Lessons from Bootstrapped Companies Founded by Software Engineers

We hear little about bootstrapped companies, despite bootstrapping being an effective way to get up and running. We cover five successful bootstrapped firms you’ve probably not heard of – until now

Hi, this is Gergely with the monthly free issue of the Pragmatic Engineer Newsletter. In every issue, I cover challenges at Big Tech and startups through the lens of engineering managers and senior engineers.

If you’re not a subscriber, you missed the deepdive on Cloud development environment vendors and a few others. Subscribe to get issues like this every week👇

The Pragmatic Engineer usually covers Big Tech and startups which are often venture-funded businesses. A group of companies which tends to get very little coverage in this newsletter and across the media landscape, is bootstrapped ones; firms that take no external funding.

This can be by design, as many of these companies prefer to fly under the radar, and focus on building a sustainable, profitable business, and don’t seek a bigger public profile. In fact, bootstrapped startup TalkJS (offering embeddable messaging experiences, UIs and APIs, competing head-on-head with the likes of Stream and Sendbird) at the time of writing this article told me, “we're going to have to politely decline an interview at this point for focus reasons.” I respect how the company stayed heads-down, focusing on building for their customers!

Another reason for the lack of media attention is that a well-run business simply isn’t very newsworthy because it’s inherently undramatic, and the media runs on stuff that is! When a company raises a large sum of cash, it’s newsworthy, and when it announces a major hiring drive, that also is. In contrast, there’s not much to cover in a bootstrapped business that didn’t raise funding or go on a hiring spree, and did no layoffs when the economy turned.

A final reason is that the media landscape is full of stories which someone has got in the news by using money or influence. Companies can and do buy media coverage. You can hire PR agencies to pitch stories to media outlets, and this practice is so widespread that the line between genuine news and paid-for promotion is sometimes blurry. I’ve learnt this fact because I get press releases in my inbox about funds raised, product launches, and so on. I ignore practically all of them.

In this issue, we spotlight five lesser-known bootstrapped companies, founded by former engineers. Why? Venture capital funding is in decline, and so I expect bootstrapping to be an approach more entrepreneurial software engineers will take, especially because the barrier of entry for a software company is very low if you know how to code and build a first version of a digital product.

So let’s jump in. Today, we cover:

1. An overview of five bootsrapped companies 

2. Taking the leap to bootstrap a company

3. Tech stack and engineering approaches

4. Growing the company

5. The contrast to working at a large company

6. What works, consistently?

Five things that worked, consistently, for five bootstrapped companies.

1. Overview of the companies

When deciding which companies to include in this issue, I wanted to include more established companies that have been operating for 10+ years, as well as newer ones. I also aimed to include ones from both the US and Europe, which share a few things:

1. They were founded or cofounded by former software engineers, some of whom formerly worked in Big Tech

2. They have not take external investment

3. All are profitable

4. … and yet, you may never have heard of them, until now!

The businesses:

Five bootstrapped companies: Fern Creek Software, Ticket Tailor, Formspree, Friendly Captcha and Secta Labs

​​Fern Creek Software is the oldest company, founded in 2001 by former web developer and team lead M. Keith Warren. It’s a boutique development agency, focused on greenfield development. The firm serves clients ranging from startups to Fortune 500 businesses. An interesting fact about the company after 20 years is that it’s still a 100% engineer-driven firm. Even the salespeople are engineers who write code!

Ticket Tailor is an event ticketing platform, founded in 2010 by freelance web developer Jonny White. The company brings in around $6M per year in revenue, with 20 people. Ten of these folks work in tech; 8 are developers, and 10 others work across various business functions.

Formspree is a form backend, allowing the embedding of forms with no backend code. It was founded in 2015 by Rohit Datta, formerly a software engineer at Apple and Cole Krumbholz, formerly a product manager at Squarespace. The company employs 7 people, of whom 4 are engineers, and uses external agencies for things outside their core expertise, like design. One of the seven employees is a fractional Chief Marketing Officer (CMO.) This concept sounds pretty similar to the fractional CTO role we cover in this issue.

