Today is a special day. My solo side project has hit 100K in sales, which is absolutely insane. When I started this thing, I never thought this would happen.
Monetizing an open-source project is very difficult, so I decided to share my personal journey and lessons learned.
This is a visual timeline showing all the lessons learned from $0 to $100k in one year.
I hope this can be helpful and inspire your own journey :)
Let me know if you have any questions! I'm here to answer every single one of them.
But it leaves one big thing out. You didn't launch with no one paying attention. You already had a fan base - people you could reach out to. I know you had to build your email list from scratch, but at least you know who was going to be in that list, and those people already knew and had a relationship with you.
You should go back farther in time to when you had no following, no people who cared, and no one who knew your name. As you said below, it's a 7 year overnight success. For those who want to replicate your success (not your product), what advice do you have for those who want to build a following so that later they can capture that following into something that will be beneficial?
That's a very good point that people often fail to consider!
I'm far from expert in this, but I'll reply with what my wife would say. She's in a completely unrelated, non-technical field and has a pretty good online following. In short: post things people are interested in, follow up with the ones who respond or ask questions and be genuinely interested in the people who follow you. Don't be fake.
The theory is simple, putting it into practice for your particular field is the hard part :-)
Did you get many repeat customers?
I see the licence is limited to 3 machines instead of the user. Did this have the effect of getting customers to buy additional licences?
It would be interesting to know if people are willing to pay again so they can use software on multiple devices or if it is more likely to put people off buying the software.
Congrats dude! I'm impressed by this project and that you got people to pay for it. Any why shouldn't they? It takes time and skills to do design work like this.
This is very cool. Do you have any insight into how many people are spending their own money to buy this theme vs. spending out of a budget their employer might give them? I could see a product like this being a no-brainer purchase for folks who might have, say, $500+ in productivity/gadgetry spend as a perk. I wonder if you could market directly to folks like that.
Unfortunately I have zero insight about how many people spent their own money vs used their company budget to buy this.
I should definitely do a better job by creating a template and allowing people to use that to convince their bosses. This site is a good example of that: product-toolkit.com
Your post gave an interesting idea - a gallery of products that people can buy to spend their budgets in case they don't have ideas lol
Just wanted to say congrats on the success and that I really enjoyed the post. The timeline outline of it all was a lot easier to read and follow than similar posts I have read around these types of things.
Btw, purchased!
Special request number 2, can you make a ZSH theme that doesn't require Oh-my-zsh? I hate using so much bloat when all I want is a cool terminal.
In 2001, Mike Cannon-Brookes and Scott Farquhar thought “Let’s sell an equivalent to the dozen of free open-source bugtrackers, but for a price.” Today Atlassian is worth about $60bn, they owned about 75%.
People have always said that ;) but Jira is super-expensive. It was $60 per user (permanent), and they’ve increased the prices a lot. Especially since you buy Confluence and all the little addons separately.
Especially now that Jira is removing it's self hosted option, someone should be able to come and fill that niche. It'd be surprising to me if that niche is unmaintainable these days.
The irony isn't lost on me, that at a previous employer, back in the early 2000s, I introduced the self-hosted version of Jira in to the company, because the alternatives were so bad and slow.
I find the most common source of Jira-hate is a combined refusal to leave many of its features on the table by management and general ignorance about how little Jira actually imposes on you by the gangpressed users.
Most of that bloat you have to deal with? Blame your bosses, not Atlassian.
I used Jira last year for a project at my work, set up a totally new environment and it ran like garbage. Each page load took seconds and every task required several page loads, often had to do a few extra because the UX was rather horrid and had to look for stuff.
One of the things I think about is how important scale is for evaluating the viability of a product.
I suspect that there aren't enough people who would pay $99 for a IDE theme to support an ongoing company with employees. However, for a single designer, it's perfect. The fact that 99.9% of the population turns their nose up doesn't matter because 0.1% is still 600,000 people. $100k is just over 1000 sales, which seems eminently doable for nearly any project.
