Our Relationship with Internet Search Engines

By Kenning Arlitsch

Kenning Arlitsch, dean of the library at Montana State University, coauthored this blog with Patrick O’Brien, semantic web research director at Montana State University.

Forget about your existing users for a moment; they’re not important to this conversation. If you think your websites and repositories already get a good amount of use then you may not grasp the potential for increased traffic. Your existing online audience is a tiny fraction of the users you could have if they were certain that what they seek exists at your sites. The targets of this conversation are the customers of commercial search engines who submit more than 17 billion search queries per month in America alone [1], and your resources don’t exist for those customers if they can’t find them in search engines. Furthermore, these search engine customers will not be yours until search engines find your sites worthy of referral to their customers. Do we have your attention?

Forgive the hyperbole. Libraries want to reach both existing and potential users, of course, but increasingly the emphasis on assessment demands that we demonstrate use and return on investment. A recent Pew Research Center report estimates that 81% of U.S. adults, and 95% of U.S. teens (about 200 million people) use the Internet [2]. Google handles more than three billion of their searches every day, and Google controls “only” 66% of the domestic search engine market share. Internet users have developed deep and trusting relationships with search engines and have made them the world’s discovery layer. Most library discovery layers in contrast draw little traffic and most of them provide little useful metadata for search engines.

Libraries are generally not very good at optimizing for web search, and that’s because search engine optimization (SEO) is too often left as an afterthought of digital library development. Optimizing a library’s web presence only works when it’s approached holistically and driven strategically, and few if any libraries are currently approaching SEO this way. We’re not just talking about tweaking your webserver configuration, or improving metadata in your digital collections, or using the right keywords in the text of your websites. We’re not talking about gaming search engines, either. We’re talking about an entirely new approach to our relationship with search engines. It’s an approach that understands them to be advocates for their customers, and understands the products and services that they want to deliver to those customers.

Over the next year we’ll be issuing a series of blogs and articles through CLIR designed to explain this relationship and how to address it holistically. We’ll discuss organizing web-based information around specific audiences, focusing on high value activities, the various phases of SEO, and the detail of what needs to be done. We’ll write about broad-stroke issues like internal and external customers, and our funding sources and what kind of ROI they’re looking for. We’ll get into detail about setting baselines, addressing information architecture, implementing semantic web practices, using social media effectively, and making sure that analytics are set up correctly so that we can track users as they jump across our various websites and repositories.

A recent excellent Re:Thinking blog post by Jon Butler, titled “Differentiation,” predicted some possible futures for the Digital Public Library of America. An inspiring effort, the DPLA could become an enormously rich information resource for the public as well as academia. But the likelihood of the approximately 200 million US Internet users visiting the DPLA directly is small. A much greater opportunity lies in ensuring that search engine customers are referred to items in DPLA, but that will only happen if DPLA puts serious consideration into organizational strategies that leverage search engine optimization and semantic web techniques. It’s much easier and more effective to align the organization with these concepts right from the beginning. Without a strategic framework, libraries run the risk of implementing the latest SEO and semantic web fads as disjointed afterthoughts that lack an objective valuation model that stakeholders understand and trust.

We look forward to the conversation that we hope these blog posts and articles will inspire.