What We Can Learn From Mess
Gary Wolf wrote a Wired cover story about craigslist called "Why Craigslist Is Such A Mess". It's a somewhat strange article because it is highly critical of craigslist's design, management, and lack of innovation. But you cannot read that article and not come away impressed with Craig Newmark, Jim Buckmaster, and the ethos of craigslist.
Here are some stats I pulled from the piece:
Yahoo's HotJobs—but garners more traffic than all of them combined
alone—nearly a fifth of the nation's adult population—it is the most
important community site going and yet the most underdeveloped
estimates that revenue could top $100 million in 2009. Should
craigslist ever be sold, the price likely would run into the billions
more than 16,000 employees. Amazon has more than 20,000. Craigslist has
30
Only programmers, customer service reps, and accounting staff work at
craigslist. There is no business development, no human resources, no
sales. As a result, there are no meetings.
programmers have cut out all the cushioning layers that separate them
from the users they serve."
I do not believe we should criticize a company that operates like this. We should learn from it. Of all the Internet companies out there, the one that serves as the most iconic for our firm is craigslist, not Google. We dream of funding a company that can be worth a billion dollars with only 30 employees. We've never done it and I don't know if we ever will. But we are going to try again and again and again.
My partner Brad wrote a post on the USV blog recently about the freeconomics debate and he ended with this observation.
economic model for a social network that depends on the contributions
of its participants and increases in value as more people use it. One
possibility is that the economic models of these networks will look
more like Craigslist than Yahoo. Recent estimates peg Craigslist's
revenue at more than $100,000,000. Not much compared to Yahoo's
billions, but Craigslist still employs only 28 people. Even allowing
for substantial bandwidth, and server costs, it is still hard to
imagine how their costs are more than $5,000,000. Since Craigslist
collapsed a multibillion dollar classified advertising business into a
fabulously profitable $100,000,000 business, perhaps we should be
talking about the potential deflationary impact of more "zero billion
dollar" businesses. As the radical efficiencies of the web seep into
more sectors of the economy, and participants in social networks
exchange attention instead of dollars, will governments at all levels
need to make do with less tax revenue? That's a scary thought in an era
of high deficits unless traditional governments can learn from the
efficent governance systems of social networks and provide more for
less.
And of course, that is exactly Craig's point:
getting through the day," Newmark says. If most people are good and
their needs are simple, all you have to do to serve them well is build
a minimal infrastructure allowing them to get together and work things
out for themselves. Any additional features are almost certainly
superfluous and could even be damaging.
A system like Craigslist does result in a lot spam, fraud, and abuse. But it is also certainly the most efficient and cost effective way to make a market. We should all be studying this system, understanding why it works so well, and how we can replicate it on our businesses, institutions, and governments. I think we have no other choice.
Comments (Archived):
Craigslist reminds me of the Dutch Marktplaats … http://www.marktplaats.nl … it is very simple and very effective. It also was so enthusiastically supported that eBay decided to buy them, rather than compete with them in the Dutch market.
Drudge hasn’t changed in 15 years and they continue to grow. It takes certain king of personality, to stay with what works. I would tell you that Fred Wilson will not be able to do it. Fred is a curious person who always likes to try new gadgets and explore new and cutting edge ideas. Fred Willson can admire Craig’s model but this is not for him.
that may well be true Ben, but i’d love to find some entrepreneurs to fund like Matt and Craig
Fred, would you like to fund entrepreneurs like Matt & Craig OR ideas that resemble craigslist’s economic model (30 employees pushing 100,000,000 in revenue). My general sentiment from the short amount of reading I’ve done and hearing you speak at ‘The Naked Truth’ in Seattle was that market > idea > team.Being the first major player in this exploding market seems to have more of an impact than Matt and Craig (not to say they aren’t brilliant for playing in that market).
Not sure what is cause and what is effect
May be such efficient businesses don’t need funding?
Most will at some point. I wrote a post about this a few years ago called “why web 2.0 is a blessing not a curse”
i think it all depends on whether corporate finance is going to collapse or not. if the dollar system continues and the IPO market returns and things go back to “normal” i think the finance economy wins. but if not, we’re going to have a broken financial system and companies with 80%+ profit margins with no one to finance them while those dependent upon the finance economy crumble. with those types of margins and reduced competition, and with everyone and their dog throwing an API at you, they’ll find a way to get it done.but i think the real story is when these CL-type companies say to themselves, “hey, with these profit margins, why don’t we become a fund ourselves?” i know you’ve hated on corporate VCs, but i think it’s a got a future, a very bright one, and i think it’s going to be a part of disrupting you and your peeps. ironically i think that is the very type of business you guys want to fund, because i think it will have the potential to be 30 employees creating billions. imagine if CL was a fund and they gave you the platform to launch your own CL type business. all CL has to do is print the virtual currency. this of course solves the banking/money supply problem, which is already the most lucrative game in town.
Most will at some point. I wrote a post about this a few years ago called “why web 2.0 is a blessing not a curse”
Quite a remarkable interview and analysis
Reminds me of the 37 Signals philosophy. Start with the core value, try it out cheap, be your own customer and stick to the essential value. Never forget the essential values your customers live by. Their needs spoken and not must be yours.
Craigslist is a fascinating company, arguably the most interesting company ever “born” on the web. And there’s no question that despite Craig’s general lack of interest in financial reward, it’s of tremendous economic value.Yet in thinking about the wired piece and your post above, I’m inclined to liken craigslist to a virulent disease that doesn’t kill its host. Here’s why – we all use it, it continues to grow like crazy, and yet it delivers (in my view) very poor value as a product. I’ve found most of the apartments I’ve lived in over the past 12 years on craigslist, yet I find the experience of looking for housing on the site painful and irritating. But because craigslist has “infected” the majority of the suppliers of urban (rental) housing stock, it remains the best way to find something good, quickly. I’ve bought and sold things on craigslist, but it’s always a bit of a high wire act, since you know nothing about your counterparty and there’s a high failure rate.I’ve given a bunch of stuff away on craigslist, but the lack of structure, tools and reputation mean it’s a dicey proposition with a high failure rate. http://www.kashless.org is working on solving this problem, but craigslist remains the large majority of the activity here.So in a way, I agree that craigslist should continue to do what they do, because they have managed to build one of the largest, most durable properties on the internet with minimal staff and money. But they’ve also created a crappy product that continues to deliver just enough value that people aren’t quite ticked off enough to switch. Personally, I’d rather see craigslist evolve as Wired suggests or be supplanted by multiple services that address subsets of its uses in better, more useful ways.