Friendly Captcha is a GDPR-compliant bot protection service. The company was cofounded by founder and (now former) director of a digital agency Benedict Padberg and former Lyft software engineer Guido Zuidhof in October 2020. The company consists of 13 people now (with many part-time), of whom 5 are engineers all based in Europe. 

Secta Labs was founded in December 2022. The company built and launched a product to generate professional headshots from selfies, using AI tools. It was cofounded by former chief product officer Marko Jak, and former software engineer Alex Kotliarskyi, who was previously at Replit and Facebook. The company consists of 6 people: 3 work in tech, and 3 others focusing on AI research and customer support. Secta Labs launched its first product soon after founding, late in January 2023. The product went viral and has since generated more than $1M in revenue, and has more than 22,000 customers.

As always, I aim to stay independent in my coverage. I have not been paid to write about any of these companies and am not affiliated with any of them. See my ethics policy for more details on how I maintain my independence.

2. Taking the leap to bootstrap a company

From here on, founders discuss their business approaches. Each question is followed by answers from the companies. 

How did you take the leap to leave fulltime employment and go all-in on your company?

Fern Creek Software emerged from the wave of layoffs caused by the dotcom bust of 2001:

M. Keith Warren: “I graduated high school in 1997. My first job was fixing computers;  then I joined a medical software company and became a coder. I wrote one of the first SaaS apps in existence. This company was then sold.

In 2000, I was employee #2 at a startup that raised $10M in funding. They did something similar to cloud computing, but we failed. I then ended up leading the software dev team for a web design firm in 2001, when the dotcom bubble burst. My company did a round of layoffs, and was starting their second one. 

The truth was, I just needed a push to quit employment. I was thinking of starting my own company, and already had my first client lined up. The second round of layoff was the push I needed; I offered to quit, which also meant someone on my team could keep their job.”

Ticket Tailor started as a side project:

Jonny White: “I was a freelance web developer building websites and software for clients.

I wanted to stop selling my time. I saw the opportunity that event organizers weren't satisfied with using ticket agencies. They charge high fees, own your data, and brand the whole ticket buying experience – it's a really bad deal.

I built an online ticketing site as one of several side projects. When it started to take off, I gradually let go of my web development clients. It was a low risk approach.

I didn't set out with any ambition other than to stop selling my time. I never thought it would achieve the scale it has today. When I reached £2,000/month revenue I was a bit clueless as to what to aim for next!”

The story of Ticket Tailor then took an unexpected twist, as the company was acquired. But two years later, its founder regained full control.

Jonny: “In 2012 – 2 years after starting the company – I got an offer from Time Out magazine to buy the company. I sold it and joined Time Out as a product manager. The intention was to integrate ticketing into the Time Out website. In the end, it didn't play out that way. 

After two years, I left. As I was leaving, I wanted to cut my notice period short. However, Time Out didn't know what to do with the still-running Ticket Tailor business! I offered to cover the legal fees and buy it back and they accepted.

Time Out selling my company back to me is the kind of luck you can't plan for. In the end, I came out having paid off some of my mortgage, and still had 100% ownership of the company!”

Formspree also started as a side project:

Rohit Datta: “I started working on Formspree as a side project, back in 2015. At the time there weren’t good ways to add a form to a static website. I built something very simple that let you send an email from your form. 

Back then, I was still in school, and my cofounder, Cole had a full time job. Still, this side project kept growing. For three years, we worked on it, during nights and weekends. In 2018, my cofounder quit his job to work full time and I followed shortly after. 

By the time we took the leap to go fulltime, the company was already able to support us. This de-risked the decision to go fulltime.”

