Hacker News new | past | comments | ask | show | jobs | submit login
I started a one-man biz that's beating VC-backed startups (distrokid.com)
1237 points by pud on Feb 15, 2017 | hide | past | favorite | 217 comments



@Pud can undeniably build products that are laser focused and useful.

And his minimalist approach to company building is epic and I am a fan.

But it's sort of disingenuous to play David Vs. Goliath here like his line about launching:

"I figured only a small number of smart people would somehow find us among the masses."

He's just like the rest of us - just build a great product and be lean guys!

But oh wait he launched DistroKid on the back of the HUNDREDS OF THOUSANDS of registered musicians already using his other site - Fandalism - founded years before.(1)

EDIT - why I am a fan of pud is in the children to this comment he brings a data-gun to a knife fight and sorts out my impression w some facts.

(1) https://techcrunch.com/2012/04/06/philip-kaplan-fandalism/


A similar situation: Jason Grishkoff built and grew his music submission site SubmitHub[1] "on the back of" his popular music blog Indie Shuffle. People said it therefore must have been easy, but that's not actually the case. He spent 4 backbreaking months sending over 1000 personalized emails to sign up the initial collaborators he needed to make the SubmitHub marketplace possible. The moral of the story is to be wary of succumbing to availability bias[2] and underestimating the work that goes on behind the scenes.

The other lesson is, as Pud said in reply to you, always be parlaying. Spending years building up an audience that you can market to might sound like "cheating" to those of us without such an audience... but then again, spending years acquiring programming skills before starting an online business might sound like "cheating" to non-developers. In reality, it's simply a smart idea, no more and no less, to leverage existing advantages like a unique skillset, valuable knowledge, an audiences, personal connections, etc.

Specifically, building an audience is a great precursor to launching an online business. Growth is hard, and starting from zero makes it harder. And anyone can build an audience! I recommend checking out Ben Gelsey's analysis of Product Hunt's rise.[3]

[1] https://www.indiehackers.com/businesses/submithub

[2] http://www.businessdictionary.com/definition/availability-bi...

[3] https://medium.com/@theunixbeard/product-hunt-s-rise-d49249a...


This response only addresses the extreme criticisms about saying they had a blog so it "must have been easy" or "it's cheating".

I don't think it was easy nor cheating. Both of the founders mentioned here are smart, capable, and engineered whatever advantage they had.

Another perspective on self-promotional articles (not an inherently bad thing) like these is they downplay or omit the influence another venture had on building their current one. The SubmitHub article has a specific question about 'what have been your biggest advantages' and it surprised me that "running a high traffic blog in my target industry with a large email list" wasn't mentioned there. That seems like a huge advantage.

This is a sentiment I've seen in many of the comments here and in the HN comments about your SubmitHub writeup.

It's not meant as a personal attack at you, Jason, or Distrokid's founder. You're all doing great things! It's about breaking down what were likely the major factors in success so that information can be applied in one's own ventures.


Totally agree with you. I know this is a nitpick, but in Jason's response to "What have been your biggest advantages", he actually does talk about running Indie Shuffle:

> As mentioned before, being "part of the problem" provided the best advantage possible. There were a few competitors trying to solve similar problems, but none of them were actually part of the equation. As such, they struggled to convince others to sign up, and didn't have a steady funnel of users to push their way.

Specifically, the knowledge he gained from running Indie Shuffle ("being part of the problem") and the "steady funnel of users" really helped him out.


On 2nd glance, you're right. It could have been more clear ("running my music website was a huge advantage because..."), but it was indirectly acknowledged. Thanks for calling it out.


Not everyone on indie hackers needed to build a community to succeed. In the area of music you need a community to succeed but in other areas like the AI based logo design, it was driven by the developer's work experience, and then he flipped that into a novel product, but he didn't start a a blogger he just found a novel application for machine learning to automate himself out of business as it were.


For an example outside of the music industry, StackOverflow was launched via the blogs of its two founders, Joel Spolsky and Jeff Atwood. I signed up right away because I was an avid reader of both of them. Without that jumpstart the site never would have built up the critical mass of questions and answers and subject matter experts that made it useful even at the beginning.


Hate to be this guy, but, by definition, sending emails cannot be "backbreaking", unless, of course, you develop serious posture problems.


Hyperbole is occasionally used in the English language.


Never.


This is the worst response to a message I've ever seen on HN


This is the response to the worst message I've ever seen on HN


You're literally Hitler


I regret starting this reddit-style oneup chain, my bad.


Literally never.


This is the best response to a message I've ever seen on HN


Where are you getting the 4 months & 1000 personalized emails from?


I recorded a podcast episode[1] with Jason where he talked a lot more in-depth about how he got SubmitHub off the ground. Here's what he said:

> To launch SubmitHub I hand-tailored more than 1000 emails to blogs. It took me about 4-5 months of doing it everyday to get through this list. It didn't just launch overnight, there weren't people banging at my door. I had to go and email these guys, and they didn't respond. So I would tweet them, I'd Facebook message them, I'd send SoundCloud messages, I'd send another email, then I'd try to find another contact... Those 250 blogs and labels didn't come overnight.

I've consistently found it to be the case that there's a lot more information lurking under the surface when you sit down and talk to someone about what they did. What you can surmise from reading a short post or interview really only scratches the surface... which is why it's easy to underestimate how difficult it was for people to launch their successful businesses.

A related anecdote: In Jason's podcast episode he talks about how someone read his Indie Hackers interview, thought "This idea is so easy!", and cloned the look and feel of his site. Of course their clone went nowhere. They didn't do any of the hustling, marketing, selling, etc, because all of that was behind-the-scenes stuff. It was invisible to them, so they failed to factor it in, and instead gave too much weight to the obvious parts.

[1] Shameless plug: The Indie Hackers podcast launches tomorrow morning at https://IndieHackers.com/podcast! It'll be three episodes to start, 1 hour each, and Jason's is #1.


Fandalism was bootstrapped too (and didn't much affect DistroKid's growth; at least not as much as I would have hoped).


> But oh wait he launched DistroKid on the back of the HUNDREDS OF THOUSANDS of registered musicians already using his other site - Fandalism - founded years before

I think it's valid to say that's how his product gained traction quickly, but I doubt that's why DistroKid is a successful company (guess, I'm not in the know).

Did I miss where that's against the rules of legitimacy? Or am I misunderstanding what you're implying?


I am not implying his built in audience was the reason for the success.

DistroKid is successful because pud is a badass entrepreneur and a passionate musician that understood the pain point and executed beautifully.

But at the same time you can't underestimate - or edit the story to not include - the value of early traction and a built in "free" source of eyeballs while you're tweaking product market fit over four years.

The price of Pud hitting #1 on HN w/a piece of start-up content marketing is a little sort out - for the rest of us in the comments section here - about what lessons to take away from the content, and what else to consider.


