Journal of Library Administration, 53:177–188, 2013

Published with license by Taylor & Francis

ISSN: 0193-0826 print / 1540-3564 online
DOI: 10.1080/01930826.2013.853499

posIT

KENNING ARLITSCH, Column Editor

Montana State University, Bozeman, MT, USA

Column Editor’s Note. This JLA column posits that academic libraries
and their services are dominated by information technologies, and that the
success of librarians and professional staff is contingent on their ability to
thrive in this technology-rich environment. The column will appear in odd-
numbered issues of the journal, and will delve into all aspects of library-
related information technologies and knowledge management used to con-
nect users to information resources, including data preparation, discovery,
delivery and preservation. Prospective authors are invited to submit articles
for this column to the editor at [email protected].

MANAGING SEARCH ENGINE OPTIMIZATION:

AN INTRODUCTION FOR LIBRARY ADMINISTRATORS

KENNING ARLITSCH, PATRICK OBRIEN, and BRIAN ROSSMANN

Montana State University Library, Bozeman, MT, USA

INTRODUCTION

Libraries collectively spend millions of dollars each year creating Web sites
and digital repositories, but optimizing for search engines is too often an
afterthought and makes digital library use a fraction of what it could be.
Even libraries that do take search engine optimization (SEO) into account
tend to relegate its practice to a few individuals in IT departments, often
resulting in a disjointed and unproductive program that is viewed as a
limited domain rather than a primary concern of the entire organization. SEO

© Kenning Arlitsch, Patrick OBrien, and Brian Rossmann

Address correspondence to Kenning Arlitsch, Dean of the Library, Montana State Univer-
sity, P.O. Box 173320, Bozeman, MT 59717-3320, USA. E-mail: [email protected]

177
178 K. Arlitsch et al.

deserves cabinet-level attention because of its potential to help libraries reach

more users and derive assessment data. This article is aimed at giving library
administrators a high-level perspective of SEO so that they may be equipped
to ask the right questions of their technical staff, software vendors, and con-
tent suppliers. It stresses the importance of aligning SEO with institutional
priorities and integrating it into the strategic plan. SEO is most effective
when it is an organizational priority and when it is understood and driven
by administrative teams.
Libraries and their funders make “significant investments in the digitiza-
tion” (Maron & Pickle, 2013) of special and other collections each year. The
entire scope of a digital library includes far more than scanning objects and
loading them into a database, however, and the cost of developing and sus-
taining a digital library requires substantial funding. Whether costs are borne
by internal sources or from external grants and donations, funding providers
have become more interested in the value and use of their investments.
Libraries would do well to emphasize Internet search engines over their
own Web sites as a means of bringing users to digital libraries. The concept
of “inside-out” library resources (Dempsey, 2010) emphasizes the value of
search-engine discoverability, overriding the more traditional “outside-in”
expectation that users will begin research in the library. Americans submit
approximately 20 billion search queries to Google each month, and Google
represents “only” 66% of the domestic search engine market share (comScore,
2013). Most library Web sites simply do not draw much direct traffic of their
own (DeRosa et al., 2010) and it is a much better investment of time and
energy to establish good relationships with search engines and social media
sites than it is to try to draw users directly to library Web sites. Establishing
good relationships means ensuring that search engine crawlers can navigate
through Web sites and repositories without running into design, metadata
and systems barriers that negatively impact index ratios or rankings in search
engine results pages (SERP). Search engines must find machine-readable
and comprehensible text to index, and they must be convinced that their
customers will have a worthwhile experience at a library site. Slow Web
sites, over-use of graphics, dead links, and poor-quality content or redundant
metadata are factors that may contribute to a search engine’s decision to send
its customers elsewhere.
Some may dismiss the practice of SEO as “gaming” search engines, but
it’s only gaming if it is done poorly and with malicious intent. Practiced
according to the guidelines and tools provided by the major search engines
(Microsoft, Inc., 2013; Google, 2013), SEO helps libraries establish better re-
lationships with search engines, and thus with the customers of those search
engines. Libraries gain users, perhaps many thousands of users, when search
engines decide to send their customers to library sites. Previous research by
the authors at the University of Utah’s J. Willard Marriot Library produced a
500% increase in referrals from Google that resulted in a 132% increase in
 posIT 179

visits to digital collections (Arlitsch & OBrien, 2013) by implementing good

SEO practices.