Friendly Captcha kicked off thanks to savings its founders put aside while working in Big Tech:

Guido Zuidhof: “With a bit of savings, almost anyone working in tech could try their hand at starting their own thing! Having a large compensation at a company, however, is also what actually stops people from trying too, sadly.

I quit Lyft to work on open source projects. I built Friendly Captcha because I needed it for a different project. In the months after launching it wasn’t much of a success. ‘Build it and they will come’ is not true. A few months later, my now co-founder Benedict reached out saying we should turn this into a real business. That was exactly what the company needed - someone that could actually turn this side project into a business.”

Secta Labs came out of experimenting with GenAI image models:

Marko Jak: “Since I moved to the US I’ve been thinking about what I want to build next. I’ve been playing with and fine tuning open source models since Stable Diffusion 1.4. Late 2022, Alex and I launched a WhatsApp bot that does something similar to photo editor Lens just to explore the space. The application of using AI for headshots surfaced during these experiments. We launched the headshots product in January 2023 and it completely blew up!“

3. The tech stack and engineering approaches

What is your tech stack, and which engineering approaches do you use?

Ticket Tailor:

Jonny: “We started out as a PHP/MySQL MVC monolith in 2009. Today, we are still a monolith, but we are modularizing it.

While refactoring is satisfying, as a developer it’s a challenge to move as fast as we’d like to in building product features and completing architecture migrations. And it’s just as challenging to avoid a migration “tripping up” a feature we are working on!

Cloudflare workers is a technology we’ve found surprisingly useful recently. There's a bunch of problems we have been able to solve at the edge network level, rather than in our codebase: like dealing with the challenge of queuing high-demand events. We also use it for simple things like serving translations for our marketing website.

We tend to do at least 1 deploy per engineer, per day. We have 8 engineers on the team and you will often see more than 10 deployments go out per day. We serve over 10M requests per day. To put our scale into perspective: in Q2 this year we processed 19% of the ticket sales volume that Eventbrite did globally in the same period.”

Formspree:

Rohit: “Our stack is this:

• Frontend: a React app that uses Tailwind for CSS

• Backend: Python, using Flask

• Fraud/spam: for spam and fraud prevention we use machine learning to scan submission content and metadata. We build this on top of standard Python libraries like scikit-learn. We cover more on this in the article The machine learning toolset.

• Database: AWS Relational Database Service and Redis ElastiCache

• Hosting: AWS. Compute: ECS Fargate

Luckily, we moved off Heroku right before its security breach in 2022. We moved off because we outgrew the platform, and we also needed a higher level of security for our SOC 2 certification than Heroku was able to provide.

We do not complicate our software, unless absolutely necessary. I’m a big proponent of keeping things simple for as long as possible, and using speed to your advantage. We were conscious of introducing complexity, and did it to achieve better redundancy and more security benefits. At the same time, we don’t use any more complicated control plane systems that are in use at Big Tech companies (looking at you, Kubernetes!) We keep the configuration as simple as possible for as long as we can.”

Friendly Captcha:

Guido: “When bootstrapping, don’t choose complicated tech that can break in a hundred different ways. For example, stay away from  Kubernetes or excessive use of microservices. Use proven tech that a small team can understand and be productive with. We use:

• Frontend: Typescript, Vue SPA (customer dashboard).

• Backend: Monolith written in Go.

• Database: The trifecta that allows you to build anything: Postgres, Redis and Clickhouse.

• Hosting: Scaleway, Hetzner, Cloudflare.

• Misc: Github Actions, Terraform, OpenTelemetry. ”

Secta Labs:

Marko: “Serverless was the tooling we used because we had to move quickly. And it really helped us do this! The technologies we use: 

• Frontend: React NextJS, hosted on Vercel

• Database: Planetscale

• Backend: Inngest for running background jobs, AWS for storage, and AWS Lambda for workflows

• Reliability: Sentry for error logging and and Datadog for monitoring 

• AI and GPUs: Modal.com for GPU (graphics processing unit,) serverless and Google Cloud for our GPU instances and large model training “

Fern Creek Software has a somewhat different approach, being the only developer agency among the five bootstrapped companies:

Keith: “We work in the Microsoft world: C#, TypeScript, Azure, Acce​​ss, and so on. We also build native iOS and Android apps, and work with AWS at times.