He is also a good marketer. Thats the key.


Its sometimes interpreted as cheating if you leverage any other thing you may have developed. Like an audience or previous product.


^ This is always amusing to me, because people who get really successful do it by compounding earlier wins over decades. I'm reminded of @pmarca's advice for startup founders who want to meet VCs:

> The best way to develop contacts with VCs, in my opinion, is to work at a venture-backed startup, kick butt, get promoted, and network the whole way.

> If you can't get hired by a venture-backed startup right now, work at a well-regarded large tech company that employs a lot of people like Google or Apple, gain experience, and then go to work at a venture-backed startup, kick butt, get promoted, and network the whole way.

> And if you can't get hired by a well-regarded large tech company, go get a bachelor's or master's degree at a major research university from which well-regarded large tech companies regularly recruit, then work at a well-regarded large tech company that employs a lot of people like Google or Apple, gain experience, and then go to work at a venture-backed startup, kick butt, get promoted, and network the whole way.

> I sound like I'm joking, but I'm completely serious -- this is the path taken by many venture-backed entrepreneurs I know.

http://pmarchive.com/guide_to_startups_part3.html


> ^ This is always amusing to me, because people who get really successful do it by compounding earlier wins over decades.

The weird thing is how someone who does this is lauded for it, but if it's a family rather than an individual that compounds earlier wins, suddenly it's 'wealth inequality' and terribly unfair.


Because it demotivates people.

The thing about capitalism, which I feel is not widely understood, it's that as a participant, you're meant to compete, but you're not meant to win. The need for money to sustain your lifestyle is what drives people to productive work; competition is what drives companies to optimize. But the moment someone wins - a person becomes rich, a company becomes a monopoly - they actually become a problem. At best, they're no longer working and optimizing. At worst, they can disrupt lives of a lot of people.

The promise of great wealth is phony by design.


You are supposed to win, but then adjust your goals, i.e. aim higher. This is a different thing.


No; in designing a robust system, you can't rely on what people are "supposed to do" out of their own initiative. As it is on the market, the moment you win and the external incentives pushing you to "aim higher" disappear, you become a wildcard - i.e. a problem.


> out of their own initiative

aiming higher is human nature. The system is based on this evolutionary fact (sometimes called 'greed', or self-interest), not on directing human behaviour.

The market is also based on mutual self-interest. There are no 'external incentives' other than access to higher/more expensive goals.


> There are no 'external incentives' other than access to higher/more expensive goals.

There are - living in a market economy requires nontrivial amount of work just to maintain what you have. Both individuals and companies have to work to balance out their operational (living) expenses; moreover, in case of companies, competition forces you to "aim higher" because not doing so will make your profit fall with time.

When you "win" in the market game, those incentives diminish or disappear completely.


> living in a market economy requires nontrivial amount of work just to maintain what you have

This is not external, but natural.

Humans need a constant supply of food/energy, food spoils etc. Tell me what kind of economy works otherwise.

I'm also talking more about life goals e.g. getting a better job, as opposed to working at all.

> competition forces you to "aim higher" because not doing so will make your profit fall with time.

This is dependant on a lot of things. There are plenty businesses that have survived being less than innovative. But I'd argue whether being competitive counts as "aiming higher"; competition doesn't always grow, sometimes there is constant competition, and innovation is part of business.


Thinking about it, the real problem is that the market is finite in size. In an infinite market, being a monopoly is impossible (although anticompetitive behaviour should still be discouraged). The problem comes when one fish gets too big for the pond.


I can't really imagine an "infinite" market. The biggest one I can imagine is an ultra-global one, where every living human anywhere can buy anything from anywhere and get it reasonably quickly. In such a market, monopolies would be global monopolies.

Actually, come to think of it, the market for digital services is quite "ultra-global" already.

Either way, whatever the market size is, monopolies in a given market is the the victory condition - the goal - for the monopolist, and a failure mode for the entire market. Hence how it works is that the idea of "the win" is dangled in front of you so that you work, but it's meant to be snatched away in the last moment, so that you never reach it.


Kantian categorical imperative: if everyone does the former, we all win. If everyone does the latter, we stagnate.


Maybe because most virtues make sense from an individual perspective rather than a group.


Not cheating -- no-one is making a moral claim here -- but to leave running Fandalism out of your description of your success feels like you're eliding crucial information. Like possessing a great marketing channel. And good for Pud for building it and using it!

I read these sort of success threads as basically case studies, and that marketing channel is probably relevant.


When you're self reporting how you think you accomplished something and you leave information out, it's not cheating, it's leaving information out. It doesn't mean it's intentional, but not being intentional doesn't make the left out information irrelevant.


I found DistroKid through here, not Fandalism. Actually, of all the sites I follow, I can't say I actually recall visiting that one on the regular (ex: versus Billboard Biz or Complete Music Update).


ABP - Always Be Parlaying.

I just made that up. Maybe it'll become a thing.


Why not tweak out the "Our secret:" part of article to include the whole "capture a community" thing?

Or, fuck that, since this article is awesome self-admitted wantrepreneur wrapped marketing I'm just personally curious - how important was fandalism to (1) launch and (2) tipping yourself into current hockey stick mode?


There are two DistroKid banners on Fandalism.com.

According to our tracking URL, the main one has been clicked 57,543 times, converting into 207 customers.


You didn't e-mail your list? I certainly imagine that would have made your life a little easier?


True. Every Fandalism user does get 1 email, from Fandalism, about DistroKid.


Did you collect stats from that as well?


Bam. Added edit to my comment. Look forward to more posts and would love to see more on the "how we grew" side.


Thanks for keeping me highly entertained with FC back in the day!


Cumulative Advantage!


He is a good marketer. Good marketers are always building audiences and brands. Turns out that the audience for a failed brand can support a new one. So what if he uses previous audiences? The question is: why aren't you? (unless you are happy to be an anonymous employee (which is totally ok)).


> But it's sort of disingenuous to play David Vs. Goliath here like his line about launching:

Maybe not, people misunderstand David vs. Goliath as David being the underdog when David was a significantly a lot more skilled and was an archer[1]. @Pud was better, executed better and is beating the competition.

[1] - https://www.ted.com/talks/malcolm_gladwell_the_unheard_story...


A young man with no military experience vs. a giant who's killed every soldier in the army who's attempted to fight him qualifies as an underdog story in my book. Also I'd assume archery skill doesn't really transfer very much to using a sling


I heard an interesting segment on NPR one or two years ago that delve into "David vs Goliath" and they made the point that we have the situation completely backwards.

Some scholars believe that Goliath suffered from a genetic disorder that may have left him rather "simple" in intellect. And slingers during that era were vicious killers capable of hitting birds in flight along with other examples of their prodigious skills and lethality.

The segment made the case that David vs Goliath was a stone-cold killer against a relative 'innocent,' it just happened to be David who had the distinct advantage.