Eric, if this is true Craigslist should be vulnerable. That’s really the opportunity here for entrepreneurs. Fred, to me the lesson of Brad’s post is that we need a lot more entrepreneurship and public/private cooperation to hack government and not just education.
right on
Aaron – the question of vulnerability is an interesting one, but this might be a situation where “good enough is good enough”. In my usage over many years, CL is bad but not terrible, so I wonder if users who are habituated to CL are going to leave en masse.
Ideas of dooms day are good for movies. Similarly a phenomenon like users leaving a known platform en masse is never going to happen. I dont recall in the history of web something like this happening to any site.
Friendster?
Between 2006 and 2007 hordes of people left Myspace and migrated to Facebook because MySpace was becoming full of spam, fake people, and advertising.Basically — MySpace ceased being a trusted community. The difference within twelve months time was noticeable.
Still MySpace is far from being dead. Anyhow, in my opinion marketplaces are very hard to kill since you dont visit those sites every day, which effectively makes it hard for negative word to spread. Till supply and demand cycle is maintained, individual users can care less. Another marketplace example, Paypal. In early 2000s there was more than enough negative news about Paypal, but that did not kill the site.
It is very vulnerable. Both in the free world, and in the not so free world.And the idea of “OMG no social media!” is actually the wrong approach. Totally wrong approach. People want to know that someone else will be there with them in this massive buying/selling/finding/helping/marketplace thing.Short story why:So the area I grew up is divided into (roughly) two parts, the people who are orthodox Jewish, and the people who are not. There is some intermingling, but not alot.For some reason, all the people who are Orthodox Jewish in this small area have decided to put together a Yahoo group to essentially mimic Craiglist. You can sponsor a proper ad (I see some), but the majority of posts are sell, buy, I need help, looking for, type things. I’ve seen everything from carshares across the country, babycarriages buying/selling, to trying to return jewelry. People don’t check the vast majority of messages (really busy list), and yet it still seems to work, because everyone knows someone else on the list. It’s hyper local, and is aimed at a very targeted group. They trust the answers they get from the list, because even if you don’t immediately know the person, you know someone who knows them (or connection to another connection). As a result, I tend to see a crowd of people who will give first dibs to items, ideas, events, to this list than they would ever anywhere else on the web. They want confirmation that they have someone else they know seeing them (potentially). (The list is called Five Towns Shuls, please don’t join unless you have some vested interest in that community)It’s why Angie’s list still keeps running. In lieu of the fact that you can’t confirm who the person is you are looking to confirm your web presence is good and truthful and that his/hers is as well, you can confirm that the review is ungamed (ideally) by the paywall.Both sorts of systems allow my people like my mom to find a handyman she likes, and then to cause his business to grow massively. They play on the trust game. CL can’t repeat it. Too big, too anonymous.
Aaron (and Fred) you’re assuming that CL maintains it dominance by being the best at what it does. Hence the argument that it should never change, because it seems to be doing everything right.I’d argue that CL was simply in the right place at the right time. It’s like Ma Bell, an early player that dominated market and is difficult to supplant despite its flaws and frustartions. Fred, perhaps this is why your firm can’t replicate its success — it’s an exception to the free market, not an exemplar.Personally, I’d say that what made CL successful in the first place was finding a niche in a changing landscape. If they cease to change as the landscape does, they will eventually die. The rabid opposition to change amongst the faithful in the CL community is to me a sign of an empire about to fall.
hi Erici have not had the same “crappy experience”i probably don’t use it as much as others but i buy and sell stuff oncraigslist about a half a dozen to a dozen times a year and it alwaysworks great for me
Fred – perhaps crappy is too strong a word, in that I’ve also had far more successes than failures on craigslist. But I’ve also been ripped off once (early on), had people fail to show to pick up tickets they agreed to buy and have spent hours of wasted time wading through spam-like apartment listings or responding to listings only to find a bait-and-switch broker on the other end.The lack of real tools, reputation, payments and filtering don’t render CL unusable by any means, but in an era of increasingly high-quality web services, I think the user experience of CL is mediocre at best.
I think Newmark’s philosophy is that all those features create other problems such as favoring savvy power users and creating incentives to game the system resulting in an arms race. It’s true that using craigslist successfully involves a fair bit of luck, but that’s the very thing that makes the experience good for everyone rather than great for a select few and terrible for everyone else.As a web developer, designer and early adopter, I instinctively tend towards increasing usability and power features. But given the fact that this is one of the very few Internet companies trying this other path (CEO responds to email? Repetitive tasks are valued and performed manually?), I think it’s valuable in and of itself just to see how it stands the test of time.
Gotta be careful regulating free wheeling markets though. The elimination of naked short selling had some unfortunate results
“The elimination of naked short selling had some unfortunate results”hahaha….good one boss….next thing you know you’ll be telling us how great it is that govt can just print as much money as it wants…..
Nope. But I’d like to have that power
Here’s the rub: all the reasons you think Craigslist is crappy are *the* specific reasons why it is so outrageously popular/successful/etc.Craigslist competitors fail because they start with the premise that Craigslist sucks. The opposite is the truth.
“Craigslist competitors fail because they start with the premise that Craigslist sucks.”Completely agree. And if you look at ewiesen’s comments in this thread (sorry, dude), you’ll notice that he starts with “very poor value as a product”. He then says “bad but not terrible” and then finally “mediocre at best” while also saying he’s had “far more successes than failures”. He’s continually backpedaling in terms of how bad the service is (seriously, ewiesen, sorry for calling you out like this).As technologists, we often place far too more emphasis on the experience than the results. Call it the “uncanny valley” of domain knowledge — the more context you have the more you notice the little things. The rest of the world is just looking for the lowest friction mechanism for achieving a result.