SEO ACROSS THE ORGANIZATION

Search engines index digital files and their metadata, and therefore SEO is
naturally considered to be the concern of people who manage technology.
While it is true that effective SEO affects the layers of technology that manage
and deliver digital content, it really is the concern of anyone whose work is
represented on the Internet because it impacts how accessible that work is to
the intended audiences. SEO is not something that should be left exclusively
to an organization’s IT department: “an IT department should not be left
to make, often by default, the choices that determine the impact of IT on
a company’s business strategy” (Ross & Weill, 2002). Nearly every library
employee has some interest or influence in the content and the services that
a library makes available on the Web, and that makes for many stakeholders.
Both technology and people should be driven by the strategic goals of the
organization.

STRATEGIC PLANNING

Just as it takes many people to make print collections accessible to the public,
SEO practiced broadly will help ensure that items in the digital library will be
accessible as well. Some SEO solutions are technical in nature, but all aspects
of SEO require communication, management, and coordination to ensure
that people are working together to achieve common goals. Including SEO
goals and objectives in a library’s strategic plan elevates its importance and
creates a better chance that SEO techniques will be applied effectively and
its results measured accurately. Driving SEO strategically will help ensure
its success, and will set goals and objectives that allow administrators to
gather accurate use statistics so that reporting becomes an integral part of
the assessment process. Data gathered from SEO analysis tools can help
make course corrections, communicate a narrative about what the library is
trying to achieve for its users, and enhance library assessment.

Strategic Alignment With the Institution’s Mission

The success rate of a library’s SEO efforts improve dramatically when they are
aligned with the strategic goals of the institution. A carefully communicated
and well-executed institutional repository (IR) strategy can deliver value rec-
ognized by university administrators and its faculty. For example:
180 K. Arlitsch et al.

Ensure staff and faculty understand the

strategic importance of SEO efforts
Montana State University Strategic Plan

Objec ve: Enhance University Metric: MSU will increase grant-sponsored investment
infrastructure in support in centers, core facili es and resources to expand state of the art
of research, discovery and tools, exper se, and opportuni es for research and crea ve
crea ve ac vi es ac vi es

• Develop ins tu onal repository of intellectual output of campus

MSU Library Strategic Plan SEO Program Ac vi es

Objec ve: Improve Library Metric: Increase indexing rates of MSU content to improve
discovery layers and find- access for researchers worldwide
ability of library content
• Create dashboard for analy cal repor ng
• Establish a baseline for Google’s indexing of Library web content
for search engine op miza on
• Establish baseline for Google Scholar’s indexing of ScholarWorks

FIGURE 1 Strategic Plan Example (Montana State University, 2013). (Color figure available
online).

1. Create a showcase for research and scholarship produced at the institution

that can be leveraged for fundraising purposes with legislatures, alumni,
granting agencies, and local communities.
2. Increase faculty citation rates by making publications and data sets easily
accessible to a wider audience. (Piwowar, Day, & Fridsma, 2007) Increased
citation rates could, in turn, raise university rankings. (The Times Higher
Education, 2010)
3. Provide a hedge against publisher price inflation that continues to exceed
the Consumer Price index. (Bosch, Henderson, & Klusendorf, 2013)

Framing the value this way at Montana State University has resulted
in the Library’s IR efforts being elevated to a line item in the University’s
academic strategic plan and has in turn helped to drive part of the Library’s
strategic plan (see Figure 1). It is also leading to easier conversations with
MSU faculty about why they should deposit their papers in the University’s IR
and why faculty should consider adopting an open access publishing policy.
A well-executed IR strategy produces content that can be found by
search engines, and in particular by academic engines like Google Scholar
and Microsoft’s Academic Search. These search engines will only index meta-
data that they can parse so that they can deliver citations in whatever styles
 posIT 181