My take-away is that - if you're considering a solo side project that will only work if you charge a "high" price (which probably isn't as high as you think it is) or if you get a ton of sales - try charging the high price!
From the article: "...it doesn’t work, unless you have massive amounts of funding or a brilliant idea that can completely disrupt the existing the market."
That's why I said scale of the project matters. Using a 1% (or 0.1%) metric to found a company is rubbish. Using it to guesstimate if a personal project might get some buyers seems much more reasonable. This probably isn't a $100k recurring revenue project, but it's reasonable to imagine that you could get up to $100k revenue selling a $99 product to programmers.
I never thought people would pay for a color scheme too. That's why, I tried to position in a way where you'd not only get the actual themes, but also a bunch of other resources to make you more productive.
Might be just me but once I saw that "x tips to get better" ebook, the offer lost credibility in my eyes. It just feels more believable to me that there might be some guy who knows so much about colors and the corresponding theory that he can actually can come up with a theme worth 100 bucks than that there is a guy who can do this _and_ give me tips on random semi-related topics.
True. On the other hand, $79 (or even the discounted price of $29) is pretty steep for what is only one minor optional tool in a dev's toolset. Imagine if you paid a similar amount for every single tool you use, no matter how trivial.
If you have a microtransaction for everything in your computer, they start to add up pretty fast. At some point you have to pick and choose, and nice UI themes start to seem less essential...
(Also, do note outside the first world $29 is not such a small amount of money. And $79 is unthinkable for a UI theme).
I disagree with the argument that developers don't pay for things. Yes, there's a huge open source culture where people can have access to free code on GitHub. However, there's a huge market of paid products that developers will purchase. There are many examples out there like Tailwind, Chakra UI, Envato, etc.
> not developers, look at GitHub - it's all for free
I do use the free stuff (GitHub, VS Code, etc).
And I also pay for NCrunch, Wallaby, RubyMine, Rider, GoLand, PyCharm, FastMail, PCloud, and more.
Most of the stuff I pay for has free equivalents (and yes, I know JetBrains have an all-products subscription rather than separate Ruby, .Net, Go, and Python, but I have reasons).
I have a heuristic for this. When something sounds stupidly simple, then try it if the upside might be huge (and you have free time on your hands, there's always an opportunity cost). The chance is higher that you're right, but in my personal experience, less high than I think.
Empiricism > rationalism/intuition [1]
But rationalism/intuition takes less time, so there's always a trade-off :)
[1] Unless you're an expert at the topic (see the research of Kahneman, Tversky and KLein), then there are many cases where your intuition might fit the model of reality better than many empiric experiments, especially with business/social things like this HN submission.
Wow, I created a color theme for VSCode some years ago (which actually got some attention by some people and there was one guy even adapting it to Google Chrome dev console) and not even one second I thought that this kind of stuff could be monetised in any way. Well done.
Same thing happened with me. I built this theme as an open source project and forgot it existed. People started to create different forks and port them to different platforms. One day I realized how big that was and decided to monetize. One year later 100k. Crazy!
I checked out the theme itself, and immediately a message showed up saying that it detected my country and offered me a discount code to make sure that the theme is still affordable for my country.
It's simple things like this that probably help a side project like this get to where it is, great job!
I wonder how does one put a price on a product like this. For example, if I am paying for a cloud VM instance, I could calculate the energy per CPU, hardware costs, etc. and ultimately come to a reasonable price. What would be the thought process for reaching a certain price for this specific product?
That's not really how pricing of anything works though. If you buy a cloud VM instance, you're not just paying for cost of energy and hardware costs—you're paying for the service and the R&D that goes into it. If you buy a $300 Japanese chef's knife, you're certainly not just paying for the cost of materials.
Everything is priced based on what people are willing to pay. Supply and demand.
I watched Brandon Li's video yesterday on how to turn a shot into a scene. I feel like it strangely applies to marketing for solo developers. You packaged your disperse config files into a cohesive product, with a unified look, and your website even has an emotional response to it.
Here's his insight.