Whoa. Eric brings down the hammer with a killer comment. Nicely done
@pbreit, I completely disagree. Nobody thinks CL is a good experience. It’s only successful because it dominated the market early to the extreme degree that it is difficult for viable alternatives to get traction.
oh schnap….some serious truth was just dropped, people…..gotta give props. damn.
I wouldn’t call Craigslist a disease, but I agree with your general point. Craigslist’s greatest advantage is its total dominance in market share, and it overrides the (poor) user experience (in large part due to users, not the site itself).My startup is in the final stages of development on a service that revolves around listings like Craigslist. One debate among us is whether or not users would prefer a more personal experience via invites and connections (like Facebook or LinkedIn perhaps), or do they actually prefer to operate as they do on Craigslist – with open calls into “the vacuum” of the internet?Bottom line – the power and value of the internet is in connecting people and Craigslist nailed it.
My friend mark pincus tried going after craigslist with a social networking model (tribe.net) and failed miserablyHe made some mistakes and the execution was not perfect but there are lessons in his failureFortunately he’s got a homerun on his next startup
Ha… I was being (intentionally) vague in my description of what we’re doing. We are not trying to make a social network out of Craigslist – we don’t touch CL at all, nor are we a social network, but we are in a way returning results/connecting people in some cases “like Craigslist.”Thank you for the warning though (and I’d love to explain more sometime, but I don’t want to self-promote here).My points were 1. It’s interesting to examine the differences in power properties like Craigslist, Facebook, etc. and 2. Regardless of those differences, Craigslist is tremendously successful.
There are a lot of situations where people do *not* want to know the counterparty. Hence Tribe, Facebook, et al being lousy classifieds.
Great point. Made me chuckle thinking about them
almost everybody playing the social networking game is making the same mistake of focusing too much on the big pie….even the niche soc net players, i.e. ning, tribe, etc, still focus too much on the big idea of connecting it all instead of getting the niche environment perfected. connecting it all to create the big mega thing is of course where it’s at, but to get there i think we have to make the niche nodes as awesome and independent as can be first. IMHO of course.
Which is why providing the basic plumbing via an api and letting developers go crazy feels like a winning strategy to me
Craiglist’s success is a mystery to me too. The interface reminds me of a mock-up. I don’t know if their success could be replicated, since I view them as real outliers.
I don’t view them as outliers. Don’t focus on the UI. Focus on the utility they provide and the way they operate the business
I hear you Fred. It’s very similar to your SMS (lowest common denominator) thesis.But can super-basic really stand the test of time? Surely, at some point innovation must win?Hasn’t history shown us this?
I don’t really understand how craigslist isn’t innovative.And we can think of a million very profitable companies that aren’t reinventing the wheel every day. Instead they are focused on efficiency, partnerships, pivoting customer segments, and their overall business process. Why? Competitive edge.Not every product that happens post web2.0 has to be a NEW product. It just has to have a competitive edge. These days we’re seeing more and more products that bring LESS to the table to _emphasize_ the hidden utility in the multi-functional services of the past. That in itself may be a new breed of innovation, but it surely is the “super-basic” featureset.”Hasn’t history shown us this?” is 100% rhetoric. History has an infinite number of examples to choose from.
Don’t confuse innovation with complexity. The Google home page is a good example. Many commenters have pointed out problems with Craiglist. There is still scope for providing a better user experience without adding to the learning curve. Innovation frequently translates into ease of use. Just ask Apple.
Super basic is, in my opinion, the only thing that CAN stand the test of time.
History is not on your side Christian. If that were the case we’d all be driving Model Ts. Technology makes things easier not harder. Most of all it can makes things more personalized. Current web technology leads to a one-size-fits-all style interface. In the future users will be able to choose their preferred level of complexity/functionality. Many desktop applications have been doing this for a while.
I disagree. Although I have zero experience with Model T’s, how do you explain the explosive growth of Twitter? The simple stuff is what sells. Even today, consider what features you use on Facebook? Chat, photos, messages, that’s about it. I am all about the KISS method, personally.
On that basis Photoshop wouldn’t have stood a chance against MS Paint.
Photoshop serves professionals very, very well, but they’ve tried for a decade to reach the consumer market without success. They’ve never stood a chance against MS Paint.Generalists and specialists have different goals, different motives, different contexts and different expectations.
I’m arguing there’s a market for Photoshop, not that there isn’t one for Paint.
Web services like picnik and to a lesser degree aviary have the opportunity to do what photoshop never could do
Twitter is a great example of this, Shane. I think you and Christian both make great points. I don’t disagree with David (his example of Google’s homepage is astute) but I think that, when it comes to community building, simplicity (and a low rate of change) fundamentally works because it makes the technology (and subsequent relationships that the tech enables) accessible to Joe Citizen. We geeks tend to make the mistake of thinking everyone in the world is (and wants to be) tech savvy. Not so. Not even close. Most people simply rise to the level of understanding/proficiency that allows them to get what they need from the system and then they simply stop learning – with Craigslist getting to that level of proficiency takes all of 10 minutes. Geeks (early adopters) are different, we try everything new (whether we need it or not) and then decide whether it’s something that belongs in our tool belt. The general population on the other hand first ask “do I need this?” THEN they evaluate whether they are willing to put in the time to learn the site/system. That’s a critical and fundamental difference. There’s an invisible threshold for most people inherent in the ratio of “perceived immediate need” to “difficulty in learning” that stops most people from ever using anything new. For Craigslist that ratio is SO high that it’s a no brainer….maybe that’s the key.Remember, with Craigslist, the initial relationship facilitation is the product, not the site. The perceived value is high enough to convince even the most basic user to take 10 minutes to become an expert, and because the UI hasn’t changed in ages those people CONTINUE to feel like experts every time they come back – they take one look at the site and think “I still know how to this”. Everything that happens after that initial point of contact that Craigslist facilitates doesn’t matter – it’s all done by more advanced tech anyway (cell phones, email etc)…that’s the brilliance of the site.
I agree 100% with everything you say.However, why not boot Craiglist with the classic UI and give savvy users the choice of switching to a more complex and powerful view. Think of it as market segmentation for GUIs. I’d really love to hear Seth Godin’s take on this…
Perfect solution. But they’ll never do it.
i finally figured out how to use Diqus (case in point!!) and saw this reply. Awesome thoughts, Steffan– and a much better articulation of what I was trying to say
We have to identify what innovation is in this context. UI innovation is often simplying the overly complex. The model T did that too
Basic cars outsell fancy cars by an order of magnitude.