their users require, and that means citation data in IR’s must be offered in
discrete fields. Earlier research has demonstrated that library-developed IR’s
often fail in this regard because they tend to lump citation data into single
fields, which is not machine-comprehensible (Arlitsch & OBrien, 2012).
Libraries can use SEO strategy to think about adding value beyond their
traditional roles and boundaries. For example, using knowledge derived
from SEO research they can help develop the metadata schema used to store
citation information in a campus faculty activity database used for evaluations
of faculty performance. Engaging at this level allows libraries to influence the
alignment between the schema employed in faculty performance reviews,
in promotion and tenure reviews and the IR metadata. This alignment can
increase faculty participation by incentivizing them to deposit published
research, reduce effort needed to populate IR citation metadata, improve
IR data quality, and ensure scholarly output is discoverable by the search
engines used by academic researchers.
Library administrators must have metrics defined and tools in place to
evaluate the results of their SEO efforts and objectively demonstrate the
value they are delivering to their stakeholders. The first step is to determine
what stakeholders value and what can be objectively measured. The best
metric (within the technical reach of most libraries) for evaluating the value
of an IR is the number of IR PDF views or downloads by a target audience.
Visitors being referred to the IR by Google Scholar, or who have an IP
address that resolves to a .edu domain, are more likely to be academics than
typical visitors from Facebook or Yahoo!. Academic visitors are more likely
to formally cite papers they choose to view or download as PDF files. It is
worth paying attention to SEO areas that will help increase faculty citation
rates.

SETTING EXPECTATIONS AND USING METRICS

Statistics have long been a part of our business. Libraries have always counted
the size of their collections—the number of items borrowed, interlibrary
loans, expenditures—and they have always used those numbers to create
comparisons to other libraries. SEO continues this tradition of measurement,
with additional complexity and some variation in the vocabulary. Gate counts
for physical visits are still needed, but counting online visitors is at least as
important, and can actually tell administrators a whole lot more about their
users (in anonymous terms) than traditional counts.
Libraries must move beyond counting basic page views and visits to
individual Web sites. They must break down silos and look at the entire
user click stream across all their Web properties to create metrics their in-
ternal and external stakeholders value. This is an important concept: it is
not enough to track user behavior on a single Web server because that
182 K. Arlitsch et al.

provides only a partial picture. One physical server may contain the library’s
Web site, another server hosts a digital repository, and a third may serve
streaming video to users, but a user who peruses a particular digital collec-
tion may touch all three servers in a visit. A single machine can also host
multiple Web servers with distinct URLs established to promote a featured
collection or program. While these servers are probably all in the organiza-
tion’s domain, reporting tools must be set up appropriately to track users
as they move from one to another. (See the section on “Traditional SEO
Components.”)
A dashboard for each digital collection could help the library’s leader-
ship team have discussions with current and future donors, as well as help
collection managers who want to demonstrate value to grant providers and
collaboration partners. Dashboard metrics can provide objective data that
support a dialogue with the university’s administration about the value of
the IR and how individual faculty members, departments, and colleges can
better utilize it.

HOW SEARCH ENGINES FIND WEB SITES

Most Web sites and digital repositories are driven by databases; very few are
still collections of static HTML pages. Some databases are relational, meaning
that the data reside in table structures, and some are flat text-based files that
incorporate markup like XML to provide structure for the data. In either
case, Web pages from database-driven sites are constructed dynamically, or
“on the fly.” A programmed script assembles various components from the
database to generate a page at the moment that it is “called” by the user who
clicks on a hyperlink or types a URL in a Web browser. Data components may
include the object itself (a digitized photograph, for example), the metadata
associated with it, and the HTML template that will include a header and
footer along with other design elements that help make a page useful and
attractive.
Search engines find, harvest, and index Web sites through the use of
programs called “crawlers,” “robots,” or “spiders.” A search engine crawler
does not actually crawl through a database. It wants to see the compiled
page just as the user sees it, so it follows the link for each object in the
database and triggers the generation of a page for each. It then harvests the
text that is generated for that page before moving on to the next link. It is
at this crucial juncture, when the page is displayed, that all the text a library
hopes will be indexed by the search engine must be present, prioritized,
and accessible to the crawler. It’s crucial to note that search engine crawlers
are not unlike visually impaired users in that they “can’t read text in images,
can’t interpret JavaScript or applets, and can’t ‘view’ many other kinds of
multimedia content” (Hagans, 2005).
 posIT 183