We all love to learn new things, but sometimes we have to make “boring” choices. I have seen lots of fads come and go over time, and have been pretty lucky not to get caught up in them. 

For 95% of software we build, a simple monolith with easy-to-understand code is the right choice. Could we be more fancy? Sure, but to what end? We have a responsibility to the people that come after us and maintain things.

I tell our team all the time that they will never impress me with clever code or architecture. A simple to read and understand chunk of code that solves a problem clearly and effectively; now that is the manifestation of mastery!”

4. Growing the company

How did your company grow, over time? 

Fern Creek Software:

Keith: “In the beginning, I was the only one working at the company. I then brought in contractors to help and I still love to code to this day. 

For a decade, I really struggled with what I wanted the business to be. This was from 2002 until 2012! Revenue grew steady and I stayed with the contractor-only model for many years because that’s what I knew best. 

Back then, most of our deals came from word of mouth, or via subcontracting to larger firms. We built a good reputation in the Business Intelligence (BI) and data space during that time. “

Ticket Tailor:

Jonny: “I started off hiring engineers, as I was seeing a lot of natural growth and product development was the biggest bottleneck. 

I started working with a coach to help develop my management skills, and he helped me realize that I needed to hire an exec team to achieve the ambitions I had. So in 2020 – 10 years after founding the company – I hired a head of ops and head of growth, and so the other side of the business was born.

One of the biggest challenges I have had with growing a team is answering the question, why are we growing this thing? We don't have investors, we are very profitable, so what's the point, and at what cost are we going to go to try and create growth?

My motivation today is to see how far we can take the idea of “business for good,” balancing profit and purpose. I wonder how many more people would be inspired to become entrepreneurs if they knew it can mean so much more than just making money.”

Formspree:

Rohit: “While Formspree, as a project, has been around since 2015, most of the growth happened after the Covid-19 pandemic and lockdowns in 2020. 

Our growth has happened in bursts. We’ve had times when we were featured in prominent blogs, and yet we didn’t see much change in traffic or revenue. And then the opposite also happened: we’ve seen strong growth without any smoking gun!

We focus on long-term trends to grow the company, and try to zoom out more from the minutiae of KPI movements.”

Friendly Captcha:

Guido Zuidhof: “We built the business first, and worried about hiring much later. This is the exact opposite of the VC playbook, where founders should be spending a lot of their time hiring or raising money, and not growing the fundamentals of their business! 

‘Do things that don’t scale’ described our early strategy. Many of our early larger customers decided on us because they really liked talking to Benedict [the CEO of the company]. Nowadays we are more established and customers often reach out to us, which is something you can’t count on when just starting out.

After the first year of the company we could pay ourselves a salary, and now three years later we are a team consisting of a dozen people or so (with most part-time).”

Secta Labs:

Marko: “We realized early on that our product had an inherent word of mouth component. So, we focused on creating a great experience and generating images that looked authentic to the user. We still have many new customers who use our service after trying competitors because of this. 

We later supplemented our growth channels to include influencers and affiliates, and added paid ads and search engine optimization (SEO).”

5. The contrast to working at a big company

What is the biggest difference between running a bootstrapped company, compared to working in a large business?

Ticket Tailor:

Jonny: “It was so slow to get anything done! I don't think I could ever work at a big company again. The experience was invaluable, but the speed was so sluggish. From my limited experience, I would say we deliver at least 10x faster than big companies do.

I never want to be a small cog again, nor do I want anyone at Ticket Tailor to be one. For now, we have capped our headcount at 25, and we won't go over that.