David being classified an underdog doesn't necessarily mean he wasn't actually at an advantage compared to Goliath in some ways, just that it appeared that way. In fact if you read the biblical passage it seems to paint the picture that David was actually extremely skilled, and he claims to have killed bears and lions before while defending his sheep. It's the fact that it's a shepherd boy taking on a warrior giant that makes in an underdog story, not the skill with which David wielded his weapon.


I am reminded of Indiana Jones against the swords wielding fanatic. :)


Malcolm Gladwell's book David and Goliath talks about that example (among others). It's a pretty interesting read.


But he entirely bootstrapped fandalism too. So his argument is completely legit. Any one of those competitors could have built a fandalism for customer acquisition too. Anyway, his point is he did this stuff all as a one man shop with no funding. Nobody can dispute that. He is David.


Which is amazing but irrelevant - if the lesson is "build something obsessively for years until you have an audience to then, also over the course of 4 years, monetize on another project - assuming you get both of these things exactly right even though they're both unbelievably hard" there's no lesson.

I know this from experience. I helped build a site used by millions of people in a particular demographic, and some of my collaborators on that project have been building products to help that demographic in deeper ways. The problem? It's still hard to get that right! 2 years into that effort they're still just starting to find product market fit for the second products.

We hit the absolute lottery to somehow get 5 million users for the original product, and they'll be hitting the lottery again if they get profitable on the second product to that audience. It's hard. Having money helps.


Why do you presume these were two separate projects? He built fandalism, then realized from running it that his users wanted something like distrokid. It is one big company called Pud Inc.


Helps that DK is one of the very few services out there worth paying for. I can build most things myself.

The lesson here in the 'hey I did it' (and the implication) you can too! No you can't. You need some edge. Not you're smart, skills, or idea. That won't cut it. Way too many wanna be moguls out there already working on your idea. Bringing a built in audience, that's an edge. DK a great service however.


Nice phrase - "brings a data-gun to a knife fight" - that encapsulated the idea quite well!


I love articles like this it's inspiring...but it really doesn't offer any new insights other than the fact that

Every one man success story on HN was largely possible due to existing audience from another entity that took years to grow

It's pretty disheartening once you read the article and it boils down to this very requirement: You Need An Audience First.

I'd like more insights and details in how lone entrepreneurs of HN were able to grow their audience.

Because when Step 1 is have a big enough audience the rest of the steps are pretty much a no brainer. It's this step that most of us are struggling with.

I've spent years in seclusion writing a SaaS that ultimately resulted less than the annual wage of a North Korean factory manager and it was rough both physically and mentally but it all boiled down to the fact that VC funded startups were able to do huge PR and buy ads, write blog posts, increase Adwords bid prices etc while I was not.

So yeah, you can succeed with a one-man biz outfit but not without an audience. It makes sense why Linkedin is valuable, why Pinterest, Whatsapp, Oculus, Twitch gets bought out even without revenues.


I was able to bootstrap and launch a successful (5 figures MRR) SaaS by myself with no previous audience. Now we are much bigger (in terms of both employees and revenue).

How I did it was very simple. I found where my customers were and I lived and breathed on the industry forums they were on. While developing my beta I had a very strong sense of the problems people were facing, how my product solved those problems, and why my product would be superior to competitors. Plus I was able to establish myself as a reputable person within those circles of people on those forums.

So when my beta was ready I was able to get people to pay for the beta by offering them a lifetime discount and the ability to talk with me throughout the development process so that they had some impact in shaping the product. Just from those beta users I was ramen profitable and I was able to bring on an intern for a couple months to help me speed up some of the development.

By the time I publicly launched I had a core group of evangelists (the beta users), a strong reputation on industry forums, and an amazing product that solved big issues my customers faced and I reached 5 figure MRR within a month of launching.


Ah, so the secret to launching without previous audience is to build up an audience prior to launch.


A lot of people think along the lines of:

Step 1: Develop product

Step 2: Market product

But both need to be done in tandem with each other. While you are developing your product you should also be working on some way to either build an email list, establish rapport, get on people's radar, or at the very least validate a product fit and a basic marketing strategy.

For instance if you plan on advertising on Adwords you should be running tests to a dummy page (or email opt-in) before your product launches so that you know if that is a remotely viable option before sinking tons of priceless developer hours into a product with no proven way to market it.

Alternatively if you are going to advertise on forums you better spend a lot of time on those forums to make sure you are creating something those people care about and that those forums have enough visitors to move the needle.

Those are two examples - there are a million ways to market a product and all of the promising ideas should be experimented on and validated before your product is even finished.

Far too often solo cofounders are driven by a cool idea and then handwave (or worse look down on) the marketing aspect of it, and then after their product is finished they are left wondering why they spent huge amounts of time creating a product that nobody is signing up for. When I launched there was no uncertainty - I had already talked to tons of people in the industry, had lots of beta testers, and had properly validated the promotional channels I thought fit best with the product.


The book 'Traction' does a great job at building upon this great comment and giving good action steps for founders/makers to market their product


The "secret" is to build something useful, go where the people are, and make it easy to buy at a price they are willing to buy. Easy to say, hard to do for reasons.


Not sure if this is sarcasm, but it _does_ make sense--if you want to sell stuff, you need people to sell it to.


Maybe.

But if the creation of a small, passionate audience happens naturally during the creation of any sufficiently good product, is it that much of an obstacle?


I see this pattern in many successful products. As soon as you have a prototype, you should show it on forums and irc/chat. What I think most people are afraid of though is that your idea will either be "stolen" or that noone will be interested in your product ...


I've noticed a pattern as well where these successful entrepreneurs also don't worry about their idea being stolen. The idea should be such that anybody with ton of money/connections wouldn't be interested in.

And if a solo developer is trying to look for prospective customers, the customer should feel exited about just hearing the product. Otherwise there is no point in pursuing the idea. So there should be no worry about the customer not being interested.


Well done, I have done similar, although less successful and not SaaS.

But the parent did say:

> Every one man success story on HN was largely possible due to existing audience from another entity that took years to grow

And I think your experience (and mine) echoes that: the audience didn't happen to be "yours" but it did exist. The problem you have to solve in our case is how to turn a "colder" audience into a warmer one (as it would be if you "owned" the audience already).


um... wow! are you saying $0RR to $xMRR in 1 month after launching?


5-figures means 10s of thousands of dollars.


Honestly I don't feel that Fandalism (which is also a one-man biz, too, I spose) added that much to DistroKid's growth. It was mostly some press, some blogs, and a good product. But yes- having a built-in user base with Fandalism didn't hurt.

In hindsight, the other distributors are so clunky and expensive, the bar is super low. Like, if you wanted to compete with Google search it would be hard. But if you wanted to compete with something moderately terrible (with all due respect. i do respect my elder competitors and are thankful for the road they've paved) -- it's much easier to grow.