TRADITIONAL SEO COMPONENTS

We distinguish “traditional” SEO from more recent “semantic” SEO develop-

ments because the traditional components form a foundational relationship
that allows search engines to find, harvest, and index Web sites and digi-
tal repositories. Without that foundation any additional value provided by
semantic Web techniques is marginal at best, and irrelevant in most cases.
Semantic SEO does hold enormous potential for bringing more accurate and
relevant search results to the user as it helps set context and meaning for
search engines, but this topic is outside the scope of this article.
Traditional SEO involves identifying the requirements of search engines,
establishing communication channels to assure goals are being met, and
monitoring the relationship for disruptive changes that can occur on either
side. The basic goals of SEO are to have Web content included in a search
engine’s index (indexing ratio) and to rise to the top of SERP (search engine
results page), otherwise known as “rank” for searches conducted by target
audiences. Specific traditional SEO actions that help create that foundational
relationship with search engines include:

1. Develop internal inventories of the organization’s logical and physical

domains.
a. Logical domains are the domains, sub-domains, and sub-directories
used to organize Web content and are typically aligned with the people
responsible for the content in a given collection or repository.
b. Physical domains describe the physical servers that host the digital
library’s logical domains. A server may host multiple logical domains,
such as an IR, a departmental Web site, or a digitized collection.
2. Designing Web sites and navigational paths that pose no barriers to
crawlers. Graphics can make a Web site attractive, but their overuse tends
to increase page load time and provides little or no text for crawlers to
index. Complex internal link structures or labyrinthine paths to objects
reduce crawler efficiency and limit the total amount of content they can
find and index on the site. Crawler efficiency can weigh heavily in how a
Web site and its content is ranked in SERPs.
3. Serving indexable text and metadata. Search engines do not tolerate place-
ment of invisible text or keyword stuffing, and redundant or repetitive
metadata lead crawlers to think that the objects being described are iden-
tical.
4. Configuring servers and software to deliver results quickly to users and
search engine crawlers. Slow server response, dead links, and failure to
communicate location changes and downtime through an accepted set of
messages is viewed as a bad user experience.
5. Setting up Webmaster Tools and analytics software for monitoring and
assessment purposes. Webmaster Tools provide the feedback loop from
184 K. Arlitsch et al.

search engines as they try to crawl a library’s sites and repositories, letting
them know what kind of problems they encounter. Analytics software
such as Google Analytics provides a wealth of information about a library’s
visitors.

ASSESSMENT

Accountability has been important in higher education for many decades

(Marrs, 2009). Now, however, in times of curtailing budgets and sparse re-
sources, colleges and universities are being more closely scrutinized than
ever. They are being expected to demonstrate that they are managing their
budgets wisely, provide good value to students and faculty, and have a pos-
itive impact on student success. Accordingly, libraries are increasingly de-
veloping cultures of assessment (Farkas, 2013). “Not only does assessment
give librarians a venue for communicating with stakeholders, it determines
“the fit” between institutional mission and achieved outcomes, articulates
effectiveness, fosters improvement, increases efficiency, and demonstrates
accountability” (Oakleaf, 2010). As libraries invest more of their resources
creating a digital presence, it is important that the findability and use of
digital collections becomes a part of libraries’ assessment efforts so that they
can demonstrate the value of their digital presence. “Institutional assessment
efforts should not be concerned about valuing what can be measured, but
instead about measuring what is valued” (Banta, 1996). Assessment librarians
and administrators involved in implementing and monitoring SEO can gather
more accurate data on the use of digital collections. That SEO data then be-
comes another metric on the library’s scorecard in the assessment cycle.
In addition to incorporating SEO into a library’s assessment cycle, it
is important to continually assess SEO activities themselves. Assessment li-
brarians, if they are provided with access to Webmaster Tools, and if they
incorporate SEO assessment into their workflow, can monitor changes in
the usage patterns of digital collections. When anomalies are observed, they
may engage administrators, digital initiatives librarians, and IT staff to iden-
tify solutions. “Assessment “strives to know . . . what is” and then uses that
information to change the status quo” (Keeling & International Center for
Student Success and Institutional Accountability, 2008).
Incorporating SEO into a library’s assessment program will pay divi-
dends both with respect to improving SEO itself and allowing libraries to
measure and demonstrate increased value to library stakeholders.

