Proxycurl is almost three years old, and we are marching towards 5 million (USD) in annual recurring revenue (ARR). In this post, I will tell you exactly how we got here.
Proxycurl is a Software-as-a-Service (SaaS), and we sell data enrichment services on people and businesses to other businesses looking to build data-driven applications. Our customers include brand-name VC firms, top banks in the US, and various pre-IPO startups. And we are happily bootstrapped and profitable :)
Did I also mention the cryptocurrency exchange I built and failed or the paywall-busting Reddit alternative I made and almost got criminally charged over?
I did not come from money.
A common theme in the startup world is that founders who do well come from money. However, I most definitely did not come from money. Not that I am doing well, but I wanted to share my background with you.
My mother works as an assistant at a humble hawker store, and my estranged father works as a delivery guy at Foodpanda. I am based in Singapore and grew up in subsidized government housing. I stopped taking money from my parents when I was 15.
However, I am very lucky. I was born in a first-world country and have had access to computers since I was a teen. I am also a reasonable programmer. I have been coding since I was 12, graduated from university when I was 26 with a computer science degree, and I built a few products with millions of (freemium) users and five digits in Monthly Recurring Revenue (MRR) less than a year after.
I always knew I was going to build a company. I am on my fourth company. The others before had shut down.
With a significant appetite for risk, I invested 100+K SGD of whatever money I had saved in cryptocurrencies in 2017 and sold them for my first pot of gold last year (2021) at the peak of the crypto-mania.
I also sold a partial stake in my last venture, NuMoney, to the Indonesian partner.
On top of that, Proxycurl is generating healthy profits.
Sure, I did not come from money. But today, I am financially de-risked from the outcome of Proxycurl or any future endeavors. My point is that successful outcomes do not necessarily require old money or "pedigree."
I was funding six employees through the income of my side projects, and revenue was dying. I needed to make money pronto.
NuMoney was a pan-Asian cryptocurrency exchange with the vision to lubricate FX and fund transfers between Singapore, Malaysia, and Indonesia. I had to shut NuMoney down in 2019 because it was crypto-winter, no one wanted to trade cryptocurrency, and we needed to make money pronto! Our Indonesian partner proposed that we do the same for cryptocurrency investments but for real estate in Indonesia. That means selling partial equity in real estate to laypeople in Indonesia. The idea was that Bitcoin is volatile, but real estate is not. The Indonesian partner then took over the reins as the CEO of NuMoney, renamed the product to LandX, and I took a step back to assist with product management and development. Long story short, I stopped believing in the executive team of LandX, and its business model. The original NuMoney team whom I had hired followed me to build our next venture.
The following product was SharedHere. The sad thing is that I don't remember what SharedHere does. I remember being motivated by my hatred for paywalls. We built SharedHere for a few long months and all that I got out of the venture was a fraudulent police investigation. My local newspaper (Singapore Press Holdings) instigated a fraudulent report about me hacking their servers. The investigations were dropped, and SharedHere still was not making any money.
I was funding six employees through the income of my side projects, and revenue was dying. I needed to make money pronto.
Then I remembered a trick I have to scrape LinkedIn profiles at scale. I knew LinkedIn data was hard to scrape and is valuable. What if I can offer fresh LinkedIn data on tap via API access? I took a month to build it alone while my team worked on SharedHere. When it was done, I told my team that our work with SharedHere was done because there was no trivial path to revenue.
And that from then on, we will be working on Proxycurl. Proxycurl was born out of desperation.
Build it, and they will not come.
I have a programmer's disease. I like to build, which is why I created the prototype of Proxycurl before I talked to any customers. Building products is the easiest thing to do. The hard work comes after launching Proxycurl.
I knew I had to talk to as many customers as possible. So I purchased a Crunchbase Pro subscription and performed a search for all recently funded companies in the space of marketing/sales automation, recruitment tech, or web scraping. Then, I exported the results in .csv format with their corresponding general emails and began a series of cold email outreach to their general emails.
It worked; I sent thousands of cold emails and received hundreds of responses. Then, with the help of Calendly and Google Meet, my calendar was booked full of calls for a whole month.
Slowly and surely, we got some users. And then some revenue. It turns out that some big-ass venture-funded companies were willing to listen to a nobody pitch a LinkedIn scraping API because they, too, were trying to do the same and failing at it. Soon, we made a few thousand dollars in monthly recurring revenue (MRR) in the first few months.
At the time, I also had a habit of writing a daily blog post to practice story-telling and communication. With the launch of Proxycurl, I transformed my efforts from writing about startup stories to SEO optimization.
With SEO and cold-email outreach, we grew the company to our first million dollars in revenue in 2021.
Building a moat
Sure, Proxycurl was a LinkedIn Scraping API at launch. It is not a sexy business, and one of the many lessons I learned from my failures is never to build my castle on top of another. So it became my mission to transform our business away from just scraping LinkedIn profiles to being a data enrichment service provider. Today, we make the majority of our revenue from activities other than LinkedIn scraping.
What we have done is turn our Achilles heel which was LinkedIn scraping, into our strength. Instead of shying away from LinkedIn scraping, we leaned into LinkedIn and embraced the fact that LinkedIn profile URLs are the best identifier for people and business entities. I cannot think of any other identifier system for people and companies with wider coverage and a global presence.
With LinkedIn Profile URLs as source inputs, we aggregate fresh and rich data on these entities from a myriad of sources, make sense of them and sell access to these rich data to developers.
In the long run, we want to be the richest and freshest (private) data store about people and companies worldwide. So we can power use cases such as KYC, credit scoring, investment signal identification, etc., through an easy-to-use API.
I am going to start building in public.
I am going to be completely honest. I am starting to build in public because I want you to know about Proxycurl. I want you to think of Proxycurl the next time you have a problem that we can solve. I want you to know that we exist, that we are a great place to work, and that you can trust us because you have seen me writing about our stories and know that we always deliver. And finally, I want you to know that Proxycurl does not have a grow-at-all-cost sickness because we are bootstrapped, profitable, and have no need to burn money just because some VCs pressured us to.
I also thought I could offer Proxycurl's growth stories from an angle less commonly shared on Indiehackers. Sure, there are plenty of stories of companies starting up on IndieHackers. But from what I understand, there are not enough stories of scaling up -- particularly as a independent entity that is bootstrapped.
I am scaling Proxycurl up as we march towards 5MM in ARR, and I will journal my progress and struggles at every step on IndieHackers. However, I might not be 100% transparent on every business metric, particularly on our product roadmap, strategy, and revenue. This is to fend off competitors. I am taking a leaf out of the lessons shared in the article "Farewell, Building in Public" by Cory Zue.
Next up: I will tell you how well this post performed on IndieHackers.
I created this brand new account on IndieHackers and spruced up my Twitter profile. I will post this post soon on IndieHackers and on our blog. In my next post, I will tell you how well this post performed and if it moved the needle in our metrics.
I will also tell you more about our recent failure in SEO. Follow me on Twitter or subscribe to our mailing list here if you want to be notified of my next post.
In the meantime, feel free to ask me questions in the comments below or tweet at me!
