YC Analysis: "37signals/DHH style" companies? | Hacker News Reader

YC Analysis: "37signals/DHH style" companies?

49 pointsdavidw91 comments

Everyone has probably read about or seen DHH's talk by now, where he advocates attempting to create small/medium sized, sustainable companies with a simple, direct business model (sell something directly to your customers).

A lot of his examples are from 'real world' products (Italian restaurants, say), that clearly have very different economics from on line businesses. The marginal cost of another bottle of Bardolino is not indifferent, even though the markup is also going to be good. The marginal cost of another basecamp customer is very small. Also, his Craigslist example was not particularly compelling for me. Those guys dominated their market, but more because people flock to the site because everyone else does (network effects). Their choice to keep things small/real seems like a very conscious one, but not necessarily something they could get away with if, say, they were kijiji and eBay were the owner of Craigslist.

So, my question: what other web or software companies make a living the 37signals way? What are they key characteristics of their markets? Obviously they can't be in a winner take all market dominated by network effects.

View on HN

Also keep in mind that DHH suggested building a product for a business, not consumers. Consumers are used to free, but businesses are more than willing to pay. In fact, a price tag is a way for businesses to justify their corporate use of the product.

I'm constantly surprised when I see companies paying for a service that I get for free (Any premium chat/voice Vs. Gtalk). The thing is, businesses don't like to spend time looking into the best and most cost effective way to solve a problem - they like to throw money at it and make it go away. DHH suggests being that person who makes it go away.

There are LOADS of companies like this. I run a few small ones myself that generate good revenue. Software-related is our www.sourceguardian.com site. Like most things we solved a problem for our own needs and then turned it into a product. We have a healthy income from this which isn't far from being 'passive income'. My suggestion would be to look at a problem that hasn't been solved, or hasn't been solved well. We pick niche markets for what we do and that works well for us. I know of many companies like this.

The Craigslist comparison comes from:

1) The fact that in SF and NYC, they charge directly per listing in certain categories.

2) The fact that they didn't take VC funding.

3) The fact that they stayed small.

4) The fact that they didn't sell the company.

All of these would have been extremely easy to stray from, but instead they stuck to the core philosophy. What DHH was talking about was just that: a philosophy. Any company can stick by that philosophy, regardless of how it attracts customers or the specifics behind their product or service.

Asking the question "what other web or software companies make a living the 37signals way?" is kind of a weird question. The 37signals way is actually the normal way to run a business. It just seems weird because the scale of the web has thus far screwed up everyone's minds to the point where everybody forgets the past.

The problem with that line of reasoning, in my opinion, is that "normal" businesses sell "normal" products. As a simple example, how much would you pay for a clone of reddit? Not much. The value of the site is entirely in its large community. Same thing for eBay, for the most part. Also, normal products always cost something. There are a ton of information goods that are free: Linux, Apache, all kinds of languages, compilers, editors, browsers - everything you need to run a business, almost, can be had for free these days.

In other words, things do work differently on line, in some cases. Understanding that, and understanding what happens why is crucial to being able to do a 'DHH style' company that doesn't get crushed by a big player, or undercut by a bit of open source software.

I highly recommend the book 'Information Rules': http://www.amazon.com/dp/087584863X?tag=dedasys-20

If 37s is the normal way to run a business, then no one should ever disparage marketing, open source contributions, or consulting as methods to start or bootstrap a business. 37s has ridiculously low customer acquisition costs (compared to competitors) because they

1) have 85K subscribers to their SvN blog

2) wrote a book that sold 75K copies

3) developed one of the most popular open source projects in the world

4) speak at conferences all the time

If you're just writing code, you'd better hope to get bought, because you'll never catch up to 37s as a business.

Maybe existing DHH style companies are a good area to look for inspiration to build free alternatives with some premiums added afterwards... Like flickr added pro accounts for $25/year. Read Wired's "Why $0 is the future" http://www.wired.com/techbiz/it/magazine/16-03/ff_free

Building less of an app also means it will be easier for others to come later and build what you made and give it away. Look at how it took 2 people at Google a few days to build HuddleChat which some said was too similar to 37signals' campfire.

this is true - but is huddle chat negatively impacting Campfire? Campfire customers are not running for the doors now that some other company has built a free alternative. Building less of an app that serves a real need and provides an elegant solution can also mean that customers love your product and stick with you. Alternatives will always exist, competition is a natural part of business, giving your product away for free is not the only way to deal with competitors.

Of course it isn't negatively affecting Camfire; they had it removed from the internet.