WEBMASTER TOOLS

Webmaster Tools are offered both by Google and Bing to assist technical
teams in identifying and addressing issues that will help sites perform better
 posIT 185

in search results. The search engines’ crawlers report the following infor-
mation and more about each site that is verified via the Webmaster Tools
product:

• Identify which parts of the site pose problems for crawlers

• Notify the search engine of new or revised XML sitemaps
• Generate and analyze the robots.txt files
• Remove URLs from the crawl when they no longer exist
• Identify issues with page titles and meta tags
• Identify the top search terms used to reach sites
• Review pages as the search engine crawler would see them
• Provide notifications of any quality guideline violations
• Provide statistics

GOOGLE ANALYTICS

Numerous commercial and free Web site analysis software tools exist. Some
of them analyze Web server logs, while others utilize page tagging techniques
that embed code into each HTML page of a Web site to set and track “cook-
ies.” With page tagging, the code sends a message to a third party system
each time a page is viewed, and it compiles visitor information concerning
sessions, page views and traffic sources (e.g., referring sites, search terms,
etc.). There are advantages and disadvantages to both methods. Google An-
alytics (GA) utilizes page tagging and is suggested for its ease of use, zero
cost, excellent support, and power. If configured properly, GA can provide
data about a library’s logical domain that will help administrators understand
where visitors are coming from and what they are looking for. Aside from
creating a Google Account, configuring the GA product and embedding a
bit of code in each Web page HTML header, there is no further overhead for
basic reporting.
Google Analytics provides powerful and anonymous data about visitors
to Web sites, information about their behavior while they are visiting, and
identifies the tools they use to view the site. It can also help to quickly
troubleshoot problems on the site. Administrators may learn that certain
page titles are inaccurate or could be written more descriptively, or that
some users have bookmarked and are still visiting an obsolete page that has
never been deleted from the server. GA can provide the following pieces of
information, among others, about a Web site and its visitors:

• Number of unique and returning visitors

• Search terms used to reach a specific page
• Operating systems and browsers used by visitors
186 K. Arlitsch et al.

• Whether a mobile device was used for the visit and, if so, what type
• What pages were landed on and from which ones users exited
• Most viewed pages, the order in which they were viewed, and for how
long
• The countries and cities from which visitors originate
• How long pages take to load (as mentioned earlier, slow-loading pages
can cause search engines to send their customers elsewhere)

Google Analytics can even establish and track goals. Administrators

might want to see which academic papers were downloaded from the in-
stitutional repository, and if configured properly, GA can provide results by
college, department, and author. This can help libraries inform faculty about
the frequency with which their papers are accessed, and perhaps will help
generate support for increased faculty IR participation.
Cross-domain tracking helps track users when they are referred from
one logical domain to another. For many digital repositories, the informa-
tion describing a collection is contained in one logical domain (e.g., a Web
site located at lib.montana.edu), while the actual objects in the collection
are located within a different logical domain (e.g., an IR located at scholar-
works.montana.edu). Without cross-domain tracking set up, all that can be
known is that visitors were referred from one logical domain to the other.
The critical information about the links and search terms used that led to
viewing or downloading the object within the IR will not be passed from the
visitors’ entry point. In other words, administrators won’t know why visitors
were referred to the IR, or how they got there. Cooperation among domains
will help identify and influence key elements concerning visibility of the
library’s content within SERPs.
The information that Google Analytics provides creates a feedback loop
that can help repository managers improve the user experience, and to
identify what is working well and what is not. It can help improve the text
describing a given page, resulting in a better fit with the traffic sources and
search queries that deliver users to the library’s sites.

SUMMARY

Digital libraries suffer from lack of visitation and use because few libraries are
proactive and strategic about search engine optimization. A well-executed
SEO strategy will connect users to the information they seek and can
play a significant role in increasing citation rates of academic research.
Library administrators can derive valuable data about their organizations
through a carefully managed SEO program that ensures Web sites and dig-
ital object metadata are harvestable and comprehensible by search engine
crawlers. SEO affects many areas of an organization, and in turn there are
 posIT 187

numerous people who play roles in its successful practice. Those roles are
best driven from a strategic plan that aligns with institutional goals and is
driven by the library’s leadership. A wealth of free software exists to help
troubleshoot problems that search engine crawlers encounter when they try
to harvest Web sites and digital repositories, and Web analytics software like
Google Analytics can provide rich data about visitor behavior that can help
organizations make course corrections for better service.