Compared to the Model T no cars are basic. Innovation made cars better, cheaper and easier to use.
Nope. Beyond a few spacing issues (center justify as screens get larger), and the message board needing a little cleaning up (just a weird layout from what I remember years ago), actually, I love the design. It’s like Bauhaus for the web.Old fashioned, ordinary grids. We’ve been using them for years, most of Manhattan, and Washington DC, for example. Their part of our everyday lives. What’s wrong with having classic grids for a website? It’s a great standard, even the Romans like it!This makes me want to take a screenshot, I appreciate the design so much. It just is so, classic, like the black dress of web. I know where everything is, what is important on my page, and I would hedge that every machine can use it, so I am not in trouble if I have some special usability issue. That’s top notch.I wish more people would design to that level of standard.
Shana, I’ll see your ‘nope’ and raise you a ‘hello?’. Even Fred isn’t defending the interface.The magic of Craiglist is that it can make so much money with so little staff. They can do this because they focus on a simple service to satisfy a common need (and also because it’s mostly free).My main point is that sooner or later someone will come in and do it better.Revolutions (as I’m sure you know) are fond of eating their own children.
For the API-yes.And not for Grids, for the page layout, the simplicty of what they did which I was praising. People often forget that one of the reasons they can afford to keep the staff so low, is that there is barely anything to break. They chose simplicity, and I praise it, because at least for a front user who has no clue what is going on, it is all that is needed. It is why we still use so many simple tools, perhaps improved on, that were invented thousands of years ago. We may have changed them radically, but the principles behind them are the same.And Oh yes, I can quote the bible about Mothers Eating Children, from Lamentations, but My favorite phrase is from the opening of Ecclesiastes, 1:4, ‘One Generation Pass Away: The Earth Stands Forever”The Grid stays. We’ve been using them for thousands of years. That’s not changing anytime soon. People like them. We still write about the best way to use them. http://www.smashingmagazine…The API will probably go. That’s the way the cards fall. Something about being human.If it were a plow, the revolution would be in making it steel, and giving it an engine. Not making it any less of a plow. You still can build and use a Bronze-Aged Plow, that discovery of plowing really can’t be changed.
“The magic of Craiglist is that it can make so much money with so little staff.”That is made possible by *not* hiring pixel-manipulating photoshopping CSS Ajaxers.
Yeah! And let’s bring back the blue lines around Superman.
Exactly! The originals are almost always better than the sequels because they focused on story and characters, not special effects and whizbangery.
I gotta say, Iron Man had a more compelling story and noticeably superior special effects to Super Man (Christopher Reeves) and it’s sequels. I realize this isn’t a fair comparison but it does show that evolutionary technology doesn’t inhibit the essence or ability to convey a great story. The value is in the complete package, and it’s a moving target based on human interests and needs.
I dont really think AJAX developers are the same as photoshop designers
No, but it’s a similar problem. A lot of Ajax is unnecessary, doesn’t improve the usability, works unpredictably and is harder to maintain and improve.
David, I see a problem with your Model T analogy. Your logic stated that iff simple “super basic” is the only interface that can stand the test of time, we would all be driving Model T’s. Cars became more powerful and more fuel efficient as innovations were made in transportation. People were able to get places faster, drive for longer ranges, and be safer. Cars have arguably become easier to drive and operate (thus, “simpler”) since Ford’s automobile.That said, craigslist’s design, and other super-minimalist designs can maintain power and usefulness amidst innovations that happen in websites around them. Point being, do you then believe that craigslist can be made any simpler through a redesign or a new approach, knowing that technological innovation or new webwide standards do not necessarily simplify usability?
At the end of the day what counts is the user experience. Boiling everything down, you could say the user experience is basically a function of the user interface combined with the underlying functionality of the system.Craiglist has prospered in part because it offers a very simple interface to control very simple functionality.One point that has not so far come up in this thread is that most people would rarely use Craiglist for an extended period of time. It’s not like MS Word, for example. In other words, there’s no point learning to use a complex product if you don’t plan on using it regularly. This means that for the casual user anything other than a totally obvious interface represents a barrier to to entry.That said, the user experience can still be improved whilst maintaining a practically flat UI learning curve. I’m talking about automated spam reduction, implied reference checking etc.If more functionality can be added without compromising the interface then all the better. When is the last time you were required to modify the air/fuel mix in your car engine, or manually specify the bit rate for your DVD player?That’s the kind of innovation I’m talking about.
Super-basic is the *only* thing that can stand the test of time. Companies and service always get into trouble or leave themselves vulnerable when the depart from their super-basic childhoods.
I think there is truth in that statement. But it doesn’t explain facebook’s soaring popularity
What is innovation though?
Making things better.
Yep. A to B to get to C.
It’s free for 99% of users, and amid the mess, it delivers value. It’s not an easy model to replicate, but it can be done. Ask Marcus Frind of PlentyOfFish.com. Estimated $10M+ in profits in 08, 1.3B pageviews in July, two employees (including Frind), all from a free dating site he built as an exercise in learning ASP.NET.
yes. anyone looking for the next iteration of craigslist should check what markus is doing.
Another mess we can learn from
Great article on Plenty of Fish and what Marcus has done with it back in January in Fast Company
I could not disagree more.Craiglist is easy to use, effective and free.The only other buying / selling site that I have used is eBay and Craigslist user experience is FAR superior to eBay. eBay is a pain to use, is costly, involves using Paypal (nightmare) and is scammer central.I really like the fact that Craiglist doesn’t change. OMG – no social media!! Please. Craigslist is a simple but powerful tool that serves its’ users well and apparently is financially successful.
more likely the crappy experience has nothing to do with cl and everything to do with the people involved and inherent nature of the transactions – buy / selling used items or finding an apartment. no site, system or social network is going to make those processes any less of a high wire act or increase the ultimate success rate.