How to turn a SHOT into a SCENE - Travel Video Storytelling
> I never had time to work on my open source projects, but now that I was monetizing, I was able to spend more time doing something I love.
That's pretty cool, but also only is true because he appears to love creating marketing pages, blogging, writing a book, creating themes for applications he doesn't use himself, etc.
I'm happy that he (/you, I see you're reading along Zeno - thanks for sharing), but it's not exactly what I'd have in mind for my "working on open source" daydream :)
It definitely took a lot of work not only from me, but also from many different open source contributors to get to this point. We are now at 164 themes which is absolutely crazy!
I'm so happy to hear you got to the checkout process. Thanks a lot for the support ;D
The refund part hits close to home. I've experience this too. What were some of the reasons for the refunds? I think theres so much to learn there. Some people are just jerks (they don't read what they're buying, and then blame you), but some people just didn't get what they expected due to some miscommunication, or product experience issue. I'd love to read more about this from this project!
It's definitely really hard to deal with refunds. You can make 10 sales in one day, but if there's only 1 refund, you might feel sad and disappointed.
About the reasons for refunds - you're right, many people buy without reading and then ask for refund later. Others might feel buyer's remorse.
In my case, the most common request was because of their personal taste. You see, I'm selling a theme for developers and they usually spend a lot of time using the same theme. It takes time to adapt to a new one color scheme, and some people ended up asking for the refund before they get used to it.
This is weird to me, because of all the free versions available...how could you not know what you are getting into when you decide to spend money on it?
From reading the name of the project I wouldn‘t have guessed it‘s a color scheme for a lot of apps.
I wouldn‘t even thought of that as a viable market. Congratulations to you. This is amazing.
I think the most valuable lesson here is, that there is a market for a lot more things than we would have thought a few years/decades back. Kind of how Spotify enable niche band to have successes outside the mainstream, just by allowing people to listen to every fringe corner of the musical ecosystem.
Are you generating the configs for all these apps? Having a method of turning a universal theme spec into specific app configs might be worth more than the theme itself.
It's really hard to build something like this since some themes like JetBrains requires a gradlew build (Java), while others might require some other languages. However, there are some attempts on this field already, https://themer.dev being the best one so far.
I accidentally half way started this type of project the other month when I wanted to quickly change themes along with toggling dark / light mode in a bunch of different terminal apps.
I got things set up to where I can run: set-theme gruvbox or set-theme one and it'll switch color themes in half a dozen apps. It also supports an optional --toggle-bg flag to flip between dark and light mode if the theme supports it.
But it wouldn't be hard to make the script more general purpose and put the configuration in a YAML file instead of in the script. Then add attributes for each supported app along with how those values would get replaced. If the app supports plain text config files it could realistically be done.
Congrats. Instead of thinking in the line of "Wow $99 for a config file with colors and fonts". My reaction is: So there is a premium market for this things. Good.
Congrats to the OP! One lesson in these graphs that is very important to internalize is how spiky they are. When you're small, you'll make most of your money on a few special days, based on events that might be out of your control.
Given that dynamic, it can be hard to distinguish cause from effect, let alone layer on things like A/B tests or other statistical analysis that might hold for things at bigger scale.
In this tweet [1] you mentioned that Dracula PRO was in the works for 7 years.
1. Would it possible to start monetizing the product earlier?
2. If yes, how much earlier in the case of Dracula PRO?
While $100k looks like an impressive number, in the grand scheme of things, it's just 14k/year, which is about $1150/month, which equates to an hourly wage of $6,50. Knowing what you know now, what would you do differently if you were to maximize revenue?
An amazing body of work.
I’m beyond impressed. Thoughtful, well designed, and well researched.
I never liked the Dracula theme much, but I’m going to snag this just to support great stuff being put out into the world.
Just want to say I am a very happy customer! Used the regular free versions for a long time and bought Pro the day I heard about it, in order to "give something back" both the free and pro versions are wonderful products that makes my work so much more pleasant so thank you.
