Hacker News new | past | comments | ask | show | jobs | submit login
Why we bootstrap (flagsmith.com)
119 points by annared on Sept 26, 2023 | hide | past | favorite | 85 comments



I think one argument against bootstrapping (as someone who is trying this myself with my code review assignment and notification system "Friendly Fire") is that the expectations are higher now. Gone are the days of hacking something ugly together over the weekend and it going viral right away. Just look at the early screenshots of Twitter - you don't see even Alpha products too rough around the edges anymore.

You spend 90% of the time doing the first half, and then 90% again for polish - because you have to. They can talk all they want about "you need to go to market before the product is even usable", but if you don't have the VC runway to keep failing and still being able to pay rent, for bootstrappers it's not an option. You get to roll that dice once.

People expect few bugs, good UI and UX. There is a LOT of upfront work. I made sure that my tech choices have almost no overhead, and it's still more than a full-time job.


I have heard this argument a lot. Let me tell you it is categorically false. I run multiple bootstrapped software businesses which culminated in a $4 million profit last tax year. After many many years of failure.

The biggest mindshift required here is you need to really think what can technology solve. Most products being developed by bootstrappers are competing in a space where the problem is already solved. For example the nth job board. Yes in that case if yours look shitty people won't use it. But that is not a problem at all if you are solving a real problem.

I will give you one example, and this is me sharing one of my products for the first time: a custom mobile app for housekeepers in rich families in the UK. The mvp for this was built over a weekend using just basic ios tools. Important thing to understand in this market (custom apps for companies) is that you have to go after companies for whom the branding difference is whether they are number 1..3 or 4.. and 4.. companies die due to competition. So if you can in anyway help them stand out for a large annual contract, they will lap it up.

There are many opportunites like this. Instead if you are building the next app that you are seeing online, ya you are too late.

The other thing is alpha products that are rough around the edges still exist. Forget the Twitter screenshots. When you used Twitter for the first time, did it look like that screenshot? No it didn't. Because by the time it became popular enough, it had been polished a bit.

So what you are doing is comparing a product which is popular now vs a screenshot of the thing before it was popular. If you actually look at the first alpha version of things of popular apps now before they were popular, you will see that they are still rough around the edges. Maybe a little bit less so as a lot of ui frameworks are available now.

Also older stuff look rough around the edges even more because ui trends themselves have changed.

So don't fall for this myth.

Solve something hard using technology with a ready to go ui kit and get the v1 out there as soon as possible.


I dunno man. I sold a small company back in 2011 and it really was easier back then. Three out of four of our first leads tried to buy us because we were doing "real" machine learning. This was almost a pure API app. We didn't even have documentation we just told people we'd integrate it for them.

It is harder these days. Also, the product you mentioned "custom mobile app for housekeepers in rich families in the UK" I don't know how you're marketing that, but that seems like the hard part to me.

As for Twitter, people forget what it was like back when they were first starting. There were a ton of people you wanted to follow, but RSS couldn't handle the sheer numbers involved. Blogging had become so big and realistically you could only follow, say, 10 bloggers before being worn out with all the content. Twitter made it possible to get little nugget sized updates from thought leaders without having to eat the whole sandwich of some massive blogpost.


> I dunno man. I sold a small company back in 2011 and it really was easier back then. Three out of four of our first leads tried to buy us because we were doing "real" machine learning. This was almost a pure API app. We didn't even have documentation we just told people we'd integrate it for them.

You're kind of agreeing with their point. Your example of success was where you were (at least perceived to be) solving a legit need with a real solution.

I agree with the grandparent - people will use a crappy looking app if it solves something they're desperate for (and by implication, isn't solved for already in the market).


I was disagreeing that it is easy these days just as it was back then. Back then it truly was much easier to get traction with a new product. Basically, we're post-land-rush now. There are still opportunities, but back then it was really easy to get through doors and startups were celebrated and a lot of easy opportunities weren't yet seized on the web yet.


Absolutely agree. Everything is much easier now to build, but harder to market/compete.


You are talking about a pretty dire need and a glaring vacuum in the market. While that's the best-case product market fit, it's not like a good execution for a moderate need in a contested space is not a thing at all.

Can I live with text files in folders as my notes? Yes. Would I rather use something like Evernote instead? Absolutely. Would I use it if it gave me problems all the time? Doubt it.


