Contact me
Modernising an outdated process used by water utilities to investigate suspected leak outbreaks. From initial concept to pilot development, on-site research, and UX optimisation, I designed a location-aware mobile survey tool and management dashboard that digitised workflows, reduced investigation time, and enabled actionable analytics.
When a dramatic drop in water pressure signals a suspected leak, water networks dispatch field technicians to investigate within the affected District Meter Area (DMA) — zones that can span several square miles. Investigations involve manually testing hundreds of valves and hydrants within that DMA. The existing system was paper-based, inefficient, and led to poor-quality data.
Based on requirements gathered from leakage managers within a UK utility, I designed an MVP mobile-based investigation tool that allowed field technicians to efficiently locate survey points and log details. The goal was to rapidly improve the efficiency and traceability of leak investigations.
I shadowed field technicians as they piloted the app during a live leakage investigation in a dynamic, real-world environment. The core concept was validated almost immediately, with early detection of leak signals and faster investigation times. Technicians found the app easy and logical to use, but the trial also revealed several opportunities for significant optimisation, and these insights directly informed its evolution into a mature, enterprise-grade solution.
Based on trial findings, I aimed to minimise user interactions to reduce friction and help technicians complete surveys as efficiently and accurately as possible, while preserving data quality and accuracy.
The vast majority of surveys have the same outcome:
Common results no longer required a form to be filled out. One simple swipe of the asset details card would log a pre-set default result to dramatically reduce repetitive actions. On swipe, the survey is confirmed and the next survey point is immediately prompted.
Technicians would sometimes mark a survey point with chalk on nearby pavement and make a note to remind themselves to re-test later, a process that was easy to forget or misinterpret.
I introduced a 'Flag' feature, where a long-press on a survey point toggled a digital marker. This simple, intuitive gesture removed the need for physical markings and ensured revisit notes were clear, consistent, and digitally recorded.
The ‘Directions’ feature was only used in some edge cases, making the persistent ‘Directions’ button a waste of screen space. This could be packaged more intuitively.
Directions were now activated contextually. If a user tapped an asset icon on the map, it implied intent to navigate, activating turn-by-turn guidance. This interaction mirrors familiar touchscreen navigation patterns. The survey button could now also be removed, with that functionality accessible by tapping the asset details card, reducing screen clutter.
Tech’s sometimes logged results in batches, not immediately after a survey. Although this was user-friendly and logical in practice, managers wanted to avoid this behaviour as it could lead to skewed results and analytics.
To reduce batch-logging, location services verified the technician’s proximity. Within 10m of an asset, the 'quick swipe' feature unlocked. Beyond 10m, manual entry was required. This invisible guardrail encouraged accurate reporting, without nagging users.
