Abhi Agarwal is the co-founder of NewsAI, a news intelligence platform to streamline the workflow of public relations professionals. He recently graduated from NYU Gallatin with a degree in Creation of Artificial Intelligence. Abhi grew up in Thailand and came to the U.S. for university. He enjoys spending time on Twitter, and at university he was a part of Tech@NYU - a student-lead technology club.

This post is part of the NYU Summer Launchpad blog series featuring NYU entrepreneurs’ first-hand accounts of challenges faced in starting a business and the lessons learned along the way. Learn more about the NYU Summer Launchpad 2016 participants here.

Hold off on building

My co-founder, Julie, and I are both engineers. As engineers and product-driven people we have a tendency to start building our product right when we have the idea. Why? Because we tend to rely on the skillset we have. There’s a common fallacy that exists in the startup world: build and they will come. It is easy to get trapped in this mindset and think that you should release your prototype before going out in the world and talking to people. Do the opposite.

We initially started off with the goal to build a better news aggregator for journalists. The idea came out of brainstorming sessions, side-projects, and internships. We ran the idea past our journalist friends – they liked it & we were off to the races! We quickly jumped into building it so we could begin showing it to journalists and get feedback and iterate quickly. We spent the next four to five weekends building our product completely isolated from customer feedback. It was great because we were in the zone and we thought we were making a lot of progress. What’s better than having a product demo?

Right after we finished our prototype, we joined the Summer Launchpad program at the Leslie eLab. In the program we were pushed to go out and discover if the problem we were solving is an actual pain point for our potential customers by talking to them.

After speaking to roughly 20 journalists we realized that the problem we were trying to solve was not a pain for them after all… but how did this happen? We had experience building tools for journalists in the past, and spoke to people who loved our idea.

We quickly understood that it was important to speak to a critical mass of people before starting to build or test your product - even an MVP. Speaking to a critical mass of people not only helps you identify if the pain exists for the solution you are building, but also helps you recognize patterns.

Understanding the behavior of your customers and the commonalities between their behaviors is critical because it helps you make better business decisions. It helps you start building a model of who they are, what problems they have, and how badly they want this problem solved.

You have to start noticing patterns

Until we got to a critical mass of interviews it was hard for us to fully understand how different kinds of people were currently solving the problem we were tackling, and the types of people who had those problems. Our idea evolved a lot depending on the number of customers we spoke to as we learnt 1) more about the problem, 2) more about how they are currently solving it, 3) more about the incentive for them to switch to a new solution. Without a deep understanding of these three ideas it becomes hard to build a business.

Through a better understanding of the problem and how they are currently solving it, we were able to gain more perspective on what exactly we are building and what products compete ours. In addition, through a better understanding of what incentivised them to switch to a new solution, we understood more deeply how they were looking to change their current work flow. What exactly were the problems in your current solution? How are you looking to fix them?

Sign people up!

At this point we identified the problem, and had potential solutions we could apply to solve these problems. We were incredibly excited at this point as we felt like we were finally at a place where we could start building. We quickly began drawing up dashboards and mockups for a MVP. We had a lot of ideas on ways we could build a great product & finally put things in front of people. We took our ideas and mockups to customer interviews and got their opinions on features that were valuable to them.

These interactions helped not just us build our platform, but also to sign them up and get excited about what we’re building. We also put up a landing page with our key value propositions on it, put up ads on Twitter, Facebook, and Reddit pointing to the landing page, and sent it to people we had already met. Setting up a landing page helped us define our key value propositions as we were able to quickly test the messaging of our brand. It also helped us get more customer interviews!

The whole process got us to a point where we had identified the problems, had a good understanding of our customers, and had people waiting to try out our product. We are excited about the next phase of NewsAI.