Portfolio of Chiara Fox

Comparative analysis for large catalog and e-commerce company

January 2001, Argus Associates

I was lead information architect on this comparative analysis project. We examined the client web site, as well as four other sites. We analyzed the information architecture of these sites in terms of their global, local, contextual and supplemental navigation, organization, labeling, search, shopping experience, overall site experience/ usability, and special features that leverage the information architecture. The sites we evaluated are all large, e-commerce sites that sell many types of products and multiple brands.

An excerpt from the final comparative analysis report is available.

I developed a series of tables and matrixes to help in the evaluation of the sites. The tables were used to rank the sites from best implementation to worse. In the final report the client was able to see where they stood in comparison to their competitors. Recommendations were also included in the final report.

An example of the overall ranking table is available.

ACIA Information Architecture Guide

December 2000 - January 2001, Argus Associates

In December 2000 I began working with Kat Hagedorn to update the Information Architecture Guide. The guide hadn't been updated in almost a year, so there was a lot of work to be done. By the time we finished, the guide went from approximately 90 entries to about 240.

The project was broken down into six parts:

  • Collection Development: We gathered all the resources we could think of to add to the Guide. We created a style guide to ensure our data entry would be consistent. The style guide included everything from the fields to be included to punctuation and capitalization within a specific field.
  • Data Entry: We entered the resources we found into the database. In addition to author, title and URL information, we also assigned uncontrolled keywords to each record. We updated the records that were already in the database so they would conform to the new style guide.
  • Asking Users: Every information architecture project must look at its users, and this project was no different. We created an online survey and asked the members of the SIGIA community to participate. The survey asked users what they did and did not like about the current Guide, how they would improve the Guide, and what types of resources they wanted to see included.
  • Revision of Indexing and Summary Creation: After all of the resources were added, Kat and I went through each record. We made sure that each record conformed to the style guide, created summaries for each item and ranked them on a scale of 1-5.
  • Controlled Vocabulary Development: While going through each record as a team, Kat and I developed the controlled vocabulary of keywords. After approximately 50-100 records we had most of our vocabulary fleshed out. It included scope notes for confusing terms (e.g., user experience design), see references, and broader and shallower terms.
  • Implementation: There was no content management software for this phase of the IA Guide revision, so Kat and I had to hand code the HTML for each page. We developed workflow policies documenting each step that was required in the process. We also created a "dreamboat" or wish list of the features and functions we wanted in a content management system.

Strategy and recommendations for an international, multi-billion dollar industrial, manufacturing and residential electrical power control and distribution products company

July 2000 - October 2000, Argus Associates

This project was a basic strategy and recommendations. It used the core pieces of the Argus methodology, including content analysis, classification scheme analysis, search log analysis, comparative analysis, opinion leader interviews, and user testing. I had a lead role in the following pieces of the methodology:

  • Content Analysis: I took a representative sample of the content available on the web site. I then examined the content for similarities, differences and relationships between the different content types. The information I learned from this shaped and informed all other phases of the project.
  • Comparative Analysis: I examined eleven sites for the comparative analysis: the client site, four competitors, two distributors, and four sites of information architecture excellence. Each site was examined in terms of their information architecture which included: global, local, contextual and supplemental navigation, search, organization, labeling, content and features such as personalization/customization and shopping.
  • Opinion Leader Interviews: We interviewed ten opinion leaders and stakeholders in the client organization, including the CEO. This was done to better understand the business context of the site. We asked each interviewee about their relationship to the web site, what they thought was good and bad about the current site and their vision of what the site should be. I wrote a separate report summarizing our findings.
  • User Testing: Two separate sessions of user testing were done for this project. The first session was exploratory in nature and included card sorting and task analysis. We used this session to learn what users are most interested in on the site and what they like and do not like about it. The tasks were performed on the client site as well as a competitor site. The second session tested ideas we had for recommendations. This session also included card sorting. We also tested proposed labeling and prototypes of new page designs. We based our final recommendations on what we learned in this session.

    An example of one of the prototypes used in the second session is available.

  • Development of Final Recommendations and Report: The final report was a culmination of all of the research we had done in the project. It included recommendations on how the information architecture of the client site could be improved in terms of organization, navigation, labeling, and search. There were also recommendations on content management, including suggested attributes, or meta data, to be used to describe the content and the types of controlled vocabularies needed to create the attribute values. The final report included blueprints, wireframes, as well as attribute tables and other diagrams.

User testing and recommendations for an international telecommunications company

June 2000 - July 2000, Argus Associates

This was my first project at Argus Associates. The project was to examine the hierarchy and organization of part of the intranet for an international telecommunications company. I joined the project half way through for the user testing and development of the final recommendations.