How many other chat/messaging competitors does campfire have?

I can guarantee you huddlechat was not the only one, so now that its gone campfire can go back to enjoying their monopoly of the chat/messaging market.

Campfire is a chat app. I don't think 37Signals sees it as a bulwark against competitors. If HuddleChat integrated with Basecamp and Highrise, maybe they'd take notice.

This issue obviously cuts both ways. 2 people at a tiny consulting shop can probably whip up most of what VC-funded 2.0 companies build too. And those consulting companies don't have boards of directors that will flip out and fire the management team when that happens.

Take a look at alot of the popular Mac software companies out there. Many of them make quite a living, like Panic, and Delicious Monster.

One characteristic is that they all provide enterprise software. The end-consumer market is very difficult since internet-users are used to get everything for free.

The term "enterprise software" usually refers to software products sold to very large companies, or at least "small and medium businesses" which have 200-500 employees. David's examples were geared more toward the "Fortune 5,000,000" -- the much larger market of small businesses with a handful of employees.

The economics of the "Fortune 5,000,000" are very different than true enterprise software, as it avoids employing a direct sales force. Obviously 37signals wouldn't turn away a team from General Motors who wanted to use Basecamp, but they don't actively market their products to large companies.

You're right about the difficulty of charging for online products in the consumer market. DHH said that they were having trouble getting customers for their personal organization product, Backpack, even at $5 a month. After they repositioned Backpack to be used by small businesses, they were able to increase the price and sales still grew tremendously.

You are right. I used the term "enterprise software" only as a contrast to "consumer software". The "Fortune 5,000,000" is probably the best target market for a startup.

Why? Targeting large enterprises (what the company I work for does) is also an extremely difficult market, but for other reasons than the consumer market. Selling to large enterprises requires professionel and experienced sales people with very good industrie contacts. You also might run into the "small company problem", i.e. you don't sell because you look too small and you don't have enough customer references. Not much fun!

Reaching a bunch of relatively small, and perhaps not well-connected companies does, of course, present some obstacles of its own.

We sell to large enterprises almost exclusively. It's "different" than selling to consumers, but not "harder". Longer sales cycle, higher average deal size.

The end-consumer market is very difficult since internet-users are used to get everything for free.

...except for Mac users, who support a healthy little industry made up of one- and two-person software companies.

In other words, consumers don't spend money for software, except for the ones who do.

So - for the analysis part of the question: why? What makes them different?

I would say what makes them different is the thing that makes them different enough to pay premium for a laptop/computer in the first place: These people place a value on quality.

Add to that that I'd venture 80% of the windows users have never "bought" software (except games). Windows is either pirated or comes with the computer... office is acquired from work, some other programs are freeware/never paid for shareware.

It's also a different culture. No one could possibly claim that vim or emacs aren't "quality" software. UNIX/Linux users also place value on quality, but the form of value is just different--historically the value you pay for something like vim or emacs is you send patches, you add syntax support for your favorite new language, you file coherent bug reports, etc. It's not in dollars, but it has a price. So, there's this somewhat smug notion that Mac users are arbiters of quality and taste and the only honest computer users in existence...but there are different means of valuing software than money.

And, saying that Windows users are universally pirates is simply disingenuous. Paint Shop Pro made millions as a small software development shop, and then sold to Corel for millions and became part of a big development shop. This wouldn't have been possible if Windows users never paid for software from small shops. In fact, I suspect one could find more Windows-based small development shops making a good living than similar Mac-based shops just based on the size of the respective markets. (I'd never want to do it...but it's a viable model.)

I'm just saying, "don't be a smug Mac weenie". Many people are making a good living building and selling software on many different platforms.

The OP asked a question, and I simply extrapolated a response from my observations. I don't see how that makes me a "smug Mac weenie", I've got three boxes running, and only one is a mac.

To suggest that there isn't rampant pirating in the windows community by holding up Paint Shop Pro (generally a b2b piece of software) is ignoring the obvious, though.

What makes you smug is that you stated that Mac users were different in that they appreciate quality, and thus that people who are not Mac users do not appreciate quality. And, you stated that 80% of Windows systems are running entirely pirated software, which is, frankly, a ridiculous claim.

Paint Shop Pro isn't b2b. It's the epitome of consumer software. Businesses buy Photoshop.

I'm not arguing piracy doesn't exist among Windows users, I'm arguing that you're being smug about being an exalted and holy Mac user which makes you uniquely capable of judging quality and uniquely honest enough to pay for the software you use--unlike the riff raff using other operating systems. Sorry you don't like the term...but that's smug.