> good execution for a moderate need in a contested space is not a thing at all

That's kinda exactly what this means :) -- don't contest that kind of space if you don't have the funding and great design/etc package.


I think it depends on the market your entering and the problem your solving.

I've brought some very, very unpolished products to market, but they were in a completely new category. They got traction because there were few-to-no other options and they were all the same level of "poor" polish. That's the the market tolerated.

Likewise, ChatGPT is arguably not a particularly polished product. Even more so when it came out. It was a slow, somewhat buggy chat experience with no bells or whistles. Overtime, they've iterated and iterated and iterated.

If ChatGPT came in as "just another search engine", they'd have been laughed at from day one. The power and flexibility of LLMs would have been completely lost against the benchmark of mature, decades old search engines.

----

In my opinion, the value of the product you built is the ease of use. All of the major source control platforms have basic levels of assignment and notification. Your system must provide significant value on top of that or be a pleasure to work with.


Right but you're basically proving his point with the ChatGPT example. There was no way that was coming out by bootstrapping. They've received tons in funding and pay-for-use was sort of a built-on based on all the hype about what it could do compared to the competition.


When did you ship those products you mentioned? In the last 2-3 years or prior to that?


I’m actively shipping one now. Parts of it need to be super polished. Others, less so.


> People expect few bugs, good UI and UX. There is a LOT of upfront work.

You might have a VC-funded product that has 0 bugs, perfect UI and UX and still nobody is using it. I've seen this in practice, even a decade ago. I believe that bootstrapping actually forces you to think about how useful the thing that you're building is to people. And if it's useful and in a good niche, users will work with you through a couple bugs. That said, some products are harder to build in a bootstrapped way than others.

Also, having show-stopping bugs in your app is as bad today as it was 10 years ago no matter if you're bootstrapped or VC-funded. It's rare that people use something that doesn't work.


> I believe that bootstrapping actually forces you to think about how useful the thing that you're building is to people

You can think all you want about it, but until it's in people's hands, it doesn't matter. And when it is, per the earlier comment above, the expectations - in many many many markets - is that the UI/UX matters. IMO it matters far more than it should in early days, but that's just me.

Functionality that solves real problems is ignored because ... well... could be any number of reasons. Didn't like the colors. No dark mode. Something not obvious to people. Not using their language/terms.

> some products are harder to build in a bootstrapped way than others.

Yes - the 'ease' of connecting to an audience of decision makers, and their expectations re: usability... it can be very different between markets and services. And again, per the earlier above comment, expectations are simply much higher than years ago. Competition is stronger in many markets, and often the competition is inertia. "Learn a new system" vs "keep emailing excel sheets around" is still a real thing in a lot of places.


UX/UI matters because people are used to it. You have to have an insufferable need for a product to deal with bad UI. In the olden days, users didn't know better. HTML forms were primitive, now they are server-powered applications in their own right.

The users have been spoiled by slick, polished products. This used to be NOT the case. You could build something stupid out of boredom in a few days and walk into riches.

Frankly, the HN comments here sound quite condescending and snooty, telling me to find a narrower niche, be a better entrepreneur, and just maybe I don't know how to write code.

Not the point.


Not true. A certain segment of the population (early adopters) do not care about polish and these are the customers you're looking for. You really do not want to spend a long time polishing something that nobody cares about. This is the YC advice even today.


If you have VC backing (they force their ventures to use each other's products), or launching a product as a known brand (Meta, Threads), yes.

Try doing that when nobody knows who you are and you ask your potential customers to connect their GitHub account to your app. You need to at least make sure it does not look like a shoddy hacking operation.


"If you have VC backing (they force their ventures to use each other's products), .."

I never met a VC that did that. Perhaps 1 VC in 20 does that? This sounds more like an excuse.


This depends a lot on the product. Some products, companies, markets, etc simply NEED capital to get off the ground. Hardware for example is fairly pricey on average. Tech's evolution has made starting software companies and products really cheap where it used to be fairly expensive. Perhaps the pendulum is swinging back in the other direction a bit. Perhaps its SO easy that in some spaces, you need the money to differentiate your offering from the hundreds of others trying to do the exact same thing in order to find that success.

But speaking as someone even lately who has produced new products in the last few years, the products don't need to be perfect. The PG (I think?) anecdote about PMF where your product should feel like selling water to someone who is lost in the desert and dying of thirst. You can DEFINITELY achieve that with a sub-optimal, still a little unpolished/buggy product. I have, myself, just in the last few months. You just need the right product and the right market.


