Published on 02/06/2023
We run a SAAS tool that converts UI designs into frontend code. As for our most signficant achievement, we built and launched our MVP in 2 weeks
Hi, I am Stephen, co-founder of FireJet, a SAAS tool that converts Figma designs into code with one click. Our Figma plugin helps developers save the weeks spent manually converting user interface designs into front-end code. This reduces their overall development time, which means they can launch websites, products and features faster. As of today (May 2023), we have about 9000 users.
Since our freshman year in SUTD, Philip and I have been trying to start a business together. I would say FireJet is our 10th attempt combined, and we have this running joke that since 9 out of 10 startups fail, this is bound to succeed. 😛
At some point in our journey, we started developing custom web applications for several clients. During this time, we found ourselves spending too much time converting our clients' Figma designs into code. We tried many existing tools to automate this process and found ourselves unsatisfied with the code generated from the current tools in the market. We also found a lot of complaints about existing design-to-code tools in online communities. Our turning point was when one of the tools generated our Figma design as a single image and we quickly realised that there was a huge gap in the market space for design-to-code tools, and decided to build our own tool to solve this pain point.
I think the first step when entering any industry at all is to talk to as many target customers as possible. You want to understand the pain point from their perspective, which is valuable because these people have more experience in the industry than you. For us, it took us many months of grinding cold calls on LinkedIn, Reddit, and other online communities. Ideally, you want to be looking for people willing to pay you to solve their problem.
After signing off on our first LOI from our first customer, our business idea was finally validated and ready for execution. Here's what we did next
Spent 2 weeks building our MVP
The initial version met all the requirements in the LOI and the customer was happy and ready to purchase a subscription
Set up our payment system ASAP
The best way to determine a product's success is to see how many were willing to pay for it. We used Stripe because it was easy to set up.
Increase our online presence
We didn't want to spend on marketing unless we could confirm product-market fit, so we looked to free options. We posted on Reddit, LinkedIn, Telegram and Discord channels. Most importantly we worked on our SEO through our landing page and the description on the Figma plugin library, which we learnt later on was how the bulk of our users found us
Establish multiple feedback channels
We needed to make it super easy for our users to connect to us. We set up several automated email flows and tried to talk to as many customers in person. We continued cold calling experienced developers, designers, and product managers on LinkedIn for a second thought on our value proposition.
Product changes and improvements
Based on the feedback we got, we improved our product. We also did a bit of UX testing to make sure that our customers could reach our key-value moment, which was the design-to-code conversion.
I want to add that incorporating was one of the least of our priorities. We decided not to incorporate it because building the business was more important than some fancy pte ltd.
Users show up daily and we haven't spent a cent on marketing. But getting them to retain is a different story and to do that you need to improve your product based on customer feedback. We have several strategies, and most of them involve making it extremely accessible for users to talk to you
Email flows
Keep the emails simple and direct. We use EmailOctopus to automate this process. It can be as simple as asking them for feedback after signing up. That means no fancy graphics and/or heavily styled emails. An example can be“Hi John, I noticed you haven't been using FireJet recently. It would be a great help to our startup direction if you could tell us the reason you stopped using FireJet.”
Not everyone will reply to your email flows, but for those that do, try your very best to funnel them into a call. If they aren't interested, at the very least you have textual feedback to rely on.
Customer support
Establish a live customer support channel in both our product and our website. People will approach you for help often and make sure you are ready anytime to answer their queries. I made my Tawk.to app ring my phone 7 times just so I know it's not a regular notification. Again, try to funnel them into a call.
Extended free trials in exchange for feedback
Don't be afraid to offer whatever it takes to receive good feedback from your customers. Good customer feedback helps your product in the long run and is infinitely more valuable than money
Signed LOI for feature requests
You will sometimes get users who want a certain feature. But some people just want to suggest but don't want to pay. Feature requests are good but it doesn't matter if the person doesn't feel a strong enough pain point to pay for it. If someone is asking for a feature that will take a large amount of resources to build in, I often try to funnel them into signing an LOI that basically says “We'll build it for you, but we need your promise to purchase the product when the thing you asked for is out”
I would say that it was our first customer. We went from 0 to 100 when he came to our call ready with a whole Notion document outlining what he was willing to pay for. It also helped that the stuff he was requesting validated our hypothesis about the gaps in the market. We told him that he would have to subscribe for a 3 months subscription, to which he responded that he wouldn't mind paying 6 months upfront. It was at that point that we were finally reassured that we could possibly be moving in the right direction.
I think it is to understand that Figma is a design tool built for designers. There are things you can do in Figma to help our product generate better-quality code. But a lot of issues with our product stem mainly from the fact that these practices are not followed just because the designers don't really need to. Ideally a product in this space should account for this issue in some way or another.
Getting accepted for the Y Combinator interview twice and rejected twice for the same reason Too many competitors. We are standing by our hypothesis that if we solve the market-wide problem of poor code quality, this will make us better than our competitors. But it's hard to convince the only investor you want that product quality is a key differentiating factor. Even so, we are still looking for ways to differentiate ourselves.
Notion - Knowledge management, Discord - User community & internal communicationStripe - Payment systemMixpanel - Product usage analyticsHotjar - Website usage analytics, Supabase - DatabaseAWS - HostingTawk.to - Customer supportEmailOctopus - Email campaignsCanny - Roadmap planning and sharingAspire - FinancesOsome - Accounting, Incorporation, Corporate Sec, etcNamecheap - Domain and company email
Right now we are focusing on improving our current product based on our customers' feedback. Once that is done, we want to look to build another product to solve a different problem in the design-to-code space.
Like this story? Read more stories of how founders started here.
The 100 Club is a private founders community to help you grow your network and make meaningful friendship with successful founders in Southeast Asia.
We offer a lifetime membership. No hidden charges or recurring fees.
$25 offer valid through 2024 (u.p. $100)
Apr 4, 2023
Oct 15, 2022
Jul 17, 2023
Nov 15, 2023
Feb 23, 2023
Feb 9, 2023