NMDC Interactive Kiosk • NMDC Limited • 2024 to 2025
Making a kiosk worth more than a screen
It started as a hardware sale. NMDC was offered a movable standee and answered bluntly: it costs more than a screen, and looping news and circulars does not justify the difference. I designed and built the software that did, now running on 16 standees across 6 floors of their Hyderabad headquarters.
Challenge
Make it worth more than a screen
The brief, as I heard it
Not "design a kiosk interface". Every feature on the home screen had to be an answer to one question: why is this better than a TV on the wall?
- Give people a reason to stop and touch it
- NMDC staff must update everything themselves
- An even number of buttons on the home screen
- Unreliable office network, years of large media
Action
Eight tiles, one person, one bad network
1. Designed a home screen where every tile earns its place
Website and Socials turned the standee from a signage box into a walk-up portal. Quiz gave people a reason to interact. Then NMDC said seven buttons felt unfinished. I proposed a terminology section as the eighth; their team asked for it in Hindi and English. Word of the Day became the feature people mention most, and it exists because a grid had to be even.
- Circulars
- Photo Gallery
- News
- Films
- Quiz
- Socials
- Website
- Word of the Day
2. Built it alone, as a system and not a mock-up
There was no team to hand off to. I built every section as a real, content-managed system in Node.js, because NMDC needed to update all of it without calling us.
The back office.
3. Designed uploads for a bad network
A normal upload screen would have failed constantly. Files are split into chunks, so a dropped connection resumes instead of restarting, and staff see what succeeded and failed file by file.
Nobody sees this on the kiosk. It is why the kiosk stays current.
Result
What got sold was the software
- 16standees deployed
- 6floors of the Hyderabad headquarters
- 140quiz plays a week
- 12content updates a week, made by NMDC staff
The standee sale succeeded, but not because of the hardware. NMDC bought a way to publish their own content, teach their own staff and give visitors something to do. The wheels were the least important part.
Sixteen units, one backend.
Reflection
What I would change
- 1Measure use from day oneI can count standees but not interactions. Simple usage logging would have told us which tiles earn their place.
- 2Treat constraints as promptsThe even-grid rule looked arbitrary and produced the best feature.
- 3Design the invisible partsThe upload system is the least visible work here and the reason the rest keeps working.
More case studies