I don't necessarily disagree with you, but it's also very easy to have something that "looks nice".

It's pretty trivial these days to go get a Bootstrap Template and Tailwind Template, often times of the type of product you're trying to make anyway, and just wire together some backend stuff.


Of course you can build something that's easy on the eyes - but you also need to build the product, the value. My point is that you need both, and you need to do it lean, without getting into the toolchain swamp that is modern UI development, because you will never be nearly done that way.


I'm not sure I'm following you. Of course you need a product. You can solve whatever problem it is your customers are trying to solve and have something decent in terms of ui/ux is my point. The UI/UX expectations like you've mentioned have gone up, but so has the ease in meeting that expectation. It's a no-op unless you decide to do everything by hand without using the ecosystem of tooling that exists these days.


And yes, hobofan, I have done recent UI development. I use CSS Grid as it essentially makes Bootstrap legacy - and I have used that since version 1.

I use ArrowJS and ES6 directly, even lighter than anything you mentioned. You are probably just assuming I am technically incompetent, but I've been down a few rodeos in 20+ years of doing this.


Honest question: Have you recently done any frontend development? I think your assessment might have been valid 3-4 years ago, but nowadays if you keep to an established framework (e.g. Next.js with Tailwind), frontend development is about as simple as it was during the Rails + Bootstrap era.


IMHO that’s a lame excuse. Of course you can’t get by with a twitter 0.1 style MVP. But then again, twitter was built in its current context with the options they had available.

We, currently, have a ton of nice things that they don’t have (thanks Bootstrap) to build very quick MVPs that look pretty darn decent.


> We, currently, have a ton of nice things that they don’t have (thanks Bootstrap)

Originally named Twitter Blueprint.


I think it's a sliding scale. The more unique your product is, and the more pain it solves, the more users are willing to put up with hard UI.

I've seen a lot of bad ui make a lot of money, because it found a market segment that had no other options.


> People expect few bugs, good UI and UX.

People/companies expect solutions for their problems. If the solution has bad UX and is still buggy but solves their problem "well enough", they'll usually pay for it.

Also empirically, you can still regularly see products both on HN and Product Hunt by "indie hackers" that end up being quite successful and were incredibly rough on initial release.


This is absolutely correct, in my view. I spent the last 4 years trying to bootstrap with a partner. I’ve kept up the day job to keep the lights on, and provide some limited support to my partner whilst they commit 100% to writing code. They are a junior developer, so there was a lot of learning, and slow going, which was fine and accounted for.

Product 1 was a coding tool for a low/no code system that is really very good and provides amazing functionality, but most devs (the target audience, really) are - unsurprisingly - put off by the GUI based approach. We built a tool that allows developers to drive the platform using a DSL we developed. This was built based off the IntelliJ community edition platform.

The majority of the two years time we worked on this was spent on polish, and there was no end in sight. It is functional and really fills a solid niche, but the product is not something I would consider ready for production use by other developers. Given the amount of work left to be done, the possibly imited target market even when finished, and the growing realisation that to make this profitable we’d have to go enterprise B2B (again a massive hurdle) we left it at that, and started looking at something else to build. We spent a year on product 2, think saltstack, but better, but my partner was already on their last legs, emotionally and financially, and they called it quits a few weeks before v1.0 of product 2, leaving me in a position where the product is nearly there, but we are unable to cross the finish line on account of me not being able to quit my job and work on this full time. It is beyond frustrating.

In hindsight, there are many things I could have done different or better, but bootstrapping is hard for anything but the simplest products. I have two major products that are totally cool but unfinished due to the need for polish, life altering bills left over from this adventure, and I am facing a job/project market that is aggressively ageist (I’m officially an Old) and no outlook on better professional and financial prospects (boo-hoo, poor me, I know).


Simply wrong info


What is insane to me is seeing people turning things which are pretty simple to build - like feature flags - into products which can hit $1.5M in ARR. This is pretty trivial to build once you have given it some thought, it sits nicely next to the work you do building health-checks into the various tiers of your infrastructure, yet there are enough people out there willing to pay for it.

I guess that this is why I don't have a bootstrapped startup - I wouldn't even consider building something like this out as a SaaS.


