05Crime Shield
Case studyPersonal safetyiOS app
Press the side button three times and Crime Shield sends your location and your voice to the police and to the people you trust. It is designed for everyone who can’t explain an emergency out loud.
01 — Overview
People with disabilities, or with limited English, can struggle to explain what is happening to the police, and to prove it afterwards. Crime Shield gets help moving, and captures evidence, without a single word.
The documentary Victim/Suspect followed young women who struggled to prove they were assaulted, and who got into trouble for passing wrong information to the police. It raised one question: how can someone get help fast, and back up their claim, in the worst moment?
An app built on Apple device features: instant camera and voice recording to capture evidence in real time, and a call to emergency services with a triple-click of the side button.
02 — Process
The UX and UI designers worked as one team, and a Business and Growth team stayed in constant contact, adding insight at every stage.
Aligning on the problem, the users and the goals.
Interviews, affinity map, personas, user story and journey map.
Sketches, then high-fidelity wireframes.
Prototype, two rounds of usability testing and the final screens.
The finished interface and what we learned.
03 — Discover
Reporting a crime and asking the police for help is hard, especially for people with disabilities or limited English. Claims are hard to prove and help arrives late.
The gapA mobile app that works with Apple devices. Capture evidence with the camera and voice recording, and call emergency services with a triple-click of the side button.
The answerPacked into one easy-to-use app, these features let people document an incident, prove what happened and get help from the police fast. The hope is that communities feel safer overall.
Target audience
Everyone who lives in the United States.
Interviews
Together with my research partner in the US, we interviewed eight people who have lived in Chicago for more than ten years. We started by asking whether they feel safe in the city and which safety concerns they have, then went deeper.
Affinity map
With the interview and survey data collected, we sorted every insight into an affinity map and let the themes appear.
04 — Define
The survey and interview data became an empathy map and a gains-and-pains list for each persona. Empathy maps let us see Crime Shield’s world through its users’ eyes.
I’ve been searching for an app that’s designed with accessibility in mind.
I wish there was a better way to communicate during emergencies. It’s scary not being able to explain my situation in English.
User story & journey
A user story map organised the interactions, prioritised the features and kept the design tied to what people actually need. Then we followed Lydia through the app, stage by stage.
05 — Ideate
Sketches and wireframes are the fastest way to share a design idea, and we can make them in minutes. In a brainstorming session we ended up with two versions of sketches. For each element we asked: why is it here, and how did what we know about our users and their goals shape it?
Hi-fi wireframes
Before jumping into the visual design, we created high-fidelity wireframes of every step: facing a crime, mobile verification, sign-up, emergency contacts, permissions and the Help Me button.
06 — Test
We ran two rounds of usability studies. The first guided the move from wireframes to mockups. The second used a high-fidelity prototype and showed which parts of the mockups still needed work.
07 — Prototype
The low-fidelity prototype connected the main user flow, from sign-up through to calling the police, and asked for permission to use the device’s location, recorder and contact list.
08 — Iterate
After empathising with our audience and learning what they feel and need, we ran a second round of interviews. It showed what the design needed and how we could help them better.
09 — Final design
Onboarding
Register and login
Main page
Profile settings
10 — What’s next
The next steps we planned:
