UX case study

Designing what happens when things go wrong

SmartPet solves the major pain points of similar geofencing apps across the industry: it removes annoyance, panic and jargon from experience by making tech working seamlessly and invisibly in the background.

The ideal story

Smart Pet is a geofencing app that tracks your pet via a GPS tracking device. Its main functionality is a virtual fence warning you when your pet leaves the fenced area. So instead of constantly checking on your pet, you can focus fully on what matters to you while the app is taking care of your pet.

The ideal story is:
1. You create a fence
2. When pet escapes the fence, the app warns you with an alarm.
3. Following app directions, you go after your pet and find it. Great!

The real story

1. You create a fence
2. Because of accuracy problems, the app tells you the pet has run away and it gives you directions to it, but the pet is actually near you, on the sofa.
3. You start wondering “How reliable the app is? Instead of peace of mind, I got annoyed and worried — what if the app is misleading when I will lose my pet for real?”

The problems—what, how, why

What is the human problem being solved?
SmartPet solves the major pain points of similar apps across the industry by:
- removing annoyance and panic from the experience.
- removing tech jargon from interaction by making tech working seamlessly and invisibly in the background.

How do you know it’s a real human problem? (i.e. what research insights or data backs it up?)
App reviews show that technology inaccuracies create two major pain points:
1. GPS inaccuracies produces false alarms (indoors, outdoors at home) — this is the annoying side of safe,  which at the same time raises questions about how reliable such products are in case a pet would really get lost;
2. Existent products leave users in panic whenever there is no coverage or accuracy. Except for articles buried inside Help section, these products are not designed to help people if:
- they have lost a pet while the accuracy is low;
- they have no coverage when walking with your pet;
- they lose a pet when there is no network coverage.


Once I understood the real model (why and how these problems happen), I could integrate the findings back into the "ideal" version of the story. The design helps to: 
- reduce false alarms indoors and outdoors at home;
- improve accuracy;
- know what to do if you have lost a pet;
- know what to do if you have no coverage after having lost a pet.


As an IDF Bootcamp student, I wanted to:
- understand and solve geofencing issues across the industry;
- understand and solve pain points users of GPS apps have ;
- design what happens when things go wrong.


Discover-Define-Develop-Deliver (the double diamond process): 
Discover: quantitative and qualitative research to understand the context, users’ needs and business objectives;
Define: product vision, user stories for persona building, scenarios and tasks for usability testing.
Develop: iterate with prototyping and usability testing.
Deliver: prototype and design documentation for development.

Discover: Statistics. Interviews. Pain points. Affinity map

Define: Key tasks, persona. journey map. user stories

Four flows covering most current apps issues

Develop: Prototypes and usability testing

Paper prototype, mid-fidelity and high-fidelity prototypes, each with its own moderated usability testing (UT) session.

Deliver: Design documentation

Results / Outcomes

While I'm happy that I went beyond the competitors' solutions to the top problems their clients face and while I guess I've solved a common problem to all geofencing apps, I wish I could have done significantly more contextual inquiry work and usability testing with people actually using pet trackers.

Final screens

Interactive prototype

Lessons learned

So far I've covered the two contexts in which Chris (our persona) is taking care of his pet in person. What's left unexplored is the context in which Chris is away from his pet (eg working in an office during the day, or being abroad for holidays) and needing someone to take over. Being able to share the tracker (for a whole range of reasons) is definitely a substantial chapter.

Read the full story on Medium (32 min read)

Related work


Fixing usability for specialty coffee guides

Best Coffee Guide saved me from bad coffee, so I wanted to save them from bad UX. This project is a give-back to the people who contributed to putting specialty cafés on a map.

See this project