It goes both ways:

It's incredible to see something that would take 2-3 days make $1.5M ARR :(

It's incredible to see something that would take 2-3 days make $1.5M ARR :)

Transcend ego and come drink your fill at the font of easy money. Don't let pride hold you back from what is probably the best ROI you can make as a SWE.


As someone who introduced our org to a SaaS feature flag tool, we did it cause the cost of building out and maintaining the system in house would have been too costly. In addition we needed both FE and BE flags (multi language support).

In addition, I needed to transform the culture fast, and not wait around for people to implement a custom library for internal use.

Having said that, I have built in house flag implementations or used OSS projects and would much prefer those to using a third party.


There are a lot of things that are trivial to build. That's often not the question at a startup. At a startup, it's "what is the most valuable to build". Implicit in this is often, "what's unique to my product that I must build".

If I can take X to drop in a 3rd party feature flag system, that is almost always going to be more valuable than building it in house. It gets me back to mainline work.


I agree mostly but I would say it's a bit more nuanced. Instead of what's unique to my product, maybe also consider what's unique to my market. Business is about selling stuff. If you are selling to a unique subset, all you need is the same feature set but which feels different.


Haha we (Author here!) thought feature flags would be a pretty simple product to build 6 years ago. 6 years later we are still hard at it!


Building a feature custom to one company is usually simpler than building an entire product that meets the diverse needs of many companies.

It may very well be that the polished product is a better fit for many businesses than the quickly put together custom feature that does the bare minimum, though.


This. You don’t need to create the first thing in a category or choose something complex to be successful.

You may be able to hack together a basic feature flagging product in 2-3 days, but like many domains, it goes incredibly deep and there’s a lot of hidden nuance if you focus on it. If the value doesn’t drop off for those long-tail “deep” features and people will pay extra for you to handle the nuance - voila, you have an amazing SaaS, potentially.


Oh as a platform it's certainly a different level compared to building something for a set of products (versioning etc for a start). The thing is, if you build something yourself then it's an order of magnitude or two less complex. Unless you have little / no skills in backend.

That's the market; it's shocking (to me) how little knowledge even the largest and most highly rated "Mobile Agencies" actually have with systems which have a lifetime measures in decades vs months. Silly as I've worked with these "wordpress agencies" a lot and they've always been thin/wide skill wise (devs skills which are teachable in a few months but with designers + marketing added to the mix).

Your story is inspiring, thanks for sharing. I hope that I can eventually get over my blind spot as it is a career goal to build a small/medium sized company (with annual revenue around the $1M+ mark).


> This is pretty trivial to build once you have given it some thought, it sits nicely next to the work you do building health-checks into the various tiers of your infrastructure, yet there are enough people out there willing to pay for it.

I think the reason you don't have a bootstrapped startup is because you don't see the difference between trivial and valuable.

Would it be easy for you to build out a feature flag app? Yes, most devs probably can without a huge problem. But it will take them time to do it, and it will require maintenance over time. If it takes a senior dev two weeks of dev time in sum total over 4 years, it would be "cheaper" to pay $100/mo for this service. This type of product product has value as it allows us to spend our limited dev time more on forward feature development that drives future revenue and less on infrastructure.

This is the math that companies buying SaaS products will make. I don't want my eng team to spend hours on trivial stuff, I want them to work on the things that move the needle for the company. I'll gladly pay $$$ for certain features and functionalities that are not part of our core competencies so that the team can focus on the things that are. And once we're a Big Company paying Enterprise tier prices to a feature flag company or whatever, we can have the chat about in-housing it or building Feature #137,221


Well, some of the main players are companies like launch darkly. Which don’t have the most stellar track record. It’s easier to stand out when the competition is mediocre.


What's wrong with Launch Darkly's track record? I don't have much experience with their product, but was considering it for something upcoming.


The tough thing to about bootstrapping for people working in BigTech is opportunity cost. Assuming you do not work on your startup while working your full time job, the revenue you'd have to hit to justify the effort is generally just too high. Considering even this post: if you had a team of ex-BigTech consisting of just 4 people, collectively they were probably making at least a million assuming they were all located in a HCOL area in United States and were mid-level. Making 1.5ARR wouldn't be worth it considering that there would've been a long period of no revenue.

In a sense it's better to be broke because the opportunity cost is lower and incentive is higher. Golden weights slow you down. This is why most BigTech founders want VC money to make the reward potentially bigger and justify the effort.

On the other hand, if you make a lot of money - some types of bootstrapped companies are slow burn and you can subsidize yourself and just wait until it surpasses your income, but it's hard to have a healthy life balance.


