A Vital Mission for the USDA
The USDA’s Food Safety Inspection Service performs a vital mission for our country - protecting the public’s health by ensuring the safety of our food supply.
My team at Bixal gave the agency the digital presence it needed for this mission. We worked to help it build and maintain trust with the public and we gave it the tools it needs to inform policy.
You have never heard of it, but you encounter it every day.
The USDA’S Food Safety & Inspection Service (FSIS) is the organization the government tasks with ensuring that the meat, poultry, and eggs produced within the United States are safe to eat.
FSIS relies on its website - fsis.usda.gov - to be a critical communication and engagement tool for reaching everyone from food producers, researchers, and scientists, to public health partners, FSIS employees, public consumers, and even policymakers.
a filing cabinet became a high-end bookstore
Consumers, industry, consumer advocates, and government agencies increasingly rely on the new site for up-to-date information about food safety, food recalls, and federal policy.
Working through COVID my team never lost its focus on human-centered research and design, adapting our methods to find the answers we needed.
I gave the USDA and Americans a modern and customer-centered web experience informed by a deep understanding of FSIS customers and the role that the organization plays in the decisions they make about food safety.
The new, empirically validated design worked in concert with the content strategy. Together they supported the American people’s trust and confidence in the government’s ability to keep them safe from food-borne illness.
We improved content quality, delivery, and accessibility for everyone.
Traffic to fsis.usda.gov before and after the launch of the redesign in late 2021.
human-centered research & design
When restrictions prevented traditional methods like in-person interviews and moderated user testing, we always had a backup plan.
The team used a combination of desktop research and fieldwork to understand the structural interplay between food safety practices and the audiences that FSIS serves.
We continuously improved this understanding even when COVID hit. We explored new sources of insight like artifact journeys and market data, and we doubled down on thorough testing and validation to confirm the findings from these alternative approaches.
The single thread connecting multiple audiences
Landscape analysis gave us a systems-wide view of the audiences. We identified key artifacts in this system. One, in particular, stood out: the Hazard Avoidance and Critical Control Points plan (HAACP). Every audience intersects with a HAACP in some way. Scientists and policymakers define it, food processors need it to operate, FSIS inspectors ensure that plants adhere to it, and consumers are impacted by it when it triggers a recall.
The lifecycle of a HAACP guided the refinement of our research plan and, as the common thread that connects federal policy with the food on our dinner plate, provided the initial framework for contextual inquiry.
Continuous refinement
Landscape analysis gave us the foundation we needed to understand the basic information needs of the different audiences.
Consumers may not have a specific question in mind beyond, “I want to know more about food safety,” or “I want to know more about this particular recall.” We assumed that a Consumer has never interacted with FSIS before, or that they are even aware it exists. SMEs, on the other hand, can be expected to have more of a focused need. “I have a product that I want to export,” or “Is this product labeled properly?” or “How can I have an impact on food safety policy?”
“I see that this food was recalled because of Listeria. What is Listeria? What steps do I need to take (if any)? Why is FSIS the authority to trust here?”
For Consumers, the assumption is that for FSIS content to have value it is best presented within a context that gives it meaning. A Consumer benefits from knowing the proper temperature for cooking poultry, but they also benefit in the long term by trusting FSIS as an authority on food safety. When information is presented to a Consumer, then, it is best done with a focus on education and discovery.
For SMEs, the assumption is that they have a clearer sense of their information need than Consumers. SMEs already know the type of answer they are looking for and they don’t want to be bothered by a bunch of extraneous information or interactive crud.
Rapid Iteration & Rapid Prototyping
The sooner there was something that people could see and hold, the sooner the team would have a shared understanding of the problems they needed to solve.
The first design objective for each sprint was to produce something tangible. We chose the format of the prototype based on our confidence in what we knew and the particular goals of the sprint. We used everything from a Drupal site populated with publicly available content to an actual paper prototype of the homepage.
This focus on reaching tangibility did not limit choice. What it did instead was maintain the momentum of continuous improvement by reducing the cost of iteration. We could make, test, and even abandon incremental changes without impacting the schedule.
Collaborate, corroborate, then collaborate again
Early sketches did more than capture ideas. They established a shared understanding among everyone on the project - clients, stakeholders, managers, content strategists, designers, and developers - that this would be an iterative process propelled by transparency, openness, and communication.
Paper Prototyping
In early workshops we focused on unpacking the client’s thinking behind their ideal solution.
We recreated the original website in paper with additional material pulled from initial discovery and research, and the group played with different approaches to organization and layout by moving pieces around.
We worked to answer questions such as: What were the information needs of the audiences? How did they define those needs? What journeys could we create for them that would deliver a worthwhile payoff for them?
The goals of these exercises were not deliverables. We aimed for stories.
Zeroing in on information architecture
Audience research had begun with looking for the common thread connecting federal policy with the food on our dinner plate. The results of the landscape analysis provided a system-wide view of the food safety ecosystem. The narratives of an egg and HAACPs surfaced key intersection points between the actors and institutions, including FSIS. Analysis of the intersection points confirmed our hypothesis that the FSIS website would produce the most value by serving as an educational resource for the general public and an information resource for audiences operating within the food safety ecosystem.
We knew that when food safety information is presented to consumers it is best done with an emphasis on education and discovery. We knew that when a SME visited the website, they did so with a specific question in mind. Finally, we knew that the two groups sometimes had competing needs.
Most people don’t even know that FSIS exists.
Audience analysis uncovered a key characteristic of consumers: most don’t even know that FSIS exists. It is almost certain that if a consumer is visiting the FSIS site, it is their first encounter. We determined that the top-level pages for the site would focus on consumer education. We sought to introduce consumers to FSIS and its mission in FSIS in a way that communicated the importance of food safety and increased trust in our government’s ability to keep the food supply safe.
Given the different and often competing needs of consumers and SMEs we developed a tiered approach to the page design and information retrieval schemes for the FSIS site. Each main section of the site was differentiated by the audience it prioritized. Top level pages were oriented towards Consumers (i.e. the Museum pages), while deeper pages targeted SMEs (i.e. the Library pages).
Real Data, Real Interaction
The prototyping did not stop with paper. As we refined our understanding and strategy on the worktable, we also built out a fully interactive prototype.
Built using Drupal, this prototype was populated with real data pulled from publically available sources and imported into the CMS.
Thousands of food recalls. Nearly 7,000 businesses regulated by FSIS. Hundreds of definitions.
Because we had this data (and so much of it), we could identify relationships and patterns among content that we would have missed otherwise.
Because the prototype was fully interactive and because it was accessible to everyone on the project, we were able to catch more than just the immediate pain points. We found red routes that only appeared with repeated, frequent use, shorter routes to value, and opportunities for education.
data, not hunches
Quantitative research through web analytics, card sorting, and natural language processing informed the content strategy and information architecture. Testing and validation occurred continuously and ranged from treejack testing of the information architecture to impression testing of the visual design.
We had concluded that goal for the homepage was to educate consumers about FSIS in a way that communicated the importance of food safety and increased their trust in the government’s ability to keep the food supply safe. We had developed an initial strategy to make this happen. Testing the designs as they evolved gave us the insights we needed to refine that strategy.
Early feedback showed that the homepage design needed additional curation.
From consumers, we heard “It’s doing too much,” and “I’m not sure that it meets my needs as a consumer.” We needed to know that the page - the first encounter for major share of the total audience - and the narrative it presented would perform as expected.
The outcome
Nearly two years after its launch, fsis.usda.gov continues to grow and provide value for an increasingly large number of people.
The website established a new look and feel for an organization that impacts all of our lives in critical ways. Multiple audiences with a range of different needs and expectations use the new information architecture, feature set, and content strategy to find valuable information on food safety, science, and data, policy, and inspection.
Myself, I like to think that it has helped to save lives.