Streamlining the onboarding process for law enforcement
An infographic detailing the reach that Spotlight has gained, since its national roll-out in February 2015.
BACKGROUND
Spotlight is a robust search tool that uses machine learning to help law enforcement identify victims of trafficking online. It went from beta in Summer 2014, to extended beta in Fall 2014, to a national roll-out in Winter 2015.
I was part of the team that was responsible for designing the UX/UI and rolling out the product Nationwide, along with our partners - Digital Reasoning - whose machine learning technology, Spotlight was built on.
Due to the sensitivity of the product itself, I will not be using this case study to highlight the product design or development of Spotlight, the tool. Rather, I will be detailing the UX/UI and program management that went behind the design of the on-boarding process, which I led during my time at Thorn.
CONTEXT
Designing a user experience for human trafficking investigators to help them solve cases faster than ever
My background in law enforcement and anti-human trafficking was well suited for the needs of Spotlight when I first joined the team. Initially, they had about 100 users from three U.S. states who were using Spotlight for their human-trafficking investigations. I was given the goal of on-boarding 250 users in the next six months.
User research through interviews, personas, and journey mapping of proactive and reactive searches had already been done prior to my start date. So when I came on, I got caught up to speed by reviewing that material, along with learning the product inside out by receiving, and then conducting, Spotlight demos and online trainings. This process was relatively ad hoc, however, with no identified workflow to drive rapid acquisition and adoption.
PROCESS
Identifying the best product and UX design to onboard users, so that Spotlight could become the leading machine learning tool for human trafficking investigators
While learning the ins and outs of Spotlight, and shipping features from our product backlog, I began to focus on ways to improve our on-boarding process for increased acquisition and adoption. I started by conducting a competitive and stakeholder analysis to help with the overall discovery of where we were with our current process.
Through my competitive analysis, I identified two major competing products, Traffic Jam and Rescue Forensics. Both products had been around longer than Spotlight and had a few hundred more users than us at the time. I gathered additional information through phone call interviews with customers who were familiar with Spotlight and the other two products to get a sense of what features and design they liked. From what I gathered, Spotlight was by far the most intuitive and robust, providing a better user experience, despite having a ways to go itself.
Additionally, through my stakeholder analysis, I focused on Fox Valley Technical College (FVTC) - a leading law enforcement education center - and the DOJ (both of whom I know, because I currently consult for them on human trafficking). Eventually, I reached out to them, and we agreed to partner and co-teach a Spotlight Lab workshop at major conferences across the Nation to drive awareness and adoption.
To design the curriculum for Spotlight Lab, I conducted preliminary research. I joined a superuser with the Louisiana State Police on a human trafficking sting operation in New Orleans, to observe how they used Spotlight in the field. The insights we gained - was not only beneficial for the overall product development - but for the workshop design itself (I would continue to observe Spotlight usage at future sting operations, including Operation Cross Country, in which hundreds of juveniles were identified and pimps were arrested).
After the sting operation, we sat down with FVTC and the DOJ to put together the curriculum for Spotlight Lab. A leading assumption was that users wanted to learn by doing, so we used the outcomes from the New Orleans sting operation as a hands-on case study. Moreover, users wanted to learn from a law enforcement investigator currently using our tool - so we made sure to co-teach with a superuser. Finally, we designed marketing collateral to drive further customer acquisition.
Spotlight marketing collateral, based on what law enforcement needed to know.
We piloted our workshop at the 2015 Atlanta ICAC Conference and received additional feedback on how to improve Spotlight Lab. I also conducted guerrilla usability testing to learn about the major pain points of the product itself. This was meant to identify areas in which we could make the tool more intuitive, to minimize obstacles to adoption. After I completed this phase of research, I used the information to iterate on the workshop design, and improve the overall usability of Spotlight. My goal was to not only design a workshop that increased adoption but to build a tool that added value for current users through an intuitive user experience and user interface - which would then translate to more victims identified in a shorter amount of time.
This is me at a conference booth, preparing to hold several usability testing sessions with human trafficking investigators.
After the 2015 Atlanta ICAC Conference, we quickly realized that we needed a better on-boarding process that no longer relied solely on Google Sheets to manage requests. With the approval of the Thorn CEO, I led the design of a Spotlight product website and managed our vendors, Media Cause and Digital Reasoning - who were in charge of graphic design and engineering development.
For this phase of the project, I used the data I already gathered from Spotlight users to create a journey for those seeking to learn more about Spotlight, request a demo, request access, and to log in. I also created high-fidelity prototypes of key touchpoints on our site, whose content I validated through additional user feedback.
Journey map of users navigating htspotlight.com
Hi-fidelity prototypes of key touch points on the Spotlight website.
From there, I created an information architecture for the content on the website, put together a website design inspiration board, sketched wireframes by hand and through an online tool - and then wrote the content itself. Throughout this process, I worked collaboratively with Media Cause and Digital Reasoning to set timelines and execute.
OUTCOMES
An improved product and UX journey delivers a seamless onboarding and training process for our users and admin staff
By the end of my first year at Thorn, I had played a role in developing strategic partnerships with FVTC and the DOJ, designing a Spotlight Lab to effectively train potential and new users, and designing a Spotlight product website and a corresponding on-boarding process.
Here are samples of the Spotlight product website: home page, request access page, and request demo page.
Key results from our multi-channel strategy led to 19x the projected customer acquisition goal from all 50 states, by the end of 2015 (easily surpassing the initial goal of 250 users in six months). Furthermore, by making sure users were engaged and properly trained, user surveys have shown that Spotlight has since been able to reduce the time it takes to investigate a human trafficking case by 43%, identified close to 2,000 children, and over 2,000 traffickers.
Data gathered from user surveys.
Finally, we were awarded the 2015 Cloudera Social Impact Award, for the work we did on Spotlight.