I assisted in designing the tasks that we did with users of the intranet. We did a series of open and closed card sorting, as well as observation of tasks the users usually performed on the site. We used the information we gathered in these sessions to inform our final recommendations.

An example of the affinity diagrams we used in our analysis of the card sorting data we gathered is available.

I also helped to develop the recommendations for improving the hierarchy of the site. The site was very deep and very braod, meaning users had to navigate many pages before they got to the content they were interested in. The new hierarchy bubbled up the important content so it wasn't so deep within the site.

The blueprint of the proposed site hierarchy that I created is available.

Rowland Interlibrary Loan Database

January 2000 - May 2000, Rowland Institute for Science

As Technical Services Librarian, it was my responsibility to order, receive, and track the interlibrary loans of books and articles that our scientists needed. In addition to keeping track of when an article arrived and payment, we also had to track how many times we requested an article from a journal in case we had to pay copyright royalties. Until the Interlibrary Loan Database, this was all done by hand, on a series of index cards.

I created a relational FileMaker Pro database to aid with this. The database was a collection of three separate files. The first housed information about leading libraries (e.g., phone number, address, charges). The second contained a record for each journal the Institute had ever copied a paper from. It contained information such as what library owned it, title, previous title and years of publication. The third file contained the information about each request (e.g., author, title, scientist who requested it, date of article, amount paid).

Through a series of scripts, it is possible to view the status of a request, including if it was received and payment information. A view can be generated that displays all of the request that require copyright royalties to be paid.

Kennedy Collection Image Database

January 1999 - April 1999, Rowland Institute for Science and Simmons College

In the spring of 1999 as part of the Master of Library Science program at Simmons College, I took an indexing and abstracting class. Our semester-long project was to create an index that could be used in a real-life situation. Working with a partner, I created an electronic index to a collection of photographs by Clarence Kennedy housed at the archives of the Rowland Institute for Science. The photographs were taken in the 1920's and 1930's of different works of art in Europe.

In the collection there are over 4700 negatives and prints housed in over 155 boxes. Until this project, the only access to the collection was through a paper finding aid. The project was to create a working prototype of a FileMaker Pro database that would allow greater, and more complete access to the collection.

For the project we took a representative sample of 100 photographs. Each photograph was indexed for a variety of elements such as name, location of original artwork, era of original, and subject keywords. A specialized controlled vocabulary was developed for the subject field from the Art & Architecture Thesaurus, ICONCLASS, the Library of Congress Thesaurus of Graphic Materials and the Library of Congress Authority Files. Each item in the sample was indexed and entered into the database.

An example of an individual record is available.

After the index was created, it was tested with five users. The feedback we received from the users guided the development of the steps for the second phase of the project. Workflow procedures as well as challenges and considerations were also written.

Rowland Institute for Science Web Site

January 1998 - May 2000, Rowland Institute for Science

In the winter of 1998 I took responsibility for the web site and intranet of the Rowland Institute for Science. My duties at first were limited to updating the library web pages, but they quickly expanded as I learned more about web design.

In December 1999 I managed and implemented the second of two major revisions of the site. The site has not be changed since that 1999 revision. This revision included a system of global, local and contextual navigation. It implements a full set of meta data elements (based on the Dublin Core Meta Data Element set) that is used for both description of data as well as workflow.

I also administered the web server while I was at the Institute. The Institute uses the Macintosh operating system, so we used WebSTAR as the web server. I was beginning to get into scripting with WebSiphon when I left the Institute.

I had read Lou Rosenfeld and Peter Morville's book "Information Architecture for the World Wide Web" while at Rowland. Unfortunately I did not have the resources to implement many of the ideas they proposed. If I could, I would go back and improve the information architecture of the Rowland site. For example,

  • Implement supplemental navigation in the form of a site map and global index.
  • Improve labeling throughout the site, especially in the global navigation. Labels such as "Rowland" are not descriptive or informative (that link actually goes to the staff directory).
  • Test with users. There is a wide variety of content on the Rowland site, and I would love to know how users would most like to see it organized. Card sorting with content cards, as well as task analysis would provide great insight into how the site could be better organized.

Cyrenius H. Booth Library Web Site

Summer 1997, Cyrenius H. Booth Library

In the summer of 1997 I created the Cyrenius H. Booth Library's very first web site. The site was very basic, as most sites in 1997 were. It did include meta data, as well as global navigation that was consistent throughout the site. I did all of the HTML coding, graphic design and wrote a majority of the content on that first site.

The site has since been updated, thank goodness, though some of my original content and design has continued.