I launched with exactly one customer to build my first iteration, and grew from there. So the minimal viable audience is very minimal indeed.

How we grew: Impress the hell out of your customers. Treat them with empathy and humanity. They tell their friends in the same industry. Don't spend a single dollar on advertising, it's grossly inefficient.


Agreed - warm recommendations from people reaching out to their professional networks have by far the highest conversion for us.


> I've spent years in seclusion

Don't do that. Talk to potential customers. Hang out in the places they hang out (online / offline). Start marketing early. Get users early. Never work in seclusion.


There was insight beyond what you were specifically looking for: that he's created something that is very useful for customers and a great result for him, without turning to other money. It's a very viable path.

He's noted elsewhere that Fandalism wasn't as strong an aid as you might think. AND he had to build that audience himself too.

You can write blog posts and buy (small) ads and do some PR. Everyone started somewhere.


Of course you need an audience, people who want your product.

Do you expect to make a product that no one wants and somehow succeed?


Every one man success story on HN was largely possible due to existing audience from another entity that took years to grow

That's absolutely not the case.

Some of them are - including this one, by the sounds of it - but it's not universal.


To be fair, any company making a net profit is doing better than most VC-backed startups.


I get what you're saying, but a rock in the middle of the desert makes more money (or at least loses less) than most startups. The attempt to achieve greatness, even if they fail, that startups shoot for is noteworthy. At least it is to me. Taking on something risky and convincing someone(s) that your thing is worth a multi-million dollar investment is impressive.


To add to this, any business that targets $100MM or less a year in terminal velocity revenues will basically be uncontested from VC backed start ups. At the risk of generalizing, VCs these days don't have interest in businesses whose exit price is less than $5-$10B.


I agree with you.

But VCs are still investing in startups that do what DistroKid does. Those startups all take a percentage of royalties. So there's probably a sense of "if we release the new Jay-Z record, we'll make that $100MM revenue..."

DistroKid doesn't take any percentage. While we're leaving money on the table, the hacker in me (in the HN sense) kinda feels good that we're f*cking it up for everyone. At least I hope that's what we're doing. \m/


You're competing with what's possible rather than merely what exists today. DistroKid is better technology and a hint at the great future of the cloud. Congrats!


Is the number really that high? Can one really generalize all VCs? Are we strictly talking VCs and not other types of startup investors?


I'm not a VC, so take what I say with a grain of salt, but while the numbers look a bit high to me, they are probably in the ball park for SF VCs. I use to work with startups in the Ottawa Canada area about 15 years ago. Even at that time, they might give you $1-10 million in seed money, with %20 ownership, expecting a 20:1 exit. So they'll want $20-200 million back, which means your valuation has to be in the $100 million to $1 billion range. If you're not showing how to have $10's of millions in sales in the next 5 years, they have absolutely no interest in you. (Well, in those days, the main exit was a buyout from the big guys, so having an "industry disrupting" demo was the other thing to do and was much cheaper to build).

These days, if you are looking to build a "niche" business with $10 million a year in sales, I think you'll have absolutely no competition from VCs. If you're looking at $100 million a year in sales, I think you'll butt up against the "failed" VC funded companies. And larger than that, I suspect you'll be getting quite a lot of competition.

There are definitely angels who will fund smaller projects, but I think you still have to show an open ended potential. For example, I worked with one startup that had only $100K seed money and was looking for a quick $3-5 million exit. I've never actually seen any of these be successful, because $100K represents only a year of development costs, so companies usually reimplement good ideas in house.

Again, I'm not an expert in this area, but I have always thought that bootstrapping a sustainable business has been hugely undervalued for the last 20 years or so. Everybody is so focused on putting hundreds of millions in their pockets that they turn their noses up at mere millions. Lots of opportunity in this space.


Aren't there a decent number of companies out there that take 1-5 million in funding and then 'bootstrap it' from there? This seems like a fairly reasonable middleground where you get some cash to help you get over the initial hump and form a team but still have to focus on building a sustainable business asap.

Or are companies who seem to be on that trajectory avoided by seed investors even when the fundamentals look good?


Those numbers ring true for expectations of medium-to-large ($250 M+ is probably a good threshold) SF Bay Area VCs, at this moment in early 2017.

Source: small-to-medium non-SF VC.


I've gotten so tired of the VC model. It produces nothing. Just company after company of the same rehashed ideas (or worse still, good ideas) that are doomed to failure because they can't eat the world.

I wouldn't work at a VC unless I had literally no other options, and if I ever start a business, I'm doing it the way the author here did: honest work for honest wage. Maybe I won't get any Ferrari's but at least I'll have my dignity.


You aren't alone. The current VC model is destructive because of it's wasting capital at perhaps the most riskiest and inefficient way of producing alpha on other people's money-spraying & praying.

I don't know all the ins and outs but I have to echo the sentiment that it is better to make $1 for every $0 spent than $1 for every $2 spent. The latter makes zero economic sense. Any fucking idiot can spend money and make negative ROI. But the market dissonance is being fueled by low interest capital.

We are going to see history repeat again this year and when the water goes out, we'll see who's been swimming naked. They'll be back for the next bubble.

I'm convinced these bubbles are engineered who make money from the ride up and the ride down. Pump & dump on a national level but of course the men who play Gods need not play by the same rules they enforce on lowly mortals.

You will be able to buy a used Ferrari as they depreciate fairly quickly unless they are rare. They aren't built with longevity in mind and the parts do not depreciate in prices (expect to pay a Honda s2000 for clutch and F1 paddle shift pumps).


It's not really accurate to say that the VC game makes $0.50 on the dollar. If that was true, they wouldn't be in business. They make money by hitting grand slams--throwing out a BS number here--every 30 times at-bat. That's all they need. What's unfortunate about the VC model is that 29 out of those 30 at-bats is made on the backs of people who genuinely believe that they have a strong chance of being that grand slam. It's those startup founders who end up with the grey hair and circles under their eyes. As long as their are optimistic dreamers, there will be VC patrons.


Except it's not 1/30, it's more like 1/300 which still sounds generous to me.

As for founders with grey hair, I admire their passion, but I didn't get to where I am now by lighting both ends of my life candle on fire for a 1/300 chance of being better off. Fuck that noise. I'd rather have a life and be poor.


There is nothing in the VC model that says you have to run your business like an idiot if you take VC money... You can still run your business lean and keep your dignity. However, if you take VC money you need to realize that the VC wants liquidity and a ROI in ~7 years. So, if you want to run a small business that brings in ~$2M a year, don't go try raise a huge VC round at a high valuation. That probably would not work out well for either party.


> So, if you want to run a small business that brings in ~$2M a year, don't go try raise a huge VC round at a high valuation. That probably would not work out well for either party.

I'm curious, how would that negatively impact the founder?