When I was working at a big company I did a lot more mindful stuff. I was into meditation, gardening, and found time for crafty projects. I still do creative stuff in my own time, but running my own company takes up a lot of headspace. Being the one responsible for the whole company impacts the quality of my personal life. I’m trying to address this especially as I have young kids, but the opportunity with Ticket Tailor just feels so big, right now.”

Formspree:

Rohit: “At Apple, I was a software engineer 100% of the time. Meanwhile at Formspree, I really enjoy that my job is a mix of software engineering, business development, operations, and so much more. 

I get to wear a lot more hats than at Apple. There’s new priorities that come up at every phase which challenges me to adapt. If you enjoy a dynamic, challenging work environment and want to grow, then founding a startup is great for experiencing how to run a business.

Instead of working fixed hours, I frequently spread out my work throughout the day and week. This is a distinct part of Formspree culture, that comes from how my cofounder and I prefer to work. 

We encourage our team to choose their own hours, so if you don’t feel like working a particular afternoon, you can just take it off. On the flip side, when you feel the itch to get something done on the weekend, it’s ok to push code. 

Ultimately, when you’re starting a company, your “bosses” are your customers! And they care about results, not how many hours you work.

I do miss having easy access at Apple to world-class experts in their fields. At a startup, everyone needs to be a generalist to some extent, whereas a company as large as Apple pushes services to their limit so much that they need experts in their area.

As an example, at Apple, we built a solution to provide end-to-end encryption in Kafka. I had almost no experience with encryption at the time, but Apple’s deep resources let me lean on world-renowned security experts, who helped me get up to speed quickly, and verified that our solution was secure and scalable. As a smaller company, when we engage in deeply technical problems we don’t have domain expertise in, we have to spend more time learning about how to apply concepts to the problem.”

Secta Labs:

Alex Kotliarskyi: “The biggest difference between working at Facebook or Replit is how things feel more ‘real’ here. I can trace everything I do back to the customer's problem just by following the money.

In both Replit and Facebook it felt that we had more ‘safety cushions’ – whether this safety cushion cash was venture funding, or ads revenue.

On one hand, this type of directness is really scary — if things don't go well, it's very clearly reflected by the bottom line. And we depend on our bottom line for paying others in the company, and our own bills!

On the other hand, running a bootstrapped startup feels more grounded. We are not running on a future promise of value, but rather on something that's useful and valuable at the time of transaction.”

Alex also shared details on how the engineering culture is also very different:

“We don't do mandatory code reviews like Facebook and Replit do. We don't have product managers. We have to rely on a lot of 3rd party infrastructure for things like compute scheduling, monitoring and marketing tools.

I had to dial down my tech purist views and get more pragmatic. For example, we adopted Tailwind: and I used to dislike Tailwind back in the day! We are staying away from overdrinking the ‘monorepo kool aid:’ because if you decide to go with a monorepo, then you are left with configuration hell for JavaScript tools!”​​

6. What works, consistently?

What is the thing that has worked from the start, for your business?

Fern Creek Software:

Keith: “Ship. Deliver for people. Shipping solves everything!

Ticket Tailor:

Jonny: “Keep it simple. Steve Jobs has a good quote on this:

‘Simple can be harder than complex. You have to work hard to get your thinking clean, to make it simple. But it’s worth it in the end because once you get there, you can move mountains.’

Simplicity is hard and requires being ruthless, but it allows so much scale.”

Formspree:

Rohit: “Make life easy for your devs. As a developer tools company, we invest a lot of time into the developer experience. We have things like:

• A React library to automatically detect duplicate form submissions

• A common form schema that works across our wide range of third-party plugins

Us, founders, are also developers. We started this company to scratch an itch that we had. We understand the problems developers have, and how we can make their life better.”

Friendly Captcha:

Guido: “Focus. Do one thing and do it better than anyone else.

Build a business first, ‘tech startup’ second. Think about the long term and what is sustainable for that timeframe. Don’t work 60 hours per week: take holidays at times. Hite kind people and be the best employer you can be. If your goal is to become a unicorn and sell to Google, ignore this advice.”