REFERENCES

Arlitsch, K., & OBrien, P. S. (2012). Invisible institutional repositories: Addressing

the low indexing ratios of IRs in Google Scholar. Library Hi Tech, 30(1), 60–81.
doi:10.1108/07378831211213210
Arlitsch, K., & OBrien, P. S. (2013). Improving the visibility and use of digital repos-
itories through SEO. Chicago: ALA TechSource, an imprint of the American Li-
brary Association. Retrieved from http://search.ebscohost.com/login.aspx?direct
=true&scope=site&db=nlebk&db=nlabk&AN =578551
Banta, T. W. (1996). Assessment in practice: Putting principles to work on college
campuses. San Francisco: Jossey-Bass.
Bosch, S., Henderson, K., & Klusendorf, H. (2013, April 25). The winds of
change: Periodical price survey 2013. Library Journal. Retrieved from http://lj.
libraryjournal.com/2013/04/publishing/the-winds-of-change-periodicals-price-
survey-2013/
comScore. (2013, July 12). comScore releases June 2013 U.S. search engine rank-
ings. comScore, Inc. Retrieved from http://www.comscore.com/Insights/Press_
Releases/2013/7/comScore_Releases_June_2013_U.S._Search_Engine_Rankings
Dempsey, L. (2010, January 11). Outside-in and inside-out. Lorcan Dempsey’s weblog
on libraries, services and networks. Retrieved from http://orweblog.oclc.org/
archives/002047.html
DeRosa, C., Cantrell, J., Carlson, M., Gallagher, P., Hawk, J., & Sturtz, C. (2010).
Perceptions of libraries, 2010: Context and community (p. 108). OCLC, Inc.
Retrieved from http://www.oclc.org/reports/2010perceptions.htm
Farkas, M. G. (2013). Building and sustaining a culture of assessment: best practices
for change leadership. Reference Services Review, 41(1), 13–31. doi:10.1108/
00907321311300857
Google. (2013, July 17). Webmasters—Google. Retrieved from https://www.
google.com/webmasters/
Hagans, A. (2005, November 8). High accessibility is effective search engine
optimization. A list apart, 207. Retrieved from http://www.alistapart.com/
articles/accessibilityseo
Keeling, R. P., & International Center for Student Success and Institutional Account-
ability. (2008). Assessment reconsidered: Institutional effectiveness for student
success. Washington, DC: ICSSIA.
Maron, N. L., & Pickle, S. (2013, February 21). Appraising our digital investment:
Sustainability of digital special collections in libraries. Association of Research
188 K. Arlitsch et al.

Libraries and Ithaka S+R. Retrieved from http://www.arl.org/storage/

documents/publications/digitizing-special-collections-report-21feb13.pdf
Marrs, H. (2009). Perceptions of college faculty regarding outcomes assessment.
IEJLL: International Electronic Journal for Leadership in Learning, 13(2). Re-
trieved from http://iejll.synergiesprairies.ca/iejll/index.php/ijll/article/view/688
Microsoft, Inc. (2013, July 17). Webmaster Guidelines—Bing Webmaster Tools.
Retrieved from http://www.bing.com/webmaster/help/webmaster-guidelines-
30fba23a
Montana State University. (2013). MSU Library strategic plan 2013. Montana
State University Library. Retrieved from http://www.lib.montana.edu/about/
msu_library_strategic_plan.pdf
Oakleaf, M. (2010). The value of academic libraries: A comprehensive research
review and report. Association of College and Research Libraries. Retrieved
from http://www.ala.org/acrl/sites/ala.org.acrl/files/content/issues/value
Piwowar, H. A., Day, R. S., & Fridsma, D. B. (2007). Sharing detailed re-
search data is associated with increased citation rate. PLoS ONE, 2(3), e308.
doi:10.1371/journal.pone.0000308
Ross, J. W., & Weill, P. (2002). Six IT decisions your IT people shouldn’t make.
Harvard Business Review, 80(11), 84–95.
The Times Higher Education. (2010). The Times Higher Education World University
Rankings 2010–2011. Retrieved from http://www.timeshighereducation.co.uk/
world-university-rankings/
Copyright of Journal of Library Administration is the property of Taylor & Francis Ltd and its
content may not be copied or emailed to multiple sites or posted to a listserv without the
copyright holder's express written permission. However, users may print, download, or email
articles for individual use.