If the founder still has a controlling interest what leverage do the VCs have to tell him that $2m/year in profit is not enough and to make him risk it for more?


If you can retain board control there isn't a ton of leverage unless the VC wants to hurt your business out of spite... in which case you picked the wrong partner.

What the VC can do is ignore you, in which case why did you take VC money when you intended to pass up the networking opportunities? What if you change your mind about raising another round of funding in three years but you already burned the bridge?

You should also remember that your cofounders will be on the board. Often the VC only needs to convince one of them you need a change in direction. A lifelong friend as cofounder will stab you in the back without a second thought if they smell a billion dollars at stake. Hence the advice to be careful about who you get into business with.


Or just don't take VC money in the first place. Then: Networking doesn't come with strings attached (or money), you have zero incentive to grow too fast, and there's no reason that you MUST make 2 billion dollars or shut down the business.

Seriously I've lost count of how many useful services have shut down because they couldn't eat the world, because just making really good money but not incredible ALL THE MONEY wasn't good enough. And for what? Some time in the public eye and a ton of money to blow on dumb shit like San Francisco offices that reduce productivity and a metric assload of hardware you didn't need in the first place.


It cuts both ways though. I am sure there are even more useful services out there that you never heard of because they didn't have enough money to market or distribute their product. The VC route and bootstrapping each have their advantages/disadvantages.


If you take money at a high valuation, the VC will only be happy if you try to grow the company significantly and exit at some multiple of that valuation. If the founder decides not to do that, then one of the largest shareholders in the company (the VC) will be angry. At a minimum I imagine the founder would be negatively impacted by losing some of his/her reputation in the investing space and having angst knowing that he/she took a VCs money and then ignored the VCs interests. At worst, there could be issues at the board level and loss of control of the company.


Really resonated with me, also being a micro-startup with a minimalist customer-centric philosophy, albeit in a completely different market.

Steps to reproduce:

1. Identify a poorly served professional segment where the business processes are not the primary task,

2. Automate their processes with vision and passion and empathy for the customer,

3. Make fans for life. Maybe also profit.

Minimum requirements to execute: at least one seed customer; wifi; coffee; time.

Rock on pud.


Indeed.

Rock on inopinatus


I very much like reading this type of stories where founders candidly talk about their business, how well they are doing, what did it take to get where they are (remove a good chunk of the usual BS, add some numbers). Indie Hackers is a great source (https://www.indiehackers.com) for this type of stories.

I have a feeling that this is the dream of a lot of people on HN. Having an easy to run business with a couple of people, without crazy competition, not needing to go sell it to VCs, and making a good amount of money while running it. Not judging, just observing.


Met @pud once while he was CEO of Blippy. Very nice guy and great to talk to. I also absolutely love his post: "Why Must You Laugh at My Back End" (http://pud.com/post/9582597828/why-must-you-laugh-at-my-back...). He's built lots of crazy impressive stuff by himself and DistroKid is no different.


Heh, I just wanted to checkout the DBaaS he mentiones, but it seems http://www.xeround.com/ is now a really shitty Alibaba "clone".

Edit: For reference, here's the old website: http://web.archive.org/web/20071019010255/http://xeround.com...


Yep, yep, my favorite part is using Dropbox to deploy code.


I've since discovered the joys of git. Dunno what I was thinking with Dropbox. Did the job I spose.


Couldn't you still use dropbox to keep eg server config in sync? you can put a git repo inside a dropbox folder anyway :)


Please correct me if I'm wrong - isn't this a terrible business decision? It took a small team a year to make a product that is apparently taking the market by storm. Their secret sauce is apparently not some special algorithm, but common-sense automation. If the barrier to entry is actually this low, and they advertise the fact widely enough, they will soon be joined by competitors trying to do the same. They may be ahead of the pack, but they lose their selling point and their advantage faster than they would have done otherwise.

Why give away your hand like this?


It's one thing to talk about your plan, it's another thing to actually execute it. We've seen bloated startup after bloated startup, I think running a lean (and I mean lean, 3 employees is nuts to me) ship like this is very hard to do and also must be very tough being the sole programmer.


It might be tough being the sole programmer, or it might be bliss.

I've been daydreaming about running a microISV recently - the ability to prioritise everything myself and crack on without justification appeals. Sure, I might be wrong, but that's OK.


The opportunity already came and went. Four years in means four years of catching up to do. If you come at it with more funding and a larger company structure you might be able to kill DistroKid by burning through a lot of cash, but you won't achieve the same "bottom feeder profitability", and your company might go under afterwards.

Basically, the only productive way to challenge this company is to also go in solo and be tough as nails, and there are almost certainly less contentious niches you can take on if you want to do that.


With more funding, long-term profitability is not a concern


I don't understand this. Do you mean "short-term profitability is not a concern"?

My take was that DistroKid is already taking bare minimum profits, so even if someone enters the market with cash to burn, they're only fighting for a piece of an already small pie, and there's little guarantee that will recoup their burned cash.


The argument against it from GP (if I read correctly) was it would not be profitable to do that.

My point was that profitability isn't a concern of highly-funded startups. Growth is. So if they can throw enough money at it and pull those customers away, they've met the definition of 'success' as seemingly applied by many VCs - fast growth and on-paper potential. The cost and loss is much less relevant.

I shouldn't have said "long-term" as it added confusion. I meant 'profitability' in general.

I don't necessarily think that the market is large enough to attract those well-funded startups, but we're talking hypotheticals at this point anyway...


"The Truth, The Whole Truth, and Nothing But The Truth"

"The untold story of the Microsoft antitrust case and what it means for the future of Bill Gates and his company."

https://www.wired.com/2000/11/microsoft-7/

> The only thing that surprised Microsoft about Netscape's strategy was the brazenness with which the upstarts shouted it to the world. Nathan Myhrvold told me, "There's a good analogy to bicycle racing. In bicycle racing, you don't want to be first until the end. What you want to do is draft the guy in front of you. And then, in the last minute, you dart out. The middleware gambit is about drafting the leader." Yet here was Andreessen publicly proclaiming in the summer of 1995 that Netscape's plan was to reduce Windows to "a poorly debugged set of device drivers." "They didn't save it up," Myhrvold said. "They fucking pulled up alongside us and said, 'Hey, sorry, that guy's already history.'"


Distrokid has been out for a couple of years now and it's easy to see that it's pretty damn popular. Sure someone can get a team together and create the next Distrokid (I'm sure it's happening), but they'll be behind a couple of years. By that time, @Pud will have made his millions and onto the next...


Yup, they have the userbase already. Also, they disrupted the incumbents by being an order of magnitude cheaper (or cheaper than that, based on how many songs you're selling) than their competitors, which helped their userbase grow.

It's hard to imagine a new competitor being able to compete with DistroKid on price at all, considering how cheap they are. An order of magnitude cheaper is probably out of the question. And these numbers (for the larger clients) are peanuts on DistroKid anyways, so even if you could compete on price it might not matter for much of the market.