I'll admit to being smug and thinking, like you, that Windows users are idiots. But, I take it one further and think that Mac users are idiots, too (and they paid too damned much for their computer to boot). Likewise for most Linux users. Users, in general, are idiots.

>What makes you smug is that you stated that Mac users were different in that they appreciate quality, and thus that people who are not Mac users do not appreciate quality.

That's not always a good thing, and shouldn't be construed as such. For example, there are many great free products out there that won't appeal to some mac users at all because they equate price with quality. It's the same type of mentality that buys a BMW when a chevy will suffice. Sometimes the "buy quality" mantra is accurate, sometimes not. It is an aspect of that particular set of users though.

>And, you stated that 80% of Windows systems are running entirely pirated software, which is, frankly, a ridiculous claim.

No, I said 80% of windows users have never bought software that isn't a game. Given my experience of close to 15 years in various IT roles, I think that's not that ridiculous.

I think you're assuming you are talking to a particular type of individual and responding in kind, but rest assured, I'm not that person.

"smug Mac weenie" - take that, Lisp!

That's easy. No VC will fund a Mac app. It's just Darwinism.

A large number of Mac users are also professional designers or music producers and not part of that typical Internet user. So Mac users may pay for Mac software, but not for web-services.

My point is: the consumer market offer less niche markets that allows you to build a sustainable business compared to the "Fortune 5,000,000". I know this is not good news, because building software for enterprises (no matter how big they are) is less fun.

Direct sale is an excellent model but I don't think it's a requirement. To me it underscores the main principle which is about reducing dependencies. Why convince two parties to do something (someone to visit your site, and an advertiser to run an ad targeting that visitor) when you can convince one.

I'd also add that the single most important thing I've found in creating a business like this is keeping your costs low. The same two businesses, with the same income and customers will have very different times during a downturn depending on their committed monthly burn rate. I know it sounds obvious, but its amazing to me how many businesses handicap themselves from day one with large spending commitments.

Exactly, direct sale = lower customer acquisition cost

I think the beauty of these types of sites is that we've probably haven't heard of many of them.

I had one for a while--a shareware business that sold pop up blockers, before they were free, along with a privacy suite. Many shareware businesses fit the description. The Association of Shareware Professionals, which I used to belong to, has many members living this way. Their newsletter had an article series about how to run your shareware business from anywhere in the world.

To answer the article's question, the people who do winzip, pretty good solitaire, clipmate and desktop tower defense are probably good examples.

the fortune 5,000,000 company I work at subscribes to GoToMeeting.com - .. desktop sharing.. teleconferencing for product demos and meetings

Second vote for gotomeeting. I use it a lot and have never had nothing but good feelings for it. Does what it does and does it well.

doesn't work on mac. :(

We want all of out apps to be non-media. I.e. Services without ads.

Bug.gd plans to do it by selling private P2P help desks directly.

We have another unrelated service launching soon that will directly charge users for anything above casual usage.

FogBugz...

And Eric Sink's SourceGear. [1]

[1] http://www.sourcegear.com/

Was the footnote notation REALLY necessary?

Are footnotes the subroutines of literature?

smugmug.com

rememberthemilk.com kind of goes halfway, with a freemium model.

In some ways, that seems more sane to me than DHH's pay-only ideology.

37signals' products are all freemium-based.

I'm a paid subscriber of Remember the Milk, but I don't think they do a good job differentiating their paid product from the free product. I subscribed because I felt that $24.95 was a fair price for a superior mobile interface, but there aren't many other benefits for paid customers.

37signals does this differently. Their free products are far more limited, and their paid products offer quite a few more features than their free products. This limits the value to the freeloading customer, but I think this makes for a more sustainable business in the long-run.

>In some ways, that seems more sane to me than DHH's pay-only ideology.

Well, first off, there are free versions of all 37s's apps.

Second, RTM isn't profitable (yet/ever?). 37S is. That seems more sane to me.

That's probably because RTM builds their Paid plan from their Free plan, only offering marginal changes in features. 37signals focuses on the price and builds from there, offering a free version on the side.

I'd agree with that concept, yes. 37s's "free" is just enough to have users say: "Well it doesn't suck, and I could really use that X feature".

Every Mac software company works on this model.

By definition, every software works on that model.

I think another aspect of this to consider is this:

Eventually, most things will be commoditized, but this happens slower for some things, and faster for others. How does the 'dhh style company' keep that at bay - or do they? How sustainable is the business in the face of cheaper competition?

