Building Janta: design process, challenges and learnings

Janta came into existence as a result of an increasing number of “Today my Uber…” conversations in India. Delhi, in particular.
Today my uber cancelled on me like, 4 times in a row.
Today my uber driver drove so recklessly, I was scared for my life.
Today my uber had lizards in it. LIZARDS.
This is when our client approached us. I work for ProductStudio, a company that builds in-house products and for its clients, and invests in a few as well. In this instance, the client didn’t have a well-formed solution, but he was determined to fix this problem using technology. “I’m handing over the reins to you”, he said.
The very first reactions were —
- OMG. A client willing to give us complete creative freedom?!
- Uber and Ola already ask their users for feedback, and drivers with low ratings are eliminated from the system. We can even read reviews written by actual users. How can we do anything more?
- Wait a minute... Why have we never read a negative review for a driver? Are these reviews audited?!
- These apps already have customer support built in. Are users’ grievances not being addressed?
- Is this problem even big enough to need solving?
In a world of over-populated home screens and excessive screen times, we’re wary of building apps that just don’t need to exist. And we aren’t afraid to be blunt about it with our clients, either. With this project, in particular, we considered options like citizen campaigns, PR campaigns, Facebook groups and even change.org petitions.

The whole team sat together to brainstorm with the client, and thus began a massive outpouring of complaints. We added a lot more to our list of “Today my uber…” conversations. Lots of advice was exchanged. Lots of ideas, too.
Surprisingly, at the end of our brainstorming/bitching session, we all felt this strange lightness. It turns out, telling others about bad experiences gave us a feeling of unity and belonging, like we were looking out for each other.
An idea began to take shape. The client decided he wants to build a community of people to crowdsource reviews of cabs. Building off of that, I put together a ‘press release’ for our future product.
Imagine an app that tells you every complaint there has been about the cab driver you’re about to ride with, saving yourself from an unsafe cab ride and a bad overall experience. Simply take a screenshot of your app once a driver is assigned to you, upload it into the app, and know your driver’s safety rank in under 30 seconds. If you see any red flags, immediately cancel and re-book. You can also help make your city safer by reviewing your cab after completing a ride. It works the same way, take a screenshot, upload it and review — all under 30 seconds.
Our Team
I acted as Product Manager and also oversaw the design efforts (UX, prototyping and testing). We had a UX/UI designer, a Machine Learning engineer, a back-end developer, an Android developer and a tester.
User Research
I floated a survey through Survey Monkey and conducted interviews in-person and over the phone. During this phase of the project, I practice an 80/20 distribution of time — 80% listening and 20% talking. The trick to successfully extracting information from users, in my experience, is to frame answerable, mostly open-ended questions. For example,
Instead of : “Have you ever felt uncomfortable or unsafe in a cab?”,
I ask : “Can you remember a time when your cab ride didn’t go as expected? Tell me about it.”
I actively restrain my urge to nudge a user in a particular direction. I never interrupt or prompt anyone I’m interviewing. Controlling my reactions is key. In fact, after a few moments of silence, people will usually elaborate and analyze better, and without bias.
The ‘M’ in MVP
For me, building MVPs is like baking a cake. You can’t quite call it a day just after mixing the batter. You have to bake it at the right temperature, poke it with a toothpick to check for doneness and finally, cool it down and taste. Get your friends and family to taste as well. How is it?
Too dense? Maybe it needs more baking powder. Too dry? Maybe a splash of milk next time.

It’s crucial to first get your cake right. You can bring out the frosting, piping bag and sprinkles in Phase 2!
For instance, we decided that an Android-first approach was the way to go. iOS could wait. Anyway, why invest in an iOS app if the Android one doesn’t pick up?
This is the list of features for Janta’s MVP —
- Users should be able to search for reviews of cabs immediately after booking.
- Users should be able to seamlessly write reviews for the cabs they ride in. It was unreal to expect users to manually type in their cab’s license plate numbers into the app. We decided to use machine learning to our benefit. Users can take a screenshot of their Uber or Ola app and upload it to Janta, and it’ll do the rest
Market Research and Competitive Analysis
When building unconventional products such as Janta, it’s easy to simply say “We don’t have a competitor”.
In my opinion, that’s usually an obtuse way of looking at a product. In the case of Janta, users were trying to solve the same problem we were trying to solve. Most were using in-app customer care services to complain, or even Twitter and Facebook.
The challenge was to show users the value of having another app on their phone. The Janta experience needed to convey the brand’s essence.
Jump into design
It was time to convert discussions into wireframes because that means the best part of any project comes next. (I’m talking about user testing!)



Digital Wireframes
My team almost always creates digital wireframes for mobile app projects. This way, we’re able to get client feedback from the very beginning in a structured manner without unnecessary Skype calls or emails.
Although designers in my team are partial to Sketch app, we were very happy using Figma for this project, since not everyone on the design team had a MacBook. Figma made collaborating and receiving feedback really easy.

UX goes beyond swipes and taps
My particular style of product management is heavily UX driven. Most significant product decisions are made by simply asking users what they want.
For Janta, it was clear that product success depends on a large repository of user-generated reviews.
I had to make it as easy as possible for users to write reviews.


User testing proved some assumptions wrong. And that’s a good thing.
Now, the challenge was to make sure looking up reviews for a cab was effortless.
At first, the idea of taking a screenshot and uploading it into Janta to search for reviews seemed brilliant. No need to go back and forth between two apps to enter a license plate number, and in a matter of seconds, the app will show you search results.
When I got down to testing prototypes, it turned out the process was unfamiliar to most users.
Maybe little ‘help’ screens will help?

What if there was a way for Janta to know when a user opens their Uber app and books a ride?

Wrapping up
I know I’ve focused mostly on product strategy and design in this write-up, but there’s so much more that goes on behind the scenes to make sure a product gets shipped in time. To get Janta in the Android Play store in a tight time frame meant exhaustive technical documentation, complex user flowcharting, comprehensive product specifications and superhuman project management skills.
I’d love to know what you think about Janta.
On Medium, I usually lurk around and occasionally write comments on others’ stories, but this is the first time I’m writing my own. I will be happy to hear any feedback.
Hope this was worth your 7 minutes!