I agree for price, you can't really beat Distrokid but disruption can come in other forms. Other competitors may think of better features, more connections, or simply better marketing. I don't know.

Will this likely happen? Nah.


Good question. However, I'd argue the entry to barrier is actually very high. It boils down to their audience base they've already built up over the years. This commands a big market premium because it cannot be replicated with engineering.

So they aren't showing any of their hand (how to build audiences for customer development) but able to convey the image that they are somehow david vs goliath by sharing their stories like this.


> It took a small team a year to make a product that is apparently taking the market by storm. Their secret sauce is apparently not some special algorithm, but common-sense automation

They're taking a part of the market by storm.. specifically Indies with a "do it yourself" mentality.

Speaking as someone who works for a music distributor who has one of the "Big Three" Record labels as a client (Warner, Sony, UMG), I can say that in general we haven't been a great fit for really small time acts (though we're getting better). But the breadth of Distrokid's offerings wouldn't be able to cover that of a major label (and although they claim they could, there's no proof that they could handle that kind of scale either).

For instance, the base Distrokid plan doesn't even allow you to specify a release date (presumably it just releases it right away). Major labels often run promotions in different countries (necessitating different release dates per country). There's also pre-order dates, some of which are tied to instant-grat tracks for certain release dates in certain countries. The fine grained control necessary for a big act is missing from most DIY offerings.

Distrokid is serving a different market than the incumbents, and that's cool. I'd definitely recommend it over the other flat rate model distributors. But there's a reason Distrokid isn't delivering the next <insert big act here>'s album.


I wouldn't want to compete with this. It's a whole lot of schlep [1], and even if you match their features, you'd have to charge less than $19.99 per year. I don't think there's any more room for innovation or lower prices.

[1] http://paulgraham.com/schlep.html


> Perfection is achieved, not when there is nothing more to add, but when there is nothing left to take away. - Antoine de Saint-Exupery

Not saying that DistroKid is perfect, but sometimes all it takes to be the best in any given market is to be exactly what's needed and nothing more. It may seem like giving one's hand away, but what could there be to add? There's certainly not much fat to trim. How would someone else compete?


Ideas are everywhere, it is the implementation that is the key.


So artist pay $20 per year, and to quote the article "100,000+ artists", he's looking at $2,000,000 gross per year? Three employees. Amazing


Plans start at $20 per year. $35 gets you more advanced features (ability to set release dates, customize pricing, and more).

We also have optional paid services, such as cover song licensing ($12/year per cover song), YouTube ContentID ($4.95 per single) and a few others.


How does $12 cover the license of a cover song? Aren't those fees much higher/based on number of sales?


DistroKid is charging $12/year to handle the covered song fees for you - they automatically withdraw the required amount from your royalties and send it to the original author.

[0]: https://distrokid.desk.com/customer/portal/articles/2131024


Minor complaint / remark: is there complete list of all stores and services that DistroKid supports (directly, or better even indirectly) somewhere?

All I was able to find were murky statements about 'dozen mayor services' and "150+ others" [0]. That page links "MediaNet Customers" page [0a] that displays 24 logos and links [0b] that displays list of 286 "MediaNet Content Partners". Is that it?

I was particularly interested about Bandcamp and all I've found that DistroKid mentions Bandcamp [1] and Bandcamp mentions DistrokKid [2] but no proof they really talk to each other.

[0] https://distrokid.desk.com/customer/portal/articles/1276117-... [0a] http://www.mndigital.com/about-us/customers.html [0b] http://www.mndigital.com/about-us/content-partners.html [1] https://distrokid.desk.com/customer/portal/articles/1601235 [2] https://bandcamp.com/help/selling


Ah yes! The CFML guy - @pud (that's how I always remember him from that one blog post he wrote)!

This is really cool, and love hearing about stuff like this. I love the idea of running a small team and scaling a product which doesn't have a huge overhead to make something which beats out the current market by just doing a few things better. Hats off to you man. Great stuff!


This pub, this very pub we're sitting in. I built it, with my own hands! But do they call me the Pubmaker?

Naa!

See the wall over there, that protects our town? I built it, with my own hands! But do they call me the Wallmaker?

And the bridge, you know, that crosses our river, I built it, with my own hands! But do they call me the Bridgemaker?

But I tell ya, man! YOU F*CK ONE GOAT!

- the CFML guy


Well, to be fair, you f*cked that one goat multiple times. ;)


I was wondering where that was going but I think you pulled it out at the end there.


This is the reply I was looking for.

You really got me there 'the CFML guy'.


I love what you're doing! But have to observe disparity between:

> beating VC-backed startups

and

> DistroKid intentionally has a small team and no investors. We’re here to make the world a better place for musicians — not to make billions from them. We’d make ludicrously more money if we charged what the other distributors do.

It sounds like you're coexisting quite nicely with different objectives.


> beating VC-backed startups

Just to nitpick, CD Baby isn't really a VC-backed startup either, right? I've been under the impression Derek bootstrapped it to revenue quickly similarly.


He sold it and they pivoted since. It's hardly the same company at all anymore (Derek's CD Baby was an online record store. A webshop for CDs).


@Pud

There are many people who want to know about your tech stack. I think the thinking goes: with one developer, it must be an insanely productive stack.

http://pud.com/post/9582597828/why-must-you-laugh-at-my-back...

is old, do you still use the same set up?


As someone who works in an AWS / CFML shop, I would also like to know this


@Pud, would be great to hear your take on what it has been like dealing with the record companies. According to YC:

> Because the alumni network is so large and tightly knit, investors or companies who try to maltreat a YC-funded startup can usually be made to stop. [--> footnote] Except for the record labels, which are effectively a rogue state with nuclear weapons. There is nothing we or anyone else can do to protect you from them, except warn you not to start startups that touch label music. [0]

Your service does various tasks such as cover song licensing which are record label-facing. What has that been like?

And, congratulations on getting engaged.

[0]: http://www.ycombinator.com/atyc/#n3


Eh, I go with CDBaby because the math doesn't work out for me as a small artist. I'd rather pay $50 and just have my album up forever. 9% doesn't matter much to me because I'm making peanuts anyway. Maybe if I get bigger I'll consider switching over.

It would be really slick if DistroKid had some kind of conversion process for CDBaby / Tunecore customers.


The entire thing is built on the ecosystems created by companies that were all venture backed at one point, though.


Well we're screwed, then, because it's turtles all the way down: https://en.wikipedia.org/wiki/Intel#Origins


VC cash is great for business that want to be infrastructure for the world. VC cash is a tool, to build a certain kind of large business. You just have to recognize that.


I didn't get the feeling that the point was he didn't benefit (directly or indirectly) from VC funding at all. Only that he was able to remain competitive with companies that were VC funded.