I like to consider myself a bit of an amateur asthete, and I see tangible value in this kind of stuff so I have no problem spending a little bit of money to make my work day more pleasant. It's just as important to me to enjoy what I'm looking at, as much as it is to have a useful tool.
Congratulations on your achievements. Very nice idea and good website.
Having said that I noticed on your demo that the difference between selected text and non selected one it negligible. It is very unergonomic I would say. Funny thing I did notice the same trend on practically all dark schemes from other offerings. Themes for VS code for example. Curious why is that as the inability to clearly emphasize selected text would make my get rid of said theme immediately.
Dracula is a treasure! I like light themes better but I’ve been using dracula since I can have a consistent theme across emacs, kitty, and firefox without having to figure out their individual configs. I’ve started to use Dracula themes as reference designs, it covers so many configs.
That's so nice! I agree with you that having to figure out each individual config is a LOT of work. There's also a high cognitive cost on switching context all the time, that's when a standardize work environment makes the most difference.
Without knocking the creators achievement, I find it curious how devs will rage and cry murder when LastPass wants a few bucks for a hosted product but don’t bat an eye for shelling out $99 for what is, when you come down to it, a config file for your IDE .
Do you know that it is the same people holding these conflicting stances?
Seems like HN is a big place, room enough for both cheap devs and more well-heeled devs.
Intuitively I'd guess that they're very different audiences. I'm sort of near the type of person who would shell out for ease-of-use stuff like the same theme across editors, but also the type of person who would even want that.
I suspect a person similar to me would also pay a bunch of bucks for, say, a screen arrangement app for MacOS, or a way to more quickly pair AirPods (ToothFairy).
It's a type of user that has no problem paying for convenience, I suppose?
That said, $79 is too steep for me to pay for a set of themes, even though I do really like this one.
$99 for a great config file is buying time. A good config can take years to hone. Easily adds up to 50 hours over 10 years.
How much are your 50 hours worth? What about cutting that 10 year lead time down to 1h?
LastPass on the other hand is a tool you use because the world sucks. And their UX is pretty bad. And switching away is an annoying chore. You feel trapped and extorted into paying.
I configured my own editor and people often ask for details. I use LastPass begrudgingly because at this point I’m too lazy/busy to switch.
> LastPass on the other hand is a tool you use because the world sucks
I feel like this gets missed a lot when questions about price/value come up.
Just because something's useful doesn't mean you want to buy it, and things you don't want to buy, but need, are likely to be thought of more harshly when it comes to price.
This emotion seems obvious and I wonder why we don’t have clear, sustained global consensus for creating/maintaining non-profit alternatives for strict needs.
Is this part of mainstream economics yet? I see “the economist” argue around the edges of this, but then they ran a front page hit on Bernie sanders.
I guess because the more it looks and feels like a public service, the more it feels like a tax. Or the more it feels like a non-profit, the more it feels like charity.
Each of those concepts has had a long time to become embedded, but the kind of globalised, "net-good" non-profit that still gives you as an individual benefits, has barely even reached the corners of HN yet (would assume there's people here who haven't yet discovered LetsEncrypt, for example).
I have nothing neither against LastPass, nor for Dracula, and I've never used either so I might be wrong, but here are a few differences that IMO matter a lot:
- Dracula is FOSS. LastPass has some FOSS projects, but the key ones aren't.
- Dracula is a one-time-purchase, LastPass follows the terrible trend where everything must be subscription based.
- You can use Dracula when the server is down. You can't use LastPass when the server is down. Yes, you can manually backup the vault somewhere, but (a) there's AFAIK no way to automate it, and (b) you should be updating the backup every time you edit the database, which turns into a PITA.
As a customer I really dislike LastPass' business model. I rather use KeePass, and pay a hundred bucks for some Vim rice.
> - Dracula is a one-time-purchase, LastPass follows the terrible trend where everything must be subscription based.
Let's be fair here: LastPass needs to finance storage servers, network infrastructure, security engineers etc.. Dracula needs a CDN for downloads. Of course this is a bit simplified (the author needs the editors and a payment processer, for example), but it's a lot easier to work with a one-time fee when nearly all of your cost is up front and you have basically no running costs.