Yeah, working at FAANG and investing your income is basically the highest return on a risk adjusted basis for engineers. In the last decade, there are many cases where people built a startup, exited for 10M, and still made less than they would have had they climbed the ladder in a FAANG and held onto their equity.


You don't start a company because you should and it is the most optimal choice with the current info.

You start a company because you have to. You have that idea that you must build or you cannot stand the corporate hypocrisy of large tech companies anymore (my case).


But what’s the opportunity cost of dealing with BigCo politics while your soul withers away in a meeting? There’s other goals beyond money, most bootstrappers want to live life on their own terms, even at a salary cut.


I find it fascinating that certain circles (cultures?) consider it a failure if your tech company doesn’t get a round of funding from the most “prestigious” SC VC firm.

Currently working in one of those tech companies right now, has made me appreciate my friend’s bootstrapped tech company, as it can evolve naturally with common sense decision making.

Edit: there was an article posted on HN two years ago or so about a SV tech founder coming to this exact realization after having laid off all his employees because he couldn’t get the next round of funding. I felt sorry for him and all others who were/will be in the position.


> Edit: there was an article posted on HN two years ago or so about a SV tech founder coming to this exact realization after having laid off all his employees because he couldn’t get the next round of funding. I felt sorry for him and all others who were/will be in the position.

Gumroad. Also typo, unless you actually meant South Carolinian VCs.


Thanks, yep that’s the one!

As much as I love South Carolina, yes I meant Silicon Valley. :)


There are founders that think their goal is to get funding to hire people to build their product for them and measure their personal success by how much theyve raised


This is roughly my startup's goal, slow bootstrapping with no VC cash. We also started as an "agency tool" that I built for my business partner. Over time we realized "hey, other folks in the space could use this too". It's been about a year and a half since we came to that realization and were prepping to go live in the next few weeks.

I'm really excited for what will happen next year but I'm also have to manage my expectations, and always tell myself "there's no guarantee this will work". The only silver lining to this is also knowing that because we're not taking VC cash and have a low burn, we have more time than most to "figure it out" before the thing goes bust.


According to Crunchbase and some other sites flagsmith did raise funding. Is Crunchbase wrong or is taking angel money considered bootstrapping nowadays?

https://polychromecapital.substack.com/p/our-investment-in-f...

https://www.axial.net/forum/want-to-grow-your-software-busin...

According to Axial Flagsmith is a 50/50 joint venture with a private equity firm:

> Then came Polychrome. Through a VC connection Rometsch was introduced to Polychrome Capital, a long-term investment firm focused on helping small but budding software companies grow their businesses through market outreach and strategic expertise. The chemistry between Rometsch and Polychrome’s founders was immediate via phone, and in September 2020 – during the midst of the pandemic – Rometsch and the Polychrome founders structured a 50-50 joint venture, before they even met in person.

> “We set up the deal so we didn’t get that ownership stake until we had fulfilled our promise of growing the business,” said San Francisco-based Alex Boswell, a co-founder and partner at Polychrome. Initially, Polychrome offered Flagsmith 25% equity in exchange for a cash investment, and after Flagsmith hit a certain milestone, the other 25% equity was granted. Two years later, Flagsmith has grown by 20-times revenue, he noted.


Hi there - my name is Matt Althauser and I’m a partner/founder at Polychrome. The funding that we brought to the company was $250,000 which we valued as the equivalent of the value of 3 years of development that the existing team had invested into building the initial open source project.

For us (polychrome) at the time, the money came directly from the bank accounts of myself and my two co-founders. It was pretty scary for us all!

In that sense; we do think of ourselves as bootstrapped. We (polychrome) did not (and still have not) take a dollar in compensation for the business. We’ve worked directly on the business for 3 years with the founders. I have been selling the software in Europe and my partner, Greg, in the US/Asia until we got revenue high enough to hire our head of sales, Mike and head of marketing, Anna. While doing this, we (Polychrome) paid ourselves by doing consulting for venture baked companies. Ben (Flagsmith CEO) was being paid by the business and his existing agency.

This is definitely not venture. FWIW - I have built two venture backed companies - Optimizely & Amplitude which took typical venture rounds and would not be bootstrapped IMO.