There is a big difference between having direct competitors that are VC backed versus using infrastructure that is VC backed.


I think the parent's point is that he's built his company on a VC-backed foundation. Not saying it's valid (I have no clue about this business or how it's built), but it seems he's trying to say this is a bit like Space-X claiming to be beating NASA at space travel.


PUD has a well deserved reputation among solo entrepreneurs. I think Distro Kid is his fifth startup.

All but one of his companies (the one unsuccessful one) were powered by CFML.


CFML is not a "hip" language, so I don't care /s

This a lesson for all of us developers (myself especially!) who end up fussing over what kind of language/framework/server to use before writing a single line of code.

Props to @pud.


Yes of course, choice of tech is mostly irrelevant as long as it works, and "cool" is an absurd criteria.

But you have to like it, esp. when you're a solo founder/programmer and it's what you do day in and day out.


CFML is and always was an inferior tech. Worse than PHP, together with ASP. But kudos to the Guy that makes itvwork the way he need.


In the time you've taken to miss the point, Kaplan's probably launched another business.


And now he is launching the third one. As a software developer myself, what is the point you got that I missed? CFML is awesome?


I am a bit of a language nerd myself so I'm inclined to agree with you from a technical perspective.

I just want to hammer in that last point you made about making it work the way he needs. I think this is often the most overlooked and undervalued thing in tech. Has been for a long time.

It's quite often people look at something like this or walk into a project at a new company with existing code and realize that sometimes the important thing is that it works well for what it's intended goal is not that it's written with the newest JavaScript Library or the newest implementation of Django (great stuff there btw not knocking Django).

Sometimes Perl is fine thank you. Still the best string processing language in the business In my not so humble opinion :)

I remember when CGI was all the rage and then poof everyone now thinks that it's gross. Anyone remember making money on custom joomla templates and plugins?

Edited for clarity


yes, even the worst of technologies can be used to produce value, if you pour enough energy into them.


I started a one-man biz that's beating VC-backed startups

Click-bait title in my opinion. It's not because you don't need VCs to satisfy your customers that VCs are bad. They are good when you need them to satisfy your customers. Now distrokid-competitors' problem is not that they have VCs money, it's that they "hired" bad ones. Now, I grant you, good VCs for startups, those who gonna understand whats good and not for your customers may be rare... But look at google and facebook, they did right by taking VCs' money don't you think.


One could very well argue that this success would not have been possible with VC money. They would never have this business model of $20/artist/year because VCs want loads of profits. And they would have >100 employees instead of three, because, well, aggressive scaling.

It might boil down to your definition of success I guess. Having taken VC money might have maximized their profit but at the cost of their top USP (very affordable service). Also they probably would not be loved as much by their customers and not doing as much of a good deed to the artists as they are now.


https://distrokid.com/ is a near-perfect landing page for this kind of service. Well done!

Anyone know if there is an equivalent for independent films?


Amazon Video Direct is a worthwhile enterprise to explore. It takes effort. Quite a bit of legwork to fill out forms. Also, the content MUST have a subtitle file...but those things considered, it's highly competitive and gets things on Amazon Prime and pays out real $USD.



Fandalism launch in January 2012. Distrokid in mid-2013. Considering it took @Pud near 4 years to reach these numbers with DistroKid, the year and a half of building a user base with Fandalism is not that big a deal.

This basically would be the same as all the freemium companies out there that offer everything they do for free for a while till they build their product enough to start charging for it later.

My favorite line is in the original DistroKid TC article where @Pud says the goal wasn't to make money with this but to get more people to the social network.


I've been watching DistroKid since it spawned out of Fandalism. Awesome to see how successful it's become.

Just curious, what % of artists gross over $19.99 a year?


This is pretty darn cool and the first I've heard of it. I thought there'd be a cheaper way to do what your competitors have and here you've done it. What your success also confirms is that reducing your workload with bots and cronjobs are one of the keys to succeeding with a small team.

How do you handle support?

How are you able to find infringing music?


> How do you handle support?

2 full-time support reps (who are amazing), so we're not really one-man, anymore.

> How are you able to find infringing music?

Obviously I can't get into details (lest the baddies figure out how to get in). But we have about 40 different checks to make sure the artist is who they say they are (or manually verify if the check fails) and other things.


I think it's great that you have found great success as a pipeline service.

I also think it may be disingenuous to suggest this can or should work for every type of business. While I happen to agree that services that simply do a pipeline, aggregation, or intermediary service are A) not something I feel VCs should usually spend their money on in great quantities if at all and B) are often the most ripe for disruption, I disagree that finding scaled efficiency for all types of businessses in this manner is possible.

I do however think it's wonderful you are promoting the more traditional idea of a business with this product at least which is the more canonical bootstrapping or self funding/ side job till you make it/ type thing that doesn't need tons of employees to be a good value for those that are a part of that business.

That's just my opinions though I'm glad it's been a success


Having quickly looked at the audience for DistroKid, then see what happens with artists by major labels, no wonder the product is sought after by upcoming artists. [0] Bandcamp is another company that uses technology to squeeze out the middle man. [1] The first page is really well designed. You can easily see what the product does, "let artists make music, then distribute it for a fixed cost".

Reference

[0] "52 Ways to Screw an Artist, by Warner Bros. Records" https://news.ycombinator.com/item?id=13648245

[1] "Is Bandcamp the Holy Grail of Online Record Stores?" https://news.ycombinator.com/item?id=12324350


This is another example that it's a misconception that you have to have a big team and lots of cash to get to the next level

Stories like this hearken to people like Mark Zuckerberg and Instagram's Kevin Systrom and Mike Krieger.


Mark Zuckerberg had "lots of cash" compared to most folks (his father, a wealthy dentist, and his father's friends bankrolled Mark's business). He's not the best example to use when it comes to "bootstrapping."


Interesting related statistics:

"The blockbuster effect has been even more striking on the digital platforms that were supposed to demonstrate the benefits of the long tail. On iTunes or Amazon, the marginal cost of “stocking” another item is essentially zero, so supply has grown. But the rewards of this model have become increasingly skewed towards the hits. Anita Elberse, of the Harvard Business School, working with data from Nielsen, notes that in 2007, 91% of the 3.9m different music tracks sold in America notched up fewer than 100 sales, and 24% only one each. Just 36 best-selling tracks accounted for 7% of all sales. By last year the tail had become yet longer but even thinner: of 8.7m different tracks that sold at least one copy, 96% sold fewer than 100 copies and 40%—3.5m songs—were purchased just once. And that does not include the many songs on offer that have never sold a single copy. Spotify said in 2013 that of its 20m-strong song catalogue at the time, 80% had been played—in other words, the remaining 4m songs had generated no interest at all."

http://www.economist.com/news/special-report/21716467-techno...


As a part-time musician, I absolutely love DistroKid.


Thanks!


