Meta Refresh 2018

The web in your pocket

How clumsy engineering can lead to powerful research


Dhruv Saxena


In most of user research practiced today, designers build click-through prototypes to get quick user feedback, which saves the product team a lot of time and money. The practice is extremely useful, but only during the infancy of a product.

After a few iterations the team has learnt all it could from a prototype and needs to build the real product with real data and real interactions to learn more. Months of painstaking engineering goes into building this ‘real product’. When the product is finally out, the team realises that it was built on a shaky foundation of incomplete research and has to put in many more months of work fixing problems that could have been surfaced had the research been more sophisticated.

What if we could use a tiny portion of the engineering bandwidth to turn a prototype into real software? Could the final production software with clean scalable code benefit from another piece of clumsy experimental software that can go out into the real world and test design assumptions with minimal engineering effort? If the experimental software has real interactions and real data, could it actually facilitate more sophisticated research than a prototype?

In this talk, we answer all these questions using an ongoing project at Uncommon as a case study.


  • Introduction to case study (with a small video)
  • Introduction to the common process of user research
  • Difficulties in doing user research with just a prototype
  • Benefits of building dirty but real software for user research
  • How to build real software for user research
  • How to get a buy-in from engineering for building real software just for research
  • How to collaborate with engineering and get them accustomed to the make-and-break process of design
  • How to use real software in user studies
  • How to synthesise user studies and come up with actionable results for engineering
  • How to incorporate learnings from experimental software into production software

Speaker bio

Dhruv heads the design practice at Uncommon - a digital product design consultancy that has helped many of India’s and SE Asia’s unicorn startups - Go-Jek, Flipkart, Myntra and Swiggy to name a few, achieve rapid exponential growth through design.

At Uncommon, Dhruv helps clients use design as a foundational pillar alongside business and technology to build products that people actually want to use.

In a previous life, Dhruv’s work was concentrated on designing physi-digital interfaces, which he built through collaboration with designers and professors at the Copenhagen Institute of Interaction Design (his alma-mater), MIT Media Lab and Stanford.