Secta AI:

Marko: “Create a “wow” experience. We worked hard so that when the user gets their gallery, it’s a ‘wow experience’ and makes our product remarkable and shareable. We built a few viral loops around this to make sharing easier, fun and worthwhile. 

This and combining our work into a teams plan, which is a natural progression for companies when they see the results.”

Takeaways

Bootstrapped companies will only become more popular and widespread in the coming years. Venture funding is trending down, but fortunately you don’t necessarily need investment to start a tech company. There are so many successful, bootstrapped companies out there, and we’re covered only five of them.

Things I’ve learned from these businesses include:

• Bootstrapped companies grow much, much slower. They don’t hire ahead, and only hire when they feel the pain of having too few people, and a hire is justified.

• They are rarely very big. While there are exceptions, the majority of bootstrapped companies I’ve seen don’t grow much beyond 30 people. Of course, some do grow beyond this, such as Zapier, which has had very little funding – $1.3M – and has grown to 800 people. But this is the exception, not the norm.

• The tech stack can be really cool! I used to associate “bootstrapped” with uncoolness, but looking at these companies’ tech stacks, there’s no reason to think this. These companies use the latest frameworks when it makes sense.

• They could offer more stability during a downturn. There’s no investor pressure to reach profitability targets, which is not the case with publicly traded companies or VC-funded ones. Of course, this doesn’t mean no bootstrapped company will let people go. However, the point about stability chimes with bootstrapped companies being the most likely to increase salaries in line with inflation, earlier this year.

• There’s an important tradeoff: compensation vs at Big Tech and VC-funded startups. The elephant in the room is that most bootstrapped companies pay less in total compensation than Big Tech and publicly traded companies do. And unlike Big Tech and startups that offer potentially life-changing equity, bootstrapped companies do not have this compensation component. In return, they often offer a lot more flexibility, full remote, and a more chill working environment.

Engineering choices seem – dare I say? – more pragmatic. Alex Kotliarskyi from Secta Labs said he had to dial down his “tech purist views” from when he was at Facebook and Replit, and instead just choose “good enough” tools. At Fern Creek Software, Keith also said how he noticed how it’s VC-funded companies that aspire to technical elegance, and over engineering things.

Could it be that engineering with a focus to support hypergrowth scenarios – aka building to scale – is a luxury that bootstrapped companies simply do not have? And so they are forced to make a lot more pragmatic choices: like choosing “boring” technology. I have a sense that being bootstrapped comes with this constraint. So reducing focus on technical excellence could be one of the biggest challenges for anyone leaving Big Tech or scaleups to found, or work at a bootstrapped company.

There is no “good” or “bad” between VC-funded startups and bootstrapped businesses; each has different tradeoffs. I hope this article provides some insights about what to expect at startups with no external funding.

If you work – or have worked – at a bootstrapped company, what is your experience?

Thanks very much to all five companies for sharing details and learnings. A few final notes from them:

• Ticket Tailor: if you would like Jonny to speak about bootstrapping at your event, or the principle of business for good, you can reach him at jonny@tickettailor.com. Johnny also asked me to convey that businesses that have been affected by the recent price rises at Eventbrite should take a look at their low-cost offering – business economics are easier with a team the fraction of your publicly traded competitor!

• Friendly Captcha: for bootstrapped founders in Europe that have seen some traction already, Guido is happy to chat. And for those based in Europe who want to join efforts in saving hundreds of millions of people from Google ReCAPTCHA: they are hiring. 

• Secta Labs: the team is building out an AI research team to work on some hard problems in this space around ethnicity-specific fine tuning, RLHF, and multi finetune  compositions. They’re interested in chatting with machine learning engineers with experience in generative images. Contact Marko for details.

And Fern Creek Software and Formspree is also hiring on-and-off, and growing responsibly.