Hey, does (or can) Distrokid offer an API? I've been dying to find a distributor that would let me programmatically submit/replace releases. It would only be used for my label, no bulk stuff, reasonable per-call fees or adjusting balances on a pre-paid account would be fine. Sale statistics API - even better.


> By contrast, our competitors largely have millions in funding ... And they’re owned by venture capitalists and/or private equity firms who are banking on a large exit.

Disincentives of the startup model at work!

These days it feels like you if your startup doesn't have polarized outcomes you're "doing it wrong".


Hey Pud, since you are answering comments here, I have a question: what failures have you had?


This is very motivating. I am also running a site in the indie music industry as a one-man biz. (http://www.synthshare.com) It has been a challenge to build up the community, I'm sure leveraging Fandalism helped you kickstart your growth. Do you have any advice about user acquisition? Recently I've been getting feedback from music production communities which has all been positive and converted to users; however, it sometimes feels like the site isn't growing as fast as it should be considering all the positive feedback I've received.


I see the comments where you started running with a sea of clients who trusted you to give them a fair shake. I see so many of the others where the music makers suck hind tit. Looking at your competitors, you have a clear advantage in your lean and efficient structure. In time your method will spread and and drive many others out. The problem I see is labels tied to the big companies may have restrictive agreements that exclude you. Couple this with the wish of many people to feel they are complete when they have bought into spotify/Apple/etc.

In any event, I wish you well, live long and prosper


I am surprised that the development time of 1 year was considered a long time by him compared with his 1-2 month initial estimate.

If only projects I / we had been working on for 2 years became so successful!


Is this a digital-only endeavor? I don't see any mention of physical product services such as pressing CDs & DVDs, printing covers/liner notes/etc. And what about marketing materials like posters, flyers and so on. There's still a market for that stuff, and it could be another revenue stream. Merch is also pretty big for some artists, you could explore getting t-shirts and hats and stuff printed. (just an idea, i've no idea if it's profitable, or more trouble than it might be worth)


This is amazing. How does one person build so many successful startups while everyone else struggles to just do one?

Maybe there was some momentum from his previous successes? If you're already a famous startup founder, then I guess it's easier to get users for your next startups. I've heard that this was the case for Slack, but I don't know how much truth there is in that.


Great to see this. Stoked to hear that it is the same guy who started Fandalism. I used to use Fandalism years ago to post my clips but just kind of stopped. I never got any emails or communication from them and just assumed that they faded away. Spurred by this article, I checked, and am amazed that it is still going strong. Must remember to get back to posting on there again.


Previous connections aside this is still very inspiring. Scaling this well with only three employees (and resisting the urge to add more people just to be bigger) is an amazing feat. I'd be very interested in more details. How focused were you on being profitable and reinvesting profits vs. taking venture money? My guess is profit was one of the key metrics?


3 employees, wow! Sounds like it would have taken 300 employees twenty years ago. Reads like an advertisement for basic income.


I wonder what his contingency plans look like. That's a bus factor of 1 there with just 1 programmer.


I'd be interested how you arrived at this particular pricing. $20 per year seems crazy low to me, have you experimented at all with raising the price. Intuitively, I'd think that the next cut-off point is at $49, but that in the range below that the demand is fairly inelastic.


...and please count me in among the users for the past 4 years who have seen the value, enjoyed the service, had a lot of fun playing with metrics, and feel inspired to continue my 100% rights ownership in music pursuit for 2017 and beyond.

Great product. Great support. Great example.


Does DistroKid track the usage/consumption details of the songs for artists to look at or analyze? If so, what is the probability that they will expand to also have a recommendation system for indie music based on these aggregated stats?


Yes we provide lots of stats to artists. No, we're not planning on a recommendation system (plenty of online radio thingy thangies that are already great!)


It's awesome to hear about a swarm of bots replacing millions of dollars worth of labor overhead. Are you worried about the maintenance and QA at all? Is there anything about the problem that lends itself especially to automation?


Would have been so much better if it was titled "YC startups"


>thought it would take 1–2 months to build DistroKid. It’s just plugging into some APIs and moving files around, right? I underestimated. It took a year.

Wow. I would like to understand more about this.


Any technical tasks estimation is usually 3x of the intial. So let's just put 3-6 months there.

Then imagine talking to lots of 'stores', contracts, legal stuff and so on.

Just reading about what this is, I think even one year is quick.


Great! I would like to tap your brains. But now you have captured/conquered the music world wouldn't it be wise to disrupt movies/films distributions?


A very well written and inspiring post. Congratulations and good luck. Do you mind shedding some light on the tech stack you used for distrokid?


It's Craigslist for music. Low overhead. This could do to the music industry what Craiglist did to newspaper classified advertising.


The music industry still produces/mixes/masters music.


No, they don't. That's all outsourced. The music industry "labels" are mostly marketing and funding operations. All the steps of music production are available to anybody.


Right, which is part of the industry. Who exactly is paying for all of that expensive studio time? The broke artist?


I can see how they would dominate music meta-distribution but it does not seem likely that they would get into anything further.


What's interesting to me is how he focused on doing small things and building up on them.


How did you get the music industry companies to agree for you to directly upload files to them?


Good, there are better backers out there who will buy you out whole and let you keep your soul.


I probably missed it in the article, but how does Distrokid make money?


"artist pay $20 per year" ... "100,000+ artists"


FC, oh boy, happy days :)


Many, many wasted hours.


You have dozens of bots, written by one dude ?

Really interested to know about your stack and your deployment methodology. Did you build a bit/async infrastructure because of scale issues...Or is it inherently a mental model that is easy to grok easily?


May not be up to date, but past info on his stack:

http://pud.com/post/9582597828/why-must-you-laugh-at-my-back...


Is there something like this but for books and poetry books??


So there are a few parts to this. Raw distribution you've got stuff like Draft2Digital which can help you distribute to Amazon/B&N/Kobo/etc. Limited value outside B&N requires you to be in the US so they can help you get around that if you are non-US based. But they also give some tools to make things like putting links to the right store in the back of your book for each platform.

However doing any licensing stuff if say you want to pay to put quotes from book snippets/lines from poetry/etc into your book the way distrokid lets you license a cover, nothing that I know of handles that space. I don't know if there's enough of a market for it because it isn't nearly as good a business decision for a writer as doing a cover can be for a musician, so smart writers avoid it.


Createspace? It's an Amazon company but it gets you into lots of distribution channels, worldwide. Zero setup or recurring fee, but a percentage of royalties.


Great work! I'd like to see more startups like this.


I use and love Distrokid - thanks for the product!


*one-person ftfy


a distrokid API would be pretty useful.


Congrats Pud.


nice


This is fantastic success story! Thank you for sharing. A schoolbook example of disruption howto.


Hey, nothing wrong with self-promotion - it can take you places! Look at POTUS.


OK.......self-promote your ventures ):




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

Search: