Great Asset Hunt

TL;DR

We helped make engineers jobs easier and more engaging, ultimately helping reduce the time it took for collected data to be stored from 6 months to 1 day and massively improving data quality.

Background

While at Play, we were approached by Thames Water to help improve the performance of their engineers, specifically around data collection. The current system they were using relied on giant Excel spreadsheets, was error prone and took 6 months for the data to be properly stored in their systems.

Process

We began with a 2-week Sprint 0 (design Sprint) to: help us understand the problem; propose and iterate on potential solutions; and define/ estimate a potential pilot to prove our idea. The Sprint 0 allowed us to remain agile (not committing to any estimates before we knew what we’re building) and gave our client confidence, as they only had to commit to the initial 2-weeks of work.

Following the Sprint 0 we designed and developed an MVP version for iPads to be piloted with engineers. We helped to plan and run the pilot, including defining and measuring the KPIs that would be used to measure success.

During the whole process we worked closely with stakeholders and engineers that would be using the product. This was important to make sure that we built something that was appropriate for the Thames Water culture, and would be well received by the team.

Product goals

• Make it easy for engineers to collect data while on-site
• Support the capture of high quality data
• Motivate engineers to continually collect comprehensive datasets over time
• Reduce the time spent by managers checking data - helping them spend their time more efficiently
• Allow managers to easily verify new equipment to ensure consistency across complete surveys
• Take advantage of the platform (photos, GPS) to streamline data collection

Product

We decided to focus on a gamified solution that helps to motivate engineers by creating a team-based game; where they win by improving their team’s performance. We shifted the perception by branding it “The Great Asset Hunt”, repositioning how the engineers thought about their work and making it fun. Working with Jez Fischer, we designed the UI based on the wireframes I created. We built a prototype using Marvel to test our initial designs with engineers.

Make it fun

Beyond just gamification features, we worked hard to ensure that The Great Asset Hunt didn’t look or feel like traditional ‘work’ applications that they were used to using. We also created a series of badges to add humour and reward the engineers for their long-term progress, my personal favourite was the ‘Taxidermy’ badge for finding dead animals on site! Engineers get immediate feedback on their performance with automated measures from the app and a manually entered quality score from managers. This creates a positive feedback loop to help engineers continually improve.

Reimagined data collection

We started from scratch by reviewing all the data that was currently requested, and ensured that everything had a purpose - removing hundreds of questions before we even started designing! We simplified how data is collected focusing on multiple choice questions and automated completion where possible. We completely changed the Information Architecture to make it easy for engineers to move between sections quickly as we discovered the data wasn’t collected linearly.

Human validation for quality

While we wanted to reduce human error as much as possible, it was required that some free-text areas remained for the data required. To help keep data standards high we developed a manager portal that allowed managers to review data - highlighting the free-text fields first as they’re most likely to include errors.

Results

After a successful Sprint 0 we built an MVP and ran a pilot for 3-months with a group of engineers to great success.

Key successes:

• The product was well received by engineers
• Reduced the time to collect data by 400%
• Reduced time from collected to storage from 6 months to 1 day