Thanks for clarifying. Many people (including other commenters) wonder how people manage to transition from open source project to a profitable company. I think the blogpost would have been much better with this extra background information. Normally, when people read stories about bootstrapping they assume the original founder(s) retain all the equity, but in this case Ben had to sell a 50% stake in his business in order to survive.


For extra clarity, the $250k + sweat equity made us 50/50 partners. Again - not an Angel or venture structure.


This is how your partner describes your PE firm in a podcast interview:

> Polychrome is a PE firm focused on acquiring 50% or greater equity positions in B2B software companies. That’s our primary focus and that’s what we do pretty much every day. [...] So we are open to full buyouts, we’re open to partial buyouts. Our core structure though, is typically keeping the founder in place. And we are typically partnering with technical founders who don’t have, the go-to-market expertise or who just don’t want to sell or the go-to-market is what’s keeping the company from taking that next step in their growth trajectory. And so what we’ve done and what our typical structure is we’ll do cash, let’s say for 25% of the business.

When you talk about "typically keeping the founder in place" you're not 50/50 partners. You're the senior partner.

https://blog.acquire.com/startup-acquisition-episode-26/


Wow that's impressive! Though I always thought bootstrapped was more that the product pays for itself from an early phase, not that you use debt/investment to fund a big make or break effort that's going to need a show lot of revenue fast.


The way I define bootstrap is if the money is a Debt that needs to be paid back and you personally are on the hook. If you are not required to pay it back, it is not bootstrapped whether you used your own money or someone else's.


I’ve been very impressed with the flagsmith team. Met with them a few months ago when evaluating feature flag vendors. Their features and pricing were great, but the team put them right at the top of our list. Really transparent and technically knowledgeable. We are an enterprise customer that moves ridiculously slowly, so we aren’t a customer yet… but we will be.

Congratulations!


VC backed startups are just an indicator of an accelerated planned obsolescence for customers that they should stay away from.

Gone are the days of the VC era of 'blitzscaling' and cheap money, bootstrapping looks to be the way to go really.

I hope that more companies reject the predatory nature of the VC world rushing to get an exit.


Bootstrapping until you reach the first step on the revenue latter is hard, especially if you don’t have a big network, lots of followers on social media or money to buy you the most costly resource, time.


It's much easier if you also have paying consulting clients to bring in revenue alongside the bootstrapped project.


Friendly question about this kind of company from the perspective of employer competitiveness: Is known that sweet-VC-money + equity (debatable) is the edge for Startups, BigCo the brand, and BigTech Money + Prestige.

From a pure employment perspective, what are the clear unique advantages of working in some of those bootstrap businesses?


Just guessing: better working environment and less stress? Working for a company that need to raise every x month to continue paying salaries and looking for crazy grow rates could lead to a lot of stress.


(Author here) That's a great, great question, and it is difficult.

I think I might write this up as a separate post as there's a lot to unpack. There are pros to working for a bootstrapped company, but they less concrete than the pros of working for someone with a massive B round.


Thanks for your prompt reply. I was just wondering about it because I worked 10 years ago in a bootstrapped company and of course, it does not have a substantial financial upside in comparison with liquid RSUs or payment premiums due to VC money; but was a pretty decent working experience.

At least for me was more or less felt less about an employee in the sense of not having a lot of Kool-Aid as a VC/larger company, and more about having a clear boundary that I was selling a service in a stable for-profit company in a specific niche.


I think it's would be useful for someone to list out good reasons for and against bootstrapping: When it works and when it doesn't.

For a few months I was in discussions to be CTO of a company that was kinda-sorta bootstrapping. (They had a very small amount of angel money.)

Ultimately, they wanted me to quit my job without pay; I realized they were 3 cents short on a dollar; and we parted ways.

Given their space, they really needed to raise money for what they want to do. (More than the fact that I need to get paid,) The space they're operating in needs someone to grow fast because there won't be a lot of room for lots of small players.

But, if this was five (or more) years ago, I think bootstrapping would really work in their favor. (Different market conditions and expectations.) I might have even been able to take a lot more sweat equity at the time, too.


I'm always impressed where people manage this. I bootstrapped an open source project for years before raising venture capital and still never succeeded. In the years prior I had tons of side projects that went nowhere. Turns out if you charge people for labour, it's a great starting point, but if you can then use that as an insight for a product, even better. A great example of that is basecamp but also Shopify. These things spun out of web agencies. Build things based on a need and it'll translate into something bigger, even profitable.


