Recommendations
Information Architecture
Improve clarity of labels and avoid “junk drawer” categories. These do not guide people toward the things inside them; they provide storage, but not discovery. “Resources” and “Services” are convenient catch-all words that can hold a lot of different topics, but they can mean so many different things that they don’t really guide users toward what’s within them.
Discontinue having one central blogroll and, instead, establish subject-based pages that display blog posts on relevant topics. Topics should have demonstrated interest from OPL blog readers, relevance to OPL collections, and consistent expertise among staff. We recommend pages with blog posts on the following topics: Oakland history, African American history and culture in the Bay Area, Using the library from home, Children's Literature, Kids' Activities, and School.
Content Strategy
Reduce overall reading level, and shorten text. Aim for maximum 6th grade reading level. An optimal website reading level is 8th, but Oakland serves many adults with low literacy or who are English language learners. A lower level than average will serve a broader group of OPL’s user base.
Limit the number of “lists of links” pages OPL maintains. Keep only those that are crucial, get significant traffic, and are backed by staff with expertise and OPL collections, and purge the rest. Examples: Business Reference Online, Job Search Resources. These are not original content, do not perform well statistically, require upkeep, and may be a reflection of an older model of library service and staffing. Many of these pages on OPL’s current site are outdated and have dead links, suggesting OPL does not have the staff time or resources to consistently maintain this kind of resource for the public. A best practice might be to select a small number of high statistical performers, such as the Teen “Aid and Scholarships” links, and keep only those, allowing staff to target their limited time and energy on links that have demonstrated use.
Create “parallel pages” in Spanish and Chinese. Translate all basic information on using the library and FAQs; then, continue expanding Spanish and Chinese original content. At the moment, OPL’s website really only offers an English experience. There are Spanish and Chinese versions of the site, but content is very minimal, and their extremely low use speaks to their limited usefulness. OPL has purchased Spanish and Chinese BiblioCommons modules, and has a recent influx of staff being paid to write content in those languages. We recommend OPL focus on creating original content in Spanish and Chinese, while also determining which basic content will be translated and placing that on its own page, not at the bottoms of English pages.
Create and follow a style guide for a consistent look throughout the site. Almost every test subject complained or reported confusion at the inconsistency of OPL’s site.
Determine an overall image strategy. User testing consistently revealed that subjects, especially children, liked photos.
Use metadata. Currently no metadata is applied to pages on OPL’s website. Employ this on the new site for better SEO.
Future Research
Opt into collecting aggregate data in Google Analytics. Aggregate data does not create the same privacy concerns as data that is attached to personally identifiable information. Many of the databases OPL subscribes to collect aggregate data. Opting into “Demographics” in Google Analytics would allow additional insight into OPL's website users without unreasonably compromising privacy.
Implement site search and use search logs to inform future IA iterations. BiblioCommons offers site search, so this will be available.
Continued A/B testing in order to further refine the prototyped hierarchy and navigation with data acquired. We recommend that special attention be paid to future participants' socioeconomics, race, age, and first-language so Oakland's demographics are accurately represented. We also recommend user-testing be performed on individuals who live in OPL's service area who are not currently patrons. This research can serve to develop a picture of those that OPL would like to reach, aiding OPL in furthering their stated mission and goals.
Administer large-scale survey of Alameda County residents. This will provide data of a much wider base of individuals of whom OPL desires to serve in the future.
Consider gathering racial demographics of current library users. This will serve as a baseline to determine the success of OPL's service to diverse communities. We recognize that a majority of those whom participated in user testing were middle-aged white women and believe that diversifying the user pool would make for more accurate data and results.