They don't; they have good products and tell a compelling story.

There are plenty of good products that people probably wouldn't pay for these days: web servers, unix-like OS's, and scripting languages; just to name a few. Once upon a time, companies made a lot of money selling Unixes, and for a period of time, there were commercial web servers.

You are crazy. Clorox still has more end-consumer market share than all store brand bleaches put together, even though it is chemically the same stuff and costs way more.

Yes, but Clorox isn't a commercial Unix, unless it was one of the less widely known ones, either...

The point is that you can do that kind of thing in that market sector. Can you in IT? Maybe, but certainly not in some areas. The economics and marketing are quite different.

and for a period of time, there were commercial web servers

That period of time includes the present.

companies made a lot of money selling Unixes

They still do, with Apple being the most obvious.

That is not what he meant because if OS X were just a traditional Unix-alike with traditional Unix-like GUI software consumers would not buy Macs.

>How sustainable is the business in the face of cheaper competition?

This is where branding and marketing come in. Ask Apple, BMW, Mercedes, Luis Vutton, etc etc.

These businesses all sell their products for much more than comparable products, and they do quite well. Why? Because there are two types of consumers that are sold by their marketing: those who appreciate high quality, and those that want to be seen as appreciating high quality.

You don't have to appeal to everyone.

Right, good examples. But those are all physical products. Can you think of online brands that have that kind of power? I can think of some negative examples, like AOL, that most hackers wouldn't have been caught dead using.

...

Indeed, thinking about it a minute, a lot of my experience with 'branding' online is that small and independent is cool. How much cooler is it to have a @yourowndomain.com email address than @hotmail.com? Remember geocities? Another thing that was sort of the lowest common denominator.

Maybe Google has a positive brand at this point, but I pay for very few things there. I suppose I might pay a little bit more to them for some things in the future to have everything 'tied together', but that's getting away from 'branding'.

As for brands with power, off the top of my head:

37 Signals

Salesforce

Flickr (picasa is free, and people still use flicker)

expedia/travelocity (not startups anymore mind you)

> "You don't have to appeal to everyone."

I thought that idea was one of the highlights of the presentation. If you can sell 400 accounts at 40 USD a month, you're going to live very comfortably. If you can't find a market niche where 400 people would be willing to pay 40$ (or 4000 people pay 4$), you might need to reconsider your idea.

This was a big point in the Italian restaurant analogy as well. You don't have to redefine Italian cuisine, you just have to build a restaurant far enough away from the other guys and make good enough food (among other things) to attract a large enough customer base to meet your goals.

blinksale.com followed a 37s path-- consulting firm that eventually gave up consulting, I believe.

Any informations if they're profitable?

I think they stopped consulting base on customer demand for BlinkSale... But I obviously don't know their costs. I'd wager they're profitable, or are sacrificing profit at the alter of growth in the near-term.

Harvest is time tracking app, with invoicing, that I really like (not unlike Blinksale). I think it's the only subscription software that my business pays for. AFAIK it was started by two designers, and my guess is that they're doing well, because they keep adding new features and raising their prices. :)

Github.com, Lighthouse, a lot of companies that make Mac software (think TextMate).

GitHub and Lighthouse are two examples of unproven products. 37signals makes millions of dollars a year, which is what makes them credible.

What makes you say they're proven or unproven? They've each got hundreds of users and I can say that they each have at least one paying customer (my company).

Hundreds of users, you say!

Hundreds of users, you say!

Oh, the sarcasm! Perhaps if you'd actually watched the talk before commenting on it, you'd have gotten the point and wouldn't be so unknowingly ironic. ;)

http://news.ycombinator.com/item?id=168378

Tell me right now that the talk establishes that GitHub and Lighthouse are financially successful (cash flow positive after investment) and I will:

* Apologize

* Watch the video

I don't think I'm going to have to do either.

The point I'm trying to make is, just because geeks love something (like, inexplicably, GitHub) doesn't mean it's necessarily a financial success. Also: hundreds of end-users? A tiny customer base.

It doesn't really matter much to me if you watch the video or not. I did want to take the opportunity, however, to give you a taste of your own medicine after this incident: http://news.ycombinator.com/item?id=135296. ;)

All that aside, the whole point of the talk was that one can live very well off one's web application even if it's only moderately successful. I think the example he gave was 2,000 small business users at $40/month ~= $1,000,000/year in gross revenue. Or 400 users at the same rate if you don't mind only grossing $200,000 a year.

