How this bunch of students from NIT Rourkela is building an OOH marketplace

Debi Prasad Sahoo was two years into a Chemical Engineering degree at NIT Rourkela when he started noticing billboards across Odisha sitting empty for months at a stretch. “The outdoor advertising industry is highly unorganised, managed by thousands of individuals,” Debi says. “There are many billboards that haven’t been booked for months, sometimes four to five months, and that’s a huge revenue loss for the vendor and the agency. At the same time, the client can’t get that billboard at a cheaper price.” 

That observation became HoardSpace. “We wanted to make the whole outdoor advertising industry digital, to build a marketplace just like Amazon, where a vendor can onboard their entire inventory in real time,” Debi says. On the client side, the idea works the same way: a client searches for a location, HoardSpace shows what’s available, and the client can send a booking request that the team confirms within a few hours. 

The team 

The core team is made up of second-year students from six different engineering departments at NIT Rourkela. Debi, from Chemical Engineering, is the founder. Arka Pal, from Computer Science, built the platform’s front end and design. Ayantik Sarkar, from Civil Engineering, leads the backend and AI integration, working alongside Shuvam Satapathi, from Artificial Intelligence, who co-runs the backend. Churepally Neha, also from Civil Engineering, handles vendor management and graphic design. Gagan Sharma, from Chemical Engineering, runs digital marketing and social media, and Preetam Samal, from Food Process Engineering, works on marketing. 

How the platform works 

On HoardSpace, clients can search for hoardings by name or address, and filter by state, city, hoarding type, lit or non-lit, size, and price. There’s a map view showing every listing across India, and vendors can add their own hoardings directly through the platform. Every listing includes campaign availability dates, pricing, and property details, along with a downloadable specification sheet. “These descriptions are generated using our AI,” Arka says, “and the data is pulled from the Google Maps API, which calculates the traffic and footfall of the area.” A comparison tool also lets a client select two or three hoardings and view them side by side across city, location, category, price, size, and availability. 

The platform’s central bet is HoardSpace AI, which lets a user type in a request and get back a filtered list of hoardings. “This is going to be one of our core features in the future,” Ayantik says. “Right now we’re doing it for free using free services, but when we have enough credits to integrate AI properly into the core platform, buyers and vendors will be able to manage the entire platform using our AI.” Today, the tool can filter hoardings by budget, city, and industry (categories such as real estate or education for instance), and answer general questions about how outdoor advertising works. 

Where the “Reach” numbers come from 

Each listing shows an estimated reach figure, generated by feeding location coordinates into their AI through the Google Maps API. Vendors can revise a listed reach if they feel it should be higher or lower, though Ayantik notes the change is verified on their end before it goes live: “They can do that, but we will verify it from our side whether it is actually right or wrong.” 

Debi was direct about the limits of the footfall estimate: “There can’t be any technology that can indicate if a person is looking at the hoarding or not. We can only predict the footfall to the location, how many people, how many vehicles have visited that location. That is near-accurate, as we are using the Google Maps API, and to be precise, we are using Gemini’s API, which gives us the data from Google Maps.” 

Revenue Model 

Listing on HoardSpace is currently free for vendors, as is the software they use. “We are basically a marketplace, without vendors we can’t get clients, and without clients we can’t get vendors,” Debi says. “So we onboard both of them, and eventually we will charge from both of them. Right now we don’t charge vendors, the listing is free, the software is free, but in the future, we may charge from them.” On the client side, the plan is to earn a commission on bookings. 

Traction so far 

The team started onboarding vendors about three months before this conversation, working from Odisha first. “We have 25-plus vendors, and around 60 to 70% of our vendors are from Odisha,” Debi says. “Right now we have 3,000-plus billboards in around 200-plus locations on our platform, covering 15-plus states.” 

Two client campaigns have been completed so far. In just one month of marketing for these two clients, HoardSpace generated a revenue of ₹3.91 lakhs, and the team is targeting ₹2 crore over the next two years.  

The commission model is built around actual outcomes: HoardSpace only charges a client once a billboard is successfully booked through the platform, taking a 5–6% commission on the total rent of the billboards booked. 

The team is currently in talks with incubation centres at IIT Madras, IIT Bhubaneswar, and IIM Sambalpur. “I’m hopeful we’ll get incubated in the next few months at other centres too, like IITs and IIMs,” Debi says. HoardSpace has already been recognised as a promising startup by FDBI at NIT Rourkela and is registered under Startup Odisha and Startup India. 

What’s next 

For now, the team’s focus stays on the basics, onboarding more vendors, closing more bookings, and building out the AI features as their resources allow. The larger goal for HoardSpace AI is to move it beyond search and filtering into something that can help plan and manage campaigns for both vendors and clients. Right now, though, the day-to-day work looks much the same as it did three months ago: one vendor onboarded, one booking closed, one billboard at a time.