Article

How we rebuilt a 20 year old site without starting over

Reorganizing, reimagining, and rebuilding the Death Penalty Information Center’s website and database

Share

When a website is old, it’s tempting to throw it away and start fresh without ever looking back. Aside from that temptation, when you're working with a website that's cited in Supreme Court decisions, throwing it out isn't an option. Doing the hard work of sorting through and carefully considering existing content and infrastructure is laborious, but having that deep understanding yields a better result in the end. You have to get your hands dirty before you can start cleaning up.

This was the case in our work with the Death Penalty Information Center (DPIC), and after much hard work and anticipation, we are beyond thrilled to launch DPIC’s new website. Beyond thrilled. DPIC is a national non-profit organization serving the media and the public with analysis and information on issues concerning capital punishment. Founded in 1990, DPIC promotes informed discussion of the death penalty by preparing in-depth reports, conducting briefings for journalists, and serving as a resource to those working on the issue.

When we first started working with DPIC, we found ourselves working with a well-established organization that was still delivering great value to its audience despite persistent problems due to aging technology. They had a sprawling, unwieldy website culminating from nearly two decades of active publishing. However, the website still had good traction. It averages 15,000 users a day, contains more than 15,000 pieces of unique content, and can receive up to 3 million page views in a given month.

To us, the content quantity and overall site traffic indicate the relevance and quality of DPIC’s work. Despite being very dated, the website continued to receive legitimate traffic. However, besides being in clear need of a fresh look, the old website did pose substantial obstacles and threats to daily operation for DPIC. The website’s content was very unorganized for both site administrators and visitors. Most of the content was completely uncategorized in the cumbersome Drupal implementation. Average users had a lot of difficulty finding the information they were looking for. DPIC actually possessed more data than was present on the site, but the lack of organization prevented them from having a place to put everything. Worst of all, the website was sometimes unable to handle the sporadic spikes of extremely high traffic when current events would drive users to the site (this was one of the very first things we addressed—coming up with a temporary solution to handle those traffic spikes).

And so we embarked upon a journey together, the team at the Death Penalty Information Center and our team at Foster Made. We came up with a plan to redesign and rebuild the website as well as build a completely custom database to house DPIC’s vast amounts of statistical data. Our work started with a deep dive into user research, producing an extensive 70+ page report of our findings and strategic recommendations. We conducted a content audit, competitive analysis, and considerable user survey. We developed user personas and journeys, ran tree testing for the new navigation, and created an improved information hierarchy for the navigation and the flow of content on individual site pages. 

During all of our research, we strove to understand who DPIC’s website users were in order to provide an optimal experience that allowed everyone to easily find, digest, and interact with the available content. DPIC already had a good idea about who was included in their audience: lawyers, legislators, reporters, educators, students, and the general public. As we spoke with and surveyed those individuals, one of the breakthroughs in our user research was dividing users into three groups: users who were experienced or experts in the subject matter, users who were familiar and had moderate knowledge of the subject, and users who were just curious and wanted to learn more. It seemed so simple, too simple. We kept trying to poke holes in this uncomplicated categorization, but the groupings seemed to hold true, and it helped us to not leave any major user groups out of consideration as we developed our strategy. 

With our strategy in hand (which included much more than just categorizing users into groups based on their level of knowledge about the subject matter), we moved into wireframes. As we reworked the navigation and created wireframes, a significant consideration was that we couldn’t “lose” any of the pre-existing content and data. On the old site, there were many pages that were strays, three or four levels deep into the navigation (or not in the navigation at all). During the wireframing process, we tried to clearly organize pages while also finding opportunities to connect users to related content that might be elsewhere on the site. We had our work cut out for us, but we created a flexible design system that helped provide structure and direction for where content could move and live. We partnered with Mark Wells at Antistatic Design who brought our advanced wireframes to life through design.

As we moved from design and into the technical implementation, there were a few key aspects to consider: the website build, the database build, and how to sync the two in a way that made sense. We decided from the beginning that the core of the site would be built on Craft while the main data store would be built as an independent Laravel application. A custom Craft plugin would allow the ingestion and rendering of data via GraphQL queries from the custom Laravel application.

Since DPIC houses the most comprehensive dataset collection on capital punishment within the United States, the way we handled the database and its integration with the website was integral to our development work and to the project being a success. DPIC’s preexisting “database” had long consisted of disjointed information systems, making the availability of real-time data for public consumption on the website nearly impossible. Without getting into the nitty-gritty details of the database project itself, the first step in delivering the new website was the development of the custom data management application to allow the aggregation, management, and reporting of DPIC’s vast store of information. The new database application would also make data available for real-time display on the new website through a Tableau connector as well as public and private GraphQL endpoints.

While our team was developing the new application we were also developing the new website on Craft to replace the aging Drupal site. Our focus in the CMS build was providing modular content components combined with data relationships to give DPIC the ability to create a fluid website that would scale and adapt as the demands of the website and the organization’s publishers changed. 

The new website and database have enormous potential that we are just beginning to unleash.

In conclusion, there’s so much to say about this project. What we’ve shared here just scratches the surface. It’s hard work, but it’s rewarding to take something that had already been invested in by many others for many years and to give it a new life to last for many more to come. On top of that, it was a true delight to work with the wonderful people at the Death Penalty Information Center. They are a group of extremely talented individuals, and it’s been an honor to walk alongside DPIC on this journey to achieve their digital goals of making their extensive research and data about the death penalty more accessible to a very broad audience.


Great things start with a conversation

Want to stay in the loop? Sign-up for our quarterly newsletter and we’ll send you updates with a mix of our latest content.