical discoveries, disseminate health information and demonstrate a continued high return on public investment in research.
- Foster an environment that values findability and search optimization by supporting staff efforts to incorporate accurate, descriptive and unique metadata in nih.gov domain web pages. Communicate the importance of these efforts to NIH Web managers, designers and content producers. Although metadata are "hidden," no Web is complete without them.
- Page titles are the headlines of search results, and should be treated accordingly to capture reader interest. Review page title tags to ensure they are accurate and unique. Multiple template-based pages with repetitive titles are often the norm. Don't rely on default or generic titles; either change them to reflect page content or ask content owners to review their pages to improve title quality. Titles should be relatively brief but descriptive. Managers of large sites should start working with top-level and most-visited pages and then tackle other pages at a regular but realistic pace.
- Description tags are important in search results too, giving potential site visitors more specifics about what they will encounter if they visit a page. Use description tags to their full advantage to provide an overview of page content in a few sentences. As with title tags, many pages in the nih.gov domain have descriptions tags that are generic or empty rather than providing useful information. Either change them or ask content owners for help. Without these tags, search engines are likely to use the first words on a page in place of the description, which may be inconsistent or confusing to potential visitors. Content management system administrators should make sure system workflow prompts content authors to provide useful descriptions for each page. Managers of large sites should start working with top-level and most-visited pages first and then tackle other pages at a regular but realistic pace.
- Many search engines will use the first paragraph on a Web page to represent a summary of the content that follows, especially if the page is missing a description tag. Increase the likelihood that your most relevant pages will appear in search results by including important keywords or phrases in the first sentence or two of each page.
- Use clear, meaningful descriptions for images and links, in caption or link text as well as ALT or TITLE text. For example, avoid generic text in these tags such as “image of building.” Instead use specific text such as “Johnson Neuroscience Laboratory.” If a hyperlink target or result isn't clear from the link text alone, consider rewriting the link text or adding text in the A link's TITLE attribute. These practices not only will make your Web pages easier for search engines to index accurately, but also will make them more accessible to site visitors, particularly those using assistive technology.
- Use HTML site maps and subject indexes. They not only increase findability for human visitors, they also help search engine crawlers index Web sites. Each main site in the nih.gov domain should have an HTML site map that includes links to subsite "children."
- Create an XML sitemap corresponding to each HTML sitemap to further improve search results rankings.
- Add links to other NIH websites, where appropriate, and ask other relevant NIH sites to link to your content. When possible, ask other credible organizations such as grantee institutions to link to your content.
- Regularly evaluate your search optimization efforts using Web traffic analyses, search query reports and by comparing public search rankings of key indicator pages on your site.
- Review your current page ranks in the NIH site search. If you consider certain pages to be an important source of information about a particular area and their rank is low (or if they don't show up in NIH site search results at all), contact the NIH/OD/OCPL Online Information Branch at firstname.lastname@example.org.
Google. Matt Cutts Discusses Snippets. November 20, 2007. Available at http://www.youtube.com/watch?v=vS1Mw1Adrk0.
Google. Search Engine Optimization Starter Guide. September 30, 2010. Available at http://www.google.com/webmasters/docs/search-engine-optimization-starter-guide.pdf
Google, Microsoft and Yahoo! XML Sitemap 0.9 Protocol. Sitemaps.org. February 27, 2008. Available at http://www.sitemaps.org/protocol.php
Hillman, Diane. Using Dublin Core. Dublin Core Metadata Initiative. November 7, 2005. Available at http://dublincore.org/documents/usageguide
Morville, Peter and Callender, Jeffery. Search Patterns: Design for Discovery. February 2, 2010. O’Reilly Media, Sebastopol, CA. See related information at Morville’s website, http://findability.org
Nielsen, Jakob. Top 10 Information Architecture Mistakes. Jakob Nielsen’s Alertbox. May 11, 2009. Available at http://www.useit.com/alertbox/ia-mistakes.html