Thanks for the kind words. A LOT of the Flagsmith design and approach was built around solving problems that our consulting clients had (basically big UK blue chips doing cloud transformation).

Starting with an agency can be a bit like enabling cheat mode in many ways.


Yes if you can build that insight whether from inside an existing company or external client it makes a lot of sense. Sometimes hard for Dev centric founders to turn that into a real business but the opportunity is there to learn and then execute on it as you've demonstrated.


> But you can also do really well in a market without that competitive advantage if you find your niche.

I can't stress the importance of this enough. Everything mainstream gets canibalized by endless venture capital money, killing competition and stifling innovation, and, most important, drasticly reducing the odds of success for independent businesses. Finding a niche and sticking to it is the only option. As some said: start small, stay small. I am thoroughly impressed by people that manage to achieve this.


Turning an agency into a SaaS is a often-sought dream for those in the agency space (because it's tiring to keep the sales machine going, handle clients and their needs, constantly put out fires, etc etc etc). Grats on getting a SaaS out of it!

I think from a bootstrappers perspective, there are two things to take from this:

1. Having a source of income that gives you the "margins" to also do a SaaS (time and money) is a big, important (almost required) factor 2. This might be an outlier - Not a lot of agency owners successfully convert over to a SaaS - both because SaaS is hard-mode business and because agency life is so distracting with the aforementioned work native to agency-style business.

Just because you see a need as an agency doesn't mean customers will pay for the SaaS version of the "thing" that fulfills that need. Flagsmith appears to have found a product that is a SaaS that people pay for.

A question on their distribution: I wonder what their customer acquisition story is - is it from their agency work (self-referrals) or what?


> We believe enterprise software is not a winner-takes-all game. With the current software landscape, there's room for multiple players in most software markets.

There's room and then there's room for 10 different players https://openfeature.dev/ecosystem


B2B software markets that have yet to consolidate:

- CRM

- databases

- IDEs

- security

- marketing email

- project management

- social media tools

- helpdesk

- expense management

- shipping

These are all markets that are over 20 years old, where there has been ample time to consolidate to a small number of providers. Markets are bigger than normally thought. This is especially true for bootstrapped companies, who can do well financially without having to ever hit a $1b revenue run rate.


Facts. I'm in logistics (last-mille) and there are lots of players but also lots of demand.


When you bootstrap, there can be room! We don't have half the board demanding 50% market share...


One reason for this post might be the fact that feature flags and related software/startups (Posthog, Launch Darkly and now Flagsmith) came up a lot here recently, such as

- https://news.ycombinator.com/item?id=37611136

- https://news.ycombinator.com/item?id=37547550

- https://news.ycombinator.com/item?id=37520294


Good. VCs—particularly the growth expectations thereof—have ruined more products than they have helped. By far. When we talk about enshittification, we're generally talking about the results of choices that came out of this process. Bootstrapping has its own problems, but I would never go the VC route again, or work for a company that did.


The awesome thing for bootstrap startup is that if they go well you can get 100% of earnings and you are able to distribuite your dividends whenever you want. If you are VC-funded you have lots of constraints in this. You are almost pushed to continually re-invest your earnings every year and only get a salary.


After reading this thread, looking through the side, and through LaunchDarkly's site, I'm still confused. Can someone ELI5 me through what a feature flag is? Ie how does it compare to a struct or database field?


Remote control to turn things on or off. Example: turn on new cool beta feature for 5% of users. Code ships with ‘if (new_cool.enabled)’ checks everywhere.


They enable experimentation: show this page variant to these users, that variant to those users, and make sure neither group overlaps with a third group in another experiment...

It is an application of authorization.


Compared this to Launch Darkly for personal use but only thing that steered me away was you don't get real time streaming on the lower tiers like LD. There is a lot of small biz SaaS that have enough customers to need it but not be enterprise scale.


We are building out real time streaming and do have it in beta on our JS SDK currently (hit me up if you want to get onto the beta).

HOWEVER!

In our experience, it's one of those features that people think they need, but then when you dig into the use cases, most people don't need it.


Yeah I can understand that for sure. Personal project and a friends business I consulted had the use case at that time.

Was thinking about the architecture reading your code and imagined I'd build that at its core and go with a more event/usage based model. Great business, product, and success story none the less. I'll keep it in mind for future projects. Nice job.




Consider applying for YC's W25 batch! Applications are open till Nov 12.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: