tradesumo

By Ivan Teong | Cofounder of TradeSumo

People always ask me what launching a tech startup while studying or working is like. It is definitely possible as long as you have a committed team with a united vision. One of the good things about doing this is that it gives you the time to test your assumptions and hypotheses.

It also gives you the bandwidth to explore an idea from different angles without the constraint of a runway. Usually the startup dies when it reaches the “end of the runway” but, for now, we don’t have that issue.

This is the story of how TradeSumo evolved from a hackathon project called SwitchTick, and what we did to develop the idea amidst all our commitments.

How It All Began

It was at Startup Weekend Jacksonville while on study exchange in the United States that I pitched the initial idea. This is where I met Daniel [Sudhindaran], Max, and Piotr, who were all participants in the hackathon because they were doing a venture creation course for college.

They could get a discount as it was organised on their campus, and write an essay about their hackathon experience for an assignment. They joined my team after I pitched the idea in front of the audience, and we worked together over the course of the weekend. Although we did not manage to win anything, Daniel asked us whether we wanted to continue as he thought that it was a great idea. He was willing to code up the web application from scratch (we used a prototype for the hackathon that had no functioning backend – it was just enough to show the user experience only).

The Startup Weekend Team - Daniel, Me, Max, and Piotr
The Startup Weekend Team – Daniel, Me, Max, and Piotr

I was surprised at his enthusiasm and decided to continue with Daniel after the hackathon ended, although Max and Piotr did not. A fundamental reason for that was because we worked well together during a short and intense hackathon; I wanted to explore whether this partnership could progress further beyond the hackathon.

Another reason was because he was still enthusiastic to continue even though we didn’t win anything, and on hindsight, that was important to me as I didn’t want to work with someone who gives up easily. Even if we had won the hackathon, there would have been many obstacles that would present themselves after the hackathon if we had continued.

Daniel has shown the same level of passion and commitment about the idea until today, as we initiate conversations almost every other day about the project outside of meetings. Sometimes, it would be me who started losing faith and he would sell the idea back to me, and other times it would be the opposite. I think that this kind of cofounder dynamic is really important, as it will be too draining for only one person to advocate and believe in the idea all the time.

startupweekend2

The initial startup idea was an auction site similar to eBay which I named SwitchTick. The difference from eBay was that it was an auction system where people used goods, instead of money, to bid for the goods on auction. As it is not like money where the kind of offers you get are limited within a monetary range, I thought that it would be psychologically addictive since it will be like a game where you may get random things that you never expect.

The concept of ‘one man’s trash is another man’s treasure’ is also core to the idea, as what is trash to you might be extremely valuable to someone else. Why go through the additional step of selling something to get the money for buying something you want, when you can do a direct trade?

The value of second-hand goods is also more subjective than first-hand goods, and usually you have to sell cheap (less than 50 percent of what you bought) or else nobody will want to buy. In that sense, the market’s offer of cash value for your goods may not be the same worth as what you value in your own eyes.

With a trading system, the market may offer you other goods that you value but the owner doesn’t in exchange for your own goods that you don’t value but the market does. The value difference subjectivity between people is greater as compared to cash value due to different people’s psyche. I thought that there are a lot of possibilities with this kind of trading system as compared to a cash system, and wanted to explore it. This idea will eventually evolve over time to include cash as a complement to the trading system we have, but trading will still be core to the idea.

At the Startup Weekend and afterwards, we heard from various people that a Canadian guy called Kyle McDonald had managed to trade his way from one red paper clip all the way to a house over a series of 14 trades across America. What I found fascinating about the One Red Paper Clip story was the randomness of the trades – how he traded intangibles such as rent, a movie role and even an afternoon with a rock star to get to the house that he wanted.

Although his case was a bit unique as the publicity around his story helped him get the attention he needed to get all kinds of trades, I started to wonder if it is possible to create something similar that could facilitate this trading randomness.

Lessons from One Spark

After the hackathon, we went to the One Spark Crowdfunding Festival in Jacksonville to promote the project. We were allocated a place in a room that was in some obscure part of the shopping mall. As it was a room just beside the main entrance that nobody noticed as everyone’s focus was on entering the mall, we had a lot of trouble trying to get people to come into the room to check out our project among other projects. We had to think of a way to get people to come into the room and go directly to our stall.

Daniel and I at One Spark Crowdfunding Festival
Daniel and I at One Spark Crowdfunding Festival

Daniel suggested buying some candy and chocolates to entice people into the room and using a huge banner promoting this to be placed on the outside. It turned out to work really well! The problem was that we were two people and too many people were coming to our stall to get the free stuff.

We did not have enough manpower to sell the project properly to all of them. Another mistake we made was that we did not ‘close’ them on the spot, and just expected them to vote at their own time. Obviously, they went on to see many other projects and forgot about us. This was one of the first few lessons that we learned in our startup journey – remember to close right after you sell and make sure that you have enough manpower.

Rebranding

We soon realized that “SwitchTick” was hard to pronounce and remember. What was worse was that some people even heard “SwitchDick” when we told them our startup name, and it sounded like a porn website. We decided that we needed to come up with a new name and logo, so we went online to look for potential ideas.

We wanted a name that had a .com domain, as this was something that was easier to remember and better for SEO. The problem with domain names these days is the number of domain squatters around. I wanted “Trade” to be in the name as I thought that our idea might change later (and it did), and it sounded more flexible than “Switch” or “Swap.”

gatormascotAmerica has this really unique mascot culture, where each college has its own mascot and the mascot will be a unifying identity for competing colleges during sport tournaments. It was also something that some startups like Reddit and Twitter were well known for, and I wanted something similar for our startup. It was quite random but we managed to settle on ‘TradeSumo’ as our new name. “Sumo” was easy to pronounce and the idea of a big, fat sumo for a mascot seems quite memorable.

It was also representative of all the goods that you want to trade, which would resemble a sumo when they were in a huge pile. Although we have not come up with the drawing of the mascot yet, we think it should be something that will differentiate us later to the consumer market.

We also decided not to have the mascot as part of the logo, as this will allow us more flexibility to change the mascot’s design later. The designer we hired for the logo came up with a trading symbol with different colours to represent the trading concept in a simple yet professional way.

Facebook for Customer Development

As developing the web application will take a while, since all of us have other commitments like studies and full-time jobs, I did not want to wait until it was ready before I did some customer development. I created some Facebook Groups for our target users, and got them to interact within these groups in order to understand their user behaviour. Acquiring these users from different sources was my first lesson in user acquisition. I got banned from some sites, and used all kinds of methods to drive users from other marketplaces to these Facebook Groups.

These Facebook Groups turned out to be a great idea as we were able to understand the pain points of trading by observing how they traded with each other, and develop a system that will resolve these issues. It was also good for identifying the potential early adopters of the idea, so that we know who to onboard into the application later when we launch. Most importantly, Facebook allowed me to reach out to total strangers through Facebook Messenger to initiate conversations for user interviews, as well as conduct polls on the Facebook Groups when we needed to ask for our users’ opinions.

tradesumosmsThese Facebook Groups were also great for recruiting other people to join our team to help out, as it was a way of proving the viability of the idea before the product was built. Erick and Philip joined us to help out with programming. We were also able to get Garrett, an experienced user experience designer, to help out with the initial site design. All this additional help was of great value to us, as we had limited time to work on the project due to our full/part-time jobs and studies.

tradesumopollSome of our hypotheses that we had regarding our users were being disproven just from observing them on these groups. This led us to develop a trading system with features that complemented our users’ behaviour. For one, our initial idea of an auction-based marketplace with a countdown time period set by the owner for each listing was thrown out of the window. We realized that the most enthusiastic traders were not concerned about time at all. They would go all the way to the oldest listing in the Facebook Groups just to find a good trade.

Another hypothesis that was disproven was the supposed efficiency of existing marketplace solutions. Traders would post photos of things that they wanted to offer in the Facebook Listing’s comments, but they did not post all the things that they owned. Both parties will have to keep asking what the other have in their inventory. From that, we knew that sites like Facebook, eBay and Gumtree did not have the most efficient user interface.

We knew that we had to come up with something that could allow both parties to see each other’s inventory all the time, so they could negotiate trades easily just like how you would do it in the real world: imagine two users laying out their stuff on two tables, and you would negotiate on the spot what you would like to trade easily.

Yet another disproven hypothesis was that traders only wanted to do a straight swap. Traders wanted to be able to do all kinds of trades, which meant trades where they could top up with cash their offered goods for swaps (due to imbalance in market value) and also buy/sell transactions – they would choose whatever offer from the marketplace which is considered a better deal to them.

We brainstormed over how to design a trading system that made the most sense to users. It had to show the inventory of both traders at all times, and enable all kinds of trades to happen in the most efficient manner without too much friction. It also needs to have a messaging system that helps users find out more information about each other’s things and negotiate trades, as we found out that users tend to continue their conversations through private messaging beyond the Facebook Comments.

That was when we came across some images of the trading systems that exist in computer games like Runescape and World of Warcraft, as well as various other multi-player online role-playing games. These trading systems have a trading table for both parties, where you can place multiple goods and cash to negotiate a trade. However, they usually only show the inventory of one side. We knew that it would be more efficient to show both sides’ inventories, so we adapted that trading system with the inventories of both sides shown for the real world.

tradesumo

Features, features, features

One of the most crucial things that we realized developing the web application with our limited resources and time was the importance of focusing only on the most important features. We also had to bear in mind the tons of bugs that appear with every code pushed, which further lengthened the time taken for a feature to be marked safe to go live. When the date to launch kept getting pushed back again and again even though we were committing and pushing code every week, we decided that we had to cut the number of features by half.

We had this grand plan of the web application doing all kinds of things for users, but the reality is that it will take forever to launch if we kept to that. We made a conscious decision to prioritize the features that allowed users to do the most basic kind of trades, and pushed every other feature to after launch.

I also divided our task board on Trello into three columns consisting of Feature Request, UX Improvements, and Bug Reports, as the tasks were becoming too confusing to track. We would decide as a team every week what tasks to prioritize based on their relative importance and urgency.

Ready to rock and roll

We finally managed to push the first version of TradeSumo into production for a soft launch to our network of family and friends in February 2016 after many months of hard work. Immediately, we got valuable feedback about the trading system and marketplace which disproved some of the other hypotheses that we had.

For example, we thought that it wasn’t necessary for users to be able to see their own goods in the marketplace as this would clutter the site. However, users were confused that their goods were not there and thought that the goods were not uploaded properly – even though their goods were clearly uploaded as shown in their garage inventory. They also wanted to make sure that their listings look appealing from the eyes of other users.

As we continue to realize, launching a tech startup is a neverending journey and new learnings about our users will emerge as we continue on this startup journey.

Do all kinds of trades now with TradeSumo at tradesumo.com.

Image: Daniel Sudhindaran and Ivan Teong.