I think I am one of those. I generally don't like "subscription model" products. I am OK paying whatever price one time to buy something, but having to keep paying recurrently for some service incites locking to me, as the moment I decide I don't want to pay anymore I end up empty handed.
Yeah, I feel like paying a subscription to keep using software just feels like a rip off, even if it's actually a fair price, and am much less inclined to sign up for it.
I’m asking out of curiosity, as someone who is thinking about implementing a subscription model, do you not like it even though you know the product you’re using has recurring costs for usage (e.g. charges for API/DB calls), or are you only against subscription models when the cost of usage is minimal to none (e.g. MS Office, Adobe CS)?
Usually both of those costs are added to the product specifically to have an excuse to charge a subscription, which gets especially galling when 99% of users have no real use for it.
But the most hilarious/sad model has to be Elastic's "you need to pay us based on how bloated our garbage is, regardless of whether we're actually involved in hosting it".
I have no problem paying for recurring usage costs. For example I have a paid vps and paid email. But in my mind I'm paying precisely for the compute, storage, and network infrastructure behind them, not paying, e.g. to run linux. And I'm happy to pay for someone else to manage that infrastructure because the setup is nontrivial, I dont need to own the infra, etc.
Contrast this with Office 365, or even more niche pay per use or per month products, that are not better in any way that matters to me because they are hosted elsewhere. With these products I feel like someone wants me to pay a recurring fee just to run their code. Some of it is psychological, but I feel like I'm getting ripped off, that someone is trying to find a way to get me to keep paying for something (which they are).
Spotify and Netflix have found a good balance of offering a subscription, but providing such a large catalog that the value vs. actually owning all the content is clear. Most SaaS feels more like having to pay monthly just to own a single DVD.
But TLDR for me is I dont want to pay recurringly for the privilege of executing your code, I will pay recurringly for needed* infra, support, etc that goes along with it.
*not just tacked on to make it SaaS or to deploy it as SaaS
So if a SaaS offered a prepaid consumption plan alongside a monthly subscription you'd be more likely to buy?
Like $19/month unlimited use or $50 for x amount of usage (api calls, transactions, assets, etc.)? Then you decide if you want to reup when you've used up your credits?
That's just a lesson in promising things to your customers - if you never promise free nobody bitches; changing things you've already delivered is the simplest way to a marketing problem.
Personally, I have no problem paying for it, but I hate the bait and switch crap. I'd rather pay someone who has always been paid, then wait for lastpass to decide to strip more features unless you pay up more.
I had the same question. But when I saw how many apps he supports, and it comes with decent fonts... I'm quite tempted. I'm normally one of the cheapskates. But getting consistent and good themes across my desktop? That's a hell of a lot of work for something I'd really appreciate.
FWIW... I guess, the paying customer segment might be different. Since he seems to be selling the bundle, he may not be knowing the demographics. I vaguely guess the paying segment might be from the designer ecosystem like Sketch or Figma.
For me, it's one-time purchase vs. subscription. For example, I was happily paying 1Password every couple of years for a long time, and left when they made it clear subscriptions were the future.
This is a SUPER important topic. I don't think Dracula Pro would be as successful if it was a subscription. There's definitely a subscription fatigue going on, and people who invest on one-time purchase models will be able to benefit from that
If people are buying it then the argument can be made it is both useful and fairly priced to the customer. It’s not like the company has a monopoly on IDE configuration.
Looks like the set of developers who cannot edit their own config files and the set of developers who are careful enough to use a password manager are mutually exclusive.
Today is a special day. My solo side project has hit 100K in sales, which is absolutely insane. When I started this thing, I never thought this would happen.
Monetizing an open-source project is very difficult, so I decided to share my personal journey and lessons learned.
This is a visual timeline showing all the lessons learned from $0 to $100k in one year.
I hope this can be helpful and inspire your own journey :)
Let me know if you have any questions! I'm here to answer every single one of them.