His argument was that you don't need to stress about investors, 12-hour days, and viral loops as long as you're willing to "settle" for a pleasant, financially viable, and capital efficient business with a niche of users who really appreciate your approach to solving their problems. That's the irony of making fun of an app in this thread for only having several hundred users.

You probably would enjoy the video; I certainly did.

(Incidentally, GitHub and Lighthouse didn't require massive capital investments [1] and I'm pretty sure that they're hosted "for free" on Engine Yard. Having a few hundred users would practically guarantee that they're cashflow positive.)

[1] http://www.akitaonrails.com/2008/4/21/chatting-with-chris-wa...

You're probably right about all of this. I'm obviously in DHH's camp: after several VC-funded startups, including an abortive attempt to weasel VC for our current venture, we're bootstrapping and aiming to build what we want, not what VC's tell us they want.

It's the orthodoxy that gets me. "You can't cite DHH's speech without watching it in its entirety! You can't talk to your team about DHH without everyone sitting down and watching it! What would PG think?" And yet, at the same time, we're willing to toss around totally unsubstantiated factoids about the success of brand new Rails apps as proof of a business model.

Rant off.

You could generally get 100 million free users for every 2000 40/month customers. Price curves, check them out!

Very, very, hard. See my 1% comment.

There's a big difference between courting small businesses (which is what DHH was talking about) and courting individual consumers (who are preconditioned not to spend money on webapps). Depending on the sector, people won't even use your app unless you charge for it.

Have we made enough to recoup the six months we could have been consulting full-time? No.

Are we making enough within two weeks of the site launch that we're thinking about never consulting again? Yes.

-PJ Hyett

I would be very surprised if GitHub wasn't rapidly approaching being cash flow positive. Since they've been out of beta for only a few weeks I think that's pretty good. Personally I love their product and it seems I'm not the only one.

I think they're a great example of the types of companies in DHH's talk: have a product your users love and charge a price for it. As a bonus they also offer a free version for open source work for which they have my respect.

dropsend.com - surviving in a particularly crowded market of sending big files over the web and charging for it when there are free services available. Though, they are backed by a HUGE marketing power of Carsonified.

a few people mention it but it doesn't seem to get zeroed in on.

don't just build a solution, build an elegant solution. ideally one that is practically transparent to the people using it. I think people constantly forget that the vast majority of computer users wouldn't know the difference between the internet and their desktop much less between campfire and huddlechat.

freshbooks.com

lessaccounting.com

Key characteristic: a real problem being solved and customers willing to pay for a solution. In the case of lessaccounting, mom-and-pop shops (online and off) that don't use quickbooks or find quickbooks too complicated.

Seems pretty new. We have information about things like smugmug or 37signals being profitable for a number of years...

How about all the many online shops that sell actual products, and others that fall below the radar, that sell everything from property advertising to job listings.

virtualmin

"37signals/DHH style" companies--in other words, not startups? This is interesting, because his advocacy of non-startups is an oblique way of saying "don't start startups". I wonder, did PG and others even know what the talk was gonna be about?

Are you suggesting that they should crush dissent had they known? DHH wasn't saying don't start startups. His point is that people have a warped view of the world and the average startup is more like 37signals than it is like Google/facebook.

I don't know if asdf was suggesting they "crush dissent" - but I do think "WOULD they?" is a good question to ask!

I have always wanted to do a startup that has more mid to long term success as a goal- "37signals-like" if you want to call it that- but I've always felt that there is a HUGE amount of pressure to grow grow grow and sell.

And btw- I really do get the feeling reading news.yc and other venture related sites, that many who are into the startup scene right now look down on anyone who doesn't build-to-flip... but that doesn't jive with all the positive feedback given to companies that buck the trend and hold on longer. anyway that's enough rambling for now.

37 signals style companies cannot be startups? Are you saying that a talk on starting a company that charges money for their product is not appropriate for Startup School?

The goal is not to start startups, the goal is to start companies, operating businesses. That "trend" has been happening for hundreds (if not thousands) of years and will continue after we're all gone, we need to gain some perspective.

This doesn't deserve to be at -5. Let's leave that kind of trashing of comments to other sites, unless it's something blatantly trollish.

It does. He seems to be implying that small companies (or companies with modest expectations) cannot be considered "startups" when, in fact, that's nearly always what they are.

I agree that he's wrong. So vote him to -1. No need to keep going after that.

Reply on news.ycombinator.com