25
8 Comments

Generating $300/mo in Passive Income With an API for Proxy Servers

Hello! What's your background, and what are you working on?

Hello! My name is Kaustubh Pratap Chand and I work remotely as a software developer. I am mostly self-taught and I have been hacking (in the nerdy sense) since I got my first computer in 7th grade.

I develop and run SpinProxies, which is an API to fetch public proxies. The whole system works without any active intervention and generates $300 of passive income a month.

SpinProxies is a tool I developed primarily to solve the one critical issue I faced while working on a web crawling project. I needed access to fast and cheap proxies and only way that was possible was to use public proxies. Most of the other tools I tried were not up to par. So I created SpinProxies with the hope that I could solve this problem for other people like me.

Home page

What motivated you to get started with SpinProxies?

I have worked on several data analytics systems and sometimes I have to write crawlers to aggregate data.

Fetching data is the first and most critical component of any data pipeline. Some data may only be accessible at a particular location, or perhaps the website you are targeting doesn't allow the type of crawling you want.

In all of these cases, you are going to need proxies which can solve these problems by transforming location, identity and speed.

You can rent private proxies but you quickly rack up bills as the size of your traffic grows. So I tried the plethora of public proxy lists available online and my experience was frustrating. Most of the proxies just didn't work or were extremely slow.

This was an issue I was going to face recurrently in my projects, so I thought, "Why not develop a solution that actually works?"

What went into building the initial product?

It took me around two weeks to get the system up and running. The code is all in Python. The crawler continuously scans the top ten public proxy sites and extracts proxies from them through a generic parser which can identify the structure of the page and tag all the relevant data.

Currently, it crawls around 100,000 pages per day.

These proxies are then sent to the monitoring module which tests each proxy on on a number of parameters like their speed, location, and anonymity. Since there are thousands of proxies to be tested every minute, I use the asynchronous Python framework, Eventlet, to speed up the processing without consuming too many resources.

If you think that your product solves a problem, then you should just put it in front of the world as soon as possible.

Since this an I/O bound problem, an async framework like eventlet works very well. While the code is waiting for the results from one request, it can fire up another request in the same thread. This helps in keeping the CPU usage low and keeping the server costs under control.

I worked on it during the weekends, so I didn't have to compromise on my typical work schedule, although I did have to pass on those Friday beers.

It's not your typical Indie Hackers success story, but the system requires very little manual intervention from me and generates enough to make a significant dent.

How have you attracted users and grown SpinProxies?

The user growth has been organic. I basically just launched the website and gave a free trial with limited number of proxies. This lets the users try out the product for themselves and see if it fits their needs.

Initially, I hired a few backlink services on Fiverr. But those had an insignificant effect with most users coming via organic search.

There are also a couple of other things which have helped in creating an authority:

These were very low-effort initiatives, but I believe they have been successful in their purposes.

If you don't have real feedback then there is a high possibility of putting your precious work in the wrong places.

Other things such as manual postings on forums, reddit and Twitter have also helped in getting the word out.

As a programmer, I was so fascinated with the coding part that I didn't put the effort into marketing as much as I should have. But this is an area which needs heavy improvement. I am considering new features which can be useful and it is imperative to market the product well so I can reach more people.

Pricing

What's your business model, and how have you grown your revenue?

SpinProxies' main source of revenue is the subscriptions. Users subscribe to one of the three plans with each plan differentiated by the types of proxies available. PayPal is the payment gateway I use, but I am considering adding cryptocurrency payments, too.

Other sources of revenue are some reseller websites which pay a monthly commission for the permission to display the proxies from SpinProxies on their own website.

As of now, SpinProxies is making about $300 with about 500 total users, including non-paying users. The only expense I have is the server hosting. Currently, there is a $35 VPS setup running on OVH and it works perfectly with the optimized code.

What are your goals for the future?

There are a few pain points during the process of crawling which I am looking to solve. Those are mostly programming challenges and I look to bring them online soon.

Another major opportunity for growth is SEO. I am considering going long on content marketing by adding useful materials on the blog which revolves around the crawling/proxies domain. The idea is to get people interested in SpinProxies when they search about topics related to this domain.

Blog

What are the biggest challenges you've faced and obstacles you've overcome? If you had to start over, what would you do differently?

The most challenging part was the SEO.

I have a product that works, but how do I make others realize the value of this product?

I realized that the best strategy here is to make the key product feature the main selling point. I target using the "Fastest Proxies..." phrase because I think that it's one of the solid points about SpinProxies and the most important point anyone is looking for in a proxy.

What's your advice for indie hackers who are just starting out?

Start fast.

If you think that your product solves a problem, then you should just put it in front of the world as soon as possible. A lot of times, we get trapped into chasing that mythical perfection which is always evading us.

As a programmer, I can understand the tendency to polish the product as much as possible before you put it in front of the world. But if you don't have real feedback then there is a high possibility of putting your precious work in the wrong places. Feedback from real users is the only way to know for sure where to put your work.

It's not your typical Indie Hackers success story, but the system requires very little manual intervention from me and generates enough to make a significant dent.

If it doesn't work, then you are still one step ahead, as now you know what doesn't work and consequentially, you are one step closer to what actually works.

I also believe that there are lots of untapped opportunities lying around in the B2B sphere. There is a lot to make by helping solve the problems of people who solve other people's problems.

Where can we go to learn more?

You can go to spinproxies.com to give it a try. The API documentation gives a quick start to get this tool working for you.

When I have something interesting to say, I post it here on the blog.

Please let me know of any questions or suggestions in the comment section below.

  1. 1

    Nice work. One observation though. Spin Proxy has a free 5-day trial. But it is not obvious. Initially, I thought it was a free tier. You should mention clearly that it is a 5-day free trial.

  2. 1

    How many enterprise users do you have? I feel like $30/mo for an enterprise plan is really selling yourself short. I'd shoot for at least $50, knowing how much support people tend to expect when thinking "enterprise"

  3. 1

    Really nice story! How did you build everything that is not the core app? I mean login, site, apidocs, and API key generation?

    1. 1

      The web app uses python-flask. It has some pretty good integrations for things like mail, login among other things.
      It has its advanatage in being quicker to get it up and running.

      For the api docs, I used slate to generate the docs from markdown.
      However, when I rework this, I will go with swagger based spec now.

      1. 1

        Hi,

        We are trying to reach you on [email protected] for a project we are working on. Can you check the email and respond?

  4. 1

    This is awesome! I'm a software developer as well and I'm also a digital marketer.
    I create and manage a team of editors and writers that produce top-notch blog content and then I do all the SEO to get those articles ranking in google, obviously and preferably on the first page of a search. I'd be happy to talk about SEO and give you some advice.

  5. 1

    How many hours of work do you reckon you’ve put into the project?

    1. 1

      The first 2 weeks was when I made it a primary focus. Around 6-7 hours/day.

      After that, I have kept working on it intermittently. The frequency now comes around to be an hour per month.
      This will increase in the upcoming days as I have some new ideas which I want to implement.