What an awesome thread. Nice work, folks. I’ve essentially had mostly positive experiences with Craigslist and I am a great admirer both of the simplicity of the business and of the way that we all seem to understand the core value proposition and accept it. Many of the needed features and reputation systems that some folks seem to crave seem to be a solution in search of a problem, almost like this notion of the ‘semantic web’ that most people actually never asked for but which is apparently a very pressing problem to a small group of developers.It seems to me that a powerfully simple tool/business plus (by design or by accident) a great brand management framework is a much better business and a (mostly) better experience than a vastly more complicated tool/business/system.Maybe the competitive start-ups fail because the problems they’re trying to solve are more expensive to solve than the marginal value they impart to their users and the simple system gets it right 70% of the time. And we accept that failure rate because their UI and their reputation have communicated the appropriate user expectations. What a case study.
I’m with you on the quality of this thread. This is what makes avc a must read
Thank god Craigslist made it through the 2.0 era without adding a social network on top of it as many would have suggested. What always strikes me about Craigslist is their email proxy – such a simple idea yet who else has leveraged this as well? Never underestimate the value of the underlined text link.
Good points
“People are good and trustworthy and generally just concerned with getting through the day,” Newmark says. If most people are good and their needs are simple, all you have to do to serve them well is build a minimal infrastructure allowing them to get together and work things out for themselves. Any additional features are almost certainly superfluous and could even be damaging.”Not to get off topic, but the same could be said as applying to government…(which would be a good thing)
That was brad’s point. But then he’s always been a libertarian
brad’s a libertarian? be careful, libertarian is the socially acceptable term for full blown kooki’ll have a 9/11 truth shirt mailed to your guys’ office
Kookology is alive and well at usv
It’s hard to argue with the massive success of Craigslist. Organizations both large and small use it now. They are brilliant in realizing that every new feature, piece of complexity, or controlling filter would require a disproportionate amount of additional resources.
Dead on. The idea that Craigslist has not innovated is pretty preposterous too. I see constant change on their site. I started getting spam on the e-mail forwarder…within weeks, they had switched to a reply button with a captcha.Craigslist hasn’t innovated the way a lot of people think it should. But it’s hard to argue that they made the wrong call looking at the results.Now the question is: if a VC had invested, would they have allowed them to leave so much money on the table? They probably could have hit a billion dollars in revenue if they had wanted to, and they still wouldn’t need any more people. Just flip the pay switches in more cities or categories.
I’m glad someone mentioned the “leaving money on the table” argument.To me, the key is that if VCs had been involved, then Craigslist wouldn’t have had the fantastic growth. Their “minimum profits on maximum reach” model is really a tremendous strategic weapon.Albeit one used by peaceniks…
You’re right, though I would call it “maximum profits on minimum revenue and maximum reach.”The 30 employees to generate 95% margins on $100MM in revenue is simply impressive. I want to have a company with a fraction of that potential margin and scale.
Yep, I agree. I think it comes down to people’s ego, as well as greed. I would love to be making that kind of dough without having to worry about the next board meeting, or how I need to pitch this feature or that feature…
You can!
If you are pitching features to your investors, something is wrong
I don’t agree. Look at how many VCs, including this one, are behaving these days. CL has been an inspiration
It takes a certain level of vision in the VC. I haven’t talked to many, but many of those I have would never have funded Twitter for it’s A round.
More will on the next one
All you need to look at is the restraint VCs are showing across the social media landscape to see the answer to that question
That’s a very good point.
Good Call, I’m starting to get grumpy with social media. Being social is good, but you don’t need to make everything a social media whatever to make it work.
I’ve never understood pure web companies that hire huge numbers of employees. The whole point of doing business online is increased margins largely developed by reduction of expenses.
Agreed
Precisely.
I think it’s an ego thing, people think you have to have hundreds of employees to be successful. Didn’t anyone learn anything from the late 90s?
Companies have very little restraint.
Very interesting. Brad talks about the deflationary impact of having lots more “zero billion dollar” businesses. Thinking through the economics there should be a benefit of re-deploying labour somewhere more productive. What businesses/opportunities will be created by more zero billion dollar businesses? Or are we reaching a point of maxed out productivity? O’Reilly wrote about this last year here: http://radar.oreilly.com/20…. I tend to disagree with O’Reilly’s point as I believe that there are near endless iterations to productivity improvement – before the steam engine people thought you would die at those speeds. What do you think?
I think we need a new measure of productivity
One thought was “lets measure leisure time.” New problem though- some people think that there are groups where are people that all they do is work, and the rest there isn’t enough work to do. Another problem is what counts as leisure time, since now there is actually a lawsuit going on for overtime about emails answered after work on blackberries…work and leisure are mixing too much?We got problems.
Speaking from someone who is enjoying a typical workation, I’d say we are doing great
500th comment.A) Enjoy your Workation!B)This may be my perception of time, I conceive time as objects, and when objects like to float in and out of each other, it makes feel very uncomfortable and disjointed. It’s hard to conceive in time as an object what is my time off in private, what is my time off in public, what is my time in work in private,and what is my time in work in public. It blends- sometimes well, and sometimes not so well. The moments of not so well are honestly very confusing, because it is unclear what to do, and who or what I should be connecting with.So as much as you enjoy the workation, if we were in the reverse positions, I probably would take less time off, but it would be more Walden-esqe. It is good to sit in a total blackout, or as near to it as possible, in order to refresh the spirit and re-invigorate the power of ones senses and observation. Even if it is just for six hours of total blackout. At least, then I know, it is definitely an object that is mine.
That last comment about needing a new measure was written on my blackberry sitting by my pool soaking up the sun
I like the sound of the new measure! The French might even like it 😉
You know what I hate? When people look at a product and say “this is going nowhere” (or “this is going to be big!”) without asking how it got there in the first place. It happens all the time and it really bothers me.I had a conversation a month ago with a friend of a friend who wanted to start something similar to Craigslist. It was going to focus around swapping free stuff and finding roommates. The Craigslist model, he argued. I suggested, instinctively, that he had to start by getting event listings together, in order to attract the community, and he poo pooed it as “not part of the idea.”I didn’t know it at the time, but it turns out that is basically how Craigslist started; Newmark had a small list of friends, he told them about some area events (for coders, I believe), and it took on a life of its own. So he moved that to the web. The community dictated the product, not the other way around, and here’s the lesson: There’s no reason why Craigslist should change that ethos now.Wikipedia is the same. I actually just wrote about their history on my blog — http://dlewis.net/2009/08/2… — because the amount of things we can learn from figuring out how things *actually* happened is immense.Takeaway: Articles like Wired’s upset me because they look at where we are, not how we got there, and assume we got there in a way which we didn’t.
Don’t let that wired piece upset you. Let it inspire you
I question how many VCs would fund a guy like Craig Newmark. Imagine him walking into his VC pitch meeting and saying, “We don’t care about profits. We’re going to only take the minimum we need and leave billions on the table. We’ll do no marketing or sales or biz-dev. We have no exit strategy. We’re going to give the product away for free.”How many VCs would invest? How many would do the ROI calculation and walk away?For founders of Internet companies, I’m really excited about the possibilities of building wildly profitably small-cap companies (< $100M a year in revenues) with minimal employees and infrastructure costs. You might call them “lifestyle businesses”, although I dislike the term. But does the VC math work out right? I don’t think it does, except for the earliest investors.
That is an excellent point. I’d like to say we would have invested if we heard the original vision but it’s hard to say that with a straight face.but now that craig has pioneered this trail and inspired us all that it can be done (although it’s much harder than it looks of course), i think we do need to figure this out.The “VC math” would work out if you were early.There will always be opportunities for liquidity for extremely valuable companies. It might come from M&A, IPO or even secondary offerings.
I’d go a step further Bijan. This may be the new way it is for many of our investments
I can imagine situations that are staff heavy for a variety of reasons.Not everyone needs to be the programmer guy, and if you are aiming at some verticals that have some, errr, interesting tech needs and a lot of drama, you might want to have staff around early on to be the nice pleasant people.
Am I the first person to use the ‘Like’ feature today? 🙂
Why would such a company need investment?As I recall it Craigslist didn’t need investment… I believe the only outside ownership is the stake ebay bought in 2004 when the company was 10 years old.
Usually areas like customer service, spam and abuse management, and scaling the infrastructure end up requiring investment ahead of revenues and that leads to VC investment
I agree that the “VC math” works for only the earliest investors. I expect that most founders of the lightweight companies you describe wouldn’t want professional money – they want partners who share their vision and investment horizon (e.g., Craig is in this business for life…his investors should be too). With lightweight startups like these, the internet may turn Venture Capital into a Zero Billion Dollar Industry.Is there any precedent for a VC taking dividends as a permanent exit? I can think of a number of problems with this (e.g., how do you calculate carry; LPs expect true liquidity, not a stream of dividends), but if the company has passed through its most rapid growth phase, does not require much capital to sustain expansion, has no plausible acquisition targets, has no material debts, and has a product which may be hurt by extending into other areas, why not?
I don’t buy the argument that professional investors don’t have the right passion to be good partners for entrepreneursIn fact, I’ve dedicated my career to proving that assertion false
And if a VC did invest and gain a controlling interest how would they prevent themselves from destroying the company? The golden goose theory can only offer so much protection. At the end of the day VCs are sitting in expensive offices waiting for returns. They will certainly delay returns if the risk looks good, but the worldview is not compatible with the spirit of public service that fuels a guy like Craig Newmark.
I think your view of VCs is warped
I’d love to hear that pitch when its aimed at a billion dollar market
I didn’t read this comment before I posted mine, which makes a similar point but your articulation of it if way better. I think its really important to build “wildly profitable” businesses FIRST, and large businesses SECOND. I have been in a situation, where we tried to build a large business first and 9 years later, profitability is still not predictable quarter over quarter. A lot depends on the nature of the team, their purpose and the objectives of the investors, which I agree are tough to come together to create a CL like business. I believe a lot of CL like business will be built by bootstrapped startup founders than VC-funded ones. I may be wrong, but it seems more likely.
The thing is, Craig Newmark didn’t need VC. Considering the lightweight website, as well as the lack of extraneous support staff means that the service can grow organically. In addition, remember that Craiglist was founded in 1996, the first-mover advantage is very difficult to overcome for new services taking them on.
The truly successful online businesses both in terms of revenue/engagement concentrate on function over form.Craigslist may perhaps be the most obvious example but there are many many others. Twitter only recently with $55m funding/44m monthly uniques – decided to re-design its homepage to something which could have been coded by a 12 year old.Facebook had 100m members before it gave its homepage a brand identity.ebay/amazon – all the giants – shitty form (initially at least) – but great function.Its disruptive innovation/blue ocean strategy through and through – reduce elements which are secondary to requirements (form) whilst over compensating on others (function).Those of us who are entrepreneurs need to remember that whilst we want to build great things which have both form and function, utility and beauty. We need to get the utility nailed down before going after the beauty side of things.Its hard work, we hate to release products which we know can be drastically improved in regards form – but we must must concentrate on function.There are tens of thousands of the most beautiful portfolio/design agency sites out there, glistening design, intricate styling – but bubkis business/revenue.All our businesses can grow by taking a leaf out of craigslists book – the only question is whether our design-sense can stomach it.
This whole conversation is very interesting. I especially like your point about concentrating on function over form when pursuing revenue/engagement.
Great comment and so true. I’d argue that form gets in the way of function in many cases. Form is nice to have but function is a requirement
It’s why Bauhaus still lives and is still so attractive to us. At the end of the day, it admits to bowing towards usability in order to make something beautiful.
Great point.Craigslist ain’t sexy, but she knows how to work the room.(Related: As a designer I was intrigued by these spec redesigns over at Wired. http://bit.ly/10MHQUI really liked SimplyScotts. Simple and effective, with minimal disruption to the existing layout).
Beautiful, and I agree about the SimplyScott. Most people don’t realize looking at the current page that it easily is enabled for say, Braille Displays. Solving for a phone is simple in comparison.
When I read this: “Of all the Internet companies out there, the one that serves as the most iconic for our firm is craigslist, not Google. We dream of funding a company that can be worth a billion dollars with only 30 employees. We’ve never done it and I don’t know if we ever will. But we are going to try again and again and again.” …I couldn’t help but think of Twitter.Twitter has the potential to become such a company if they can cover their costs with business/verified accounts or if they can come up with another elegant revenue model (like (ahem) my Twitter Answers idea). But Twitter certainly has that potential. It also has the potential to become Google-like if it truly becomes a great real-time search engine.Or it has the potential to be something inbetween these two extremes (and of course it has the potential to crash and burn).Anyway, when viewed through this lens, I understand Twitter’s strategy much better.
twitter needs to get away from the noise and focus on the signal to really mean something. Too much conversation – too little transaction.
I guess. It’s easier said than done. Third parties like Bit.ly and Tweetmemecan help in this regard.
While I agree with this critique, you could say the same thing about CL
Twitter is operating in that direction but they’ll be 100 people by year endSo we need to keep looking
Wow, that’s huge. Why the expansion, might I ask?
All the things you’d expect. Dealing with scaling, spam, abuse, corporate accounts, search, api, developers, customer support
No idea it would take so many people so soon — and yet Craigslist does most of that with much fewer people. Amazing.
All the things you’d expect. Dealing with scaling, spam, abuse, corporate accounts, search, api, developers, customer support
However, Twitter has already taken $55M in VC, and, depending on whether or not you choose to believe the leaked strategic documents, they have designs on employing hundreds of people. It will be interesting to watch them, surely.
I’ve been preaching contextually matching personalized (two-way) search and advertising. My Sixth Sense has taken a swing at the search portion, but they haven’t allowed for us to control all the dials (something I can’t stress enough). Users need to be able to tweak their preferences on how information sorting is done, not just the inputs.
I thought the Craigslist article was one of the most wrong headed articles I’ve read in Wired for a very long time. The author seemed to equate developing lots of features or being cool with success (rather than happy users or amazing financials).Jim Buckmaster came to Edinburgh to give a talk a couple of years ago. It was in the management school so there were lots of MBA students there. One of the questions was about what Craiglist’s mobile strategy was. Jim’s answer was “We get more companies offering there services to put us on a mobile platform than we get users asking for it.”Kudos to Craigslist for building what their users want not what the web 2.0 crowd think they should.
Better answer would be ‘we’ve got an api for mobile developers. We can’t wait to see what they come up with’
fred wilson 1, jim buckmaster 0
How would APIs help CL? The only people it helps are startups who want a share of the pie.. not CL
if you won’t share your pie, you force people to bake their own
I am one of those fans of craiglist business model – as you said, we should learn from it and see how it can be replicated in our businesses instead of envying on it’s success.
It’s a fallacy that “governments will at all levels need to make do with less tax revenue”.First of all, if craigslist has >90% net profit margins, corporate income tax will approach 30-35% of revenues. That’s 30 million a year right there. It may not be much compared to the amount of income taxes earned in the previous era of a multibillion classifieds industry, but it’s something.Secondly, and this is more important, the money that people save by using craiglist will either be spent (generating corporate income tax) or saved (to be used for income tax-generating investments). People will derive utility from using craiglist for free, and from whatever else they spend money on. This productivity increase boosts GDP, and the government will get its cut sooner or later.
I think this has a lot to do with the basic function being offered. Craiglist has managed to come close to the Einstein limit (“as simple as possible but not too simple”) as has Twitter. Only truly foundational offerings can get that simple, because they are so critical to the economy that they can afford to specialize completely. Classifieds are simple. Shipping containers are simple. Plankton are simple (and account for a big fraction of the earth’s biomass :).The ‘ugly’ I think is unnecessary though. You could keep things simple and at least pick better fonts and colors. But I think that is a deliberate choice on Craigslist’s part: to convey a brand of homespun simplicity.Companies in a space compete on complexity and diminishing marginal value features as they evolve though, so I wonder about the fundamental contradiction. What is it about the classifieds space that prevents the next set of features from adding enough value to overcome customer switching costs? Is it merely the network effect?Could a powerful open API, vastly better search or buyer/seller analytics disrupt Craigslist despite the network effect inertia they enjoy? Pair that with an entry strategy that cuts orthogonally in some way to Craigslist’ regional model, and you stand a chance.
I think I’m going to start calling this the “plankton approach”Thanks for that one!
Craigslist certainly has some issues, and dealing with spammers and non-serious buyers or sellers is a hassle, however the alternatives are all laughable. This is free, wide ranging, and ridiculously easy to learn and do. Its also extremely mobile friendly, before the mobile web was even something we talked about. I remmeber using craiglist all the time on my nokia 9300 communicator in 2003 or 2004, and it was one of the only sites that just worked.
craigslist’s very high margin profitability seems a given. still, while their revenues (or any numbers) are widely speculated about and oft cited, to my knowledge, the company has never confirmed or published any real financial data.anybody have any hard info?also, due to the infamous murder case in boston and several law enforcement officials and agencies public denunciations, 2009 has seen all but the end of one of craigslist’s biggest cash cows — paid listings for the prostitution business (the vast vast majority “adult personals”, “escorts”, etc).anybody know what percentage of craigslist’s revenues have disappeared?
You have it backwards: those listings were free until the backlash caused by the publicity resulted in CL starting to charge for “Adult Services” listings.
Ah.Silly me assumed adult classifieds just had to be a paid media modelWow! Huge revenues while giving that stuff away free!
I think the simplicity of design and functionality of craigslist is so obviously why it works. That said, as someone who’s developed and product managed a few sites now, every time I start a project, I think to myself, I want to make the craigslist of [INSERT INDUSTRY NAME HERE] then I get into developing it and before you know it I’ve added color and shading and one thing leads to another and I end up with an AJAX/Flash based site that is buggy as hell.I would assume this happens to a lot of people and I admire craigslist’s discipline and hope they never change.
was waiting for this post ;)the point is well taken, although i wonder if VCs will ever be able to invest in the Craigs Lists of the world or if their philosophy on life and modest capital requirements will always make them the kind of companies where we don’t get to play. like the girl that got away…separately, it does beg the question as to whether businesses like Twitter should be taking a more Craig-like approach…
Mo – I am now waiting for your post. A comment is good but there’s a lot of good questions in this that you should dig deeper on
fair point…
I’m curious to know what the demographic mix is re: their user base – being in the UK/Europe, I feel pretty detached from the relevance of Craigslist and must proclaim a general ignorance re: their business.Over here, I don’t even know a single person who has used it – most have never heard of it. But, Hyperlocal is where it is at and they seem to address that pretty well (in the USA at least) – in a retro UX kind of way …. ;-)What we should celebrate is a business sticking to core principles/passions – so often forsaken ASAP in a start-up and so important in even a mature company – ie, running it with passion and personality. All too often the first steps seem to be to ASAP endow a start-up with traditional Executive management baggage. Often with very dubious results and usually just to ‘tick the boxes’ – look, we’re a proper company.Also, their model is Disruptive, in a rather oblique manner – that always upsets a lot of people. Legacy values and mind-sets feel threatened.Which is good.
Traditional executive management baggage: so true.”Time to bring in the suits and kill the company”
Groovy thoughts Egoboss (especially the psychology norm of disruption and upsetting many folks). I’m trying to find common threads of success with ultra streamlined businesses (engineers, and eventually a person that answers the phones/emails).
I am not sure I agree that it works so well Fred. Sure it works well for selling a toaster or a tv but what about finding an apartment? While it would be interesting to look at why CL supposedly works – it’s not because of anything but because lunacy goes on there. Let’s just look at finding an apartment – I have very intimate knowledge of CL’s apt functionality because I am considering moving to 5 cities.I am trying to find an apt in nyc and for the last 3 yrs have tried using CL – I am working on a VERY long post about why CL is like death – and there are people i talk to who would rather have death than deal with CL. Brokers love CL because they can do whatever in the hell they want and there are no police. Fake photos? Sure. Fake listings? Sure. Bring it all to CL, we take it all. I just received a reply to an ad – the man told me to send over my credit report before he shows me the apt – yea that’s gonna happen. Why do brokers post the same ad 3x a day – why do I have to basically sit there all day and watch for posts. Why are brokers allowed to spam the fudge out of their listings with keywords to show up in every search? Why is Roosevelt Island considered midtown east one minute and LIC another?When I wrote about leaving NYC next month, one of the reasons was the fact that the apt search just sucks so bad here. Naturally no one really has made any dents against CL because the brokers love the ability to do whatever they want on CL.Why can’t I exclude a broker? Why can’t I leave a comment about the broker? Why aren’t the brokers required to post real photos (or no photos)? I could go on for a day and once I get a moment will do so on my site. I know some new startup out of one of the incubators is trying to create something in NYC – what they will realize quick is that the brokers want nothing but the hell of CL because it works best for them.I assume you didn’t use CL when you bought your home that you post photos of on the west side – if you’d like I would be happy to stop by USV and show you what it’s like to try to find an apt in NYC. 🙂
That’s a great perspective and critique. Clearly real estate on CL has issuesAnd you are right. As much as I’ve used CL, I’ve never used it for real estate
I hear your pain about the apartment search, but I think you’re misdiagnosing your problems. The root cause (for most of the issues) is the NYC rental market, not necessarily craigslist.If you want the Netflix solution for apartment hunting – it’s gonna be a while (unless Fred knows something/someone) ;)In the meantime, check out http://newyork.cribq.com/ – it’s a nice UI built on top of Craigslist. It might alleviate some of your pain. Note: CL is fully aware that this site exists. As I understand it, they have some informal agreement in place.Regarding quality control, I think the self-governing features within CL (miscategorized, prohibited, spam) are powerful tools given the maximum reach that CL wields. Bottom line: If most everyone uses CL and everyone chips in to “clean up” the garbage – shouldn’t it be cleaner than before?
I think you’ve identified an area on CL that can use additional framework help. Maybe real estate should have an offshoot with verification of some type? Are there serial renters that would be willing to organize this?
Arguments over Craigs List UI are comical — if it didn’t work, people wouldn’t use it. There is this belief that we need to monetize everything to its utmost potential. CL makes plenty of money, thank you very much. What would kill the goose is a bunch of “executives” who would pay themselves a fat salary, take the company public — ensuring that it would need to grow indefinitely to please investors, and most likely drive the company into the ground (cough Yahoo cough).Rock on Craig! Nothing better then being your own boss and being able to tell all the internet “experts” to go blow.
I was dismayed by Wired’s conclusions as well. By any measure, Craigslist is a raging success. But I wonder if the community would have been less successful if it had been VC-funded?This is not to criticize VCs, who generally help entrepreneurs like me think more strategically about creating a valuable company.But as anyone at Facebook could tell you, there is always tension between the community that creates the value for a social network and the network’s ultimate need to monetize that value. Social networks have to listen to their users in ways that other sites perhaps don’t.One reason Craigslist has been such a huge success has been that this tension between the network and its participants doesn’t exist. If Craigslist were VC funded, it would operate differently, and people would feel differently about it.Obviously Craigslist could make more money than it does now, but its ability to generate obscene profits may be limited by the extent to which it could do so without making all of us feel like suckers.In some ways, I think this is what happened to eBay — which once seemed as entrenched as Twitter, Facebook and Craigslist do now — but because of the way eBay tried to monetize its community, I just sort of feel gross about it now.Anyway, I am certainly not opposed to profits or venture capitalists seeking profits, only curious about how the profits social networks create must in some sense be licensed by their participants.
This is turning into the dominant theme of this thread. I’m not going to repeat myself glenn so I’d encourage you to go look at some of my other replies
Oops, that’s the second time in the past month I’ve repeated what someone else has said. And agreed that VCs have become more patient, and more focused on margin/leverage than revenue.
Is it an Occam’s razor solution that Craig’s List achieved? Classifieds represented incredible revenue models at a very low cost for newspapers and periodicals, the communication conduit of yester-year. Didn’t he simply move it to the Web? By treating a focused site as a generator of classified ads–a proven model– then pumping out relatively exponential amounts of them while enjoying being the first to do it on this scale while located in technology’s ground zero position can it be viewed as having been the simple next step in the interactive communication paradigm? Do I recall Craig’s list also enjoyed a branding experience tied in with dating (and let’s face it more questionable types of relationships) connections in a brand new, instantaneous, exciting and credible venue at the time? Could the explanation being sought on which to model replication lie within the sociological foundations at the core of Craig’s List’s origin? Might it be that singular? I am not the technology guy…so please, my questions are simply questions.
Craig Newmark is one of the few web pioneers that hasn’t forgotten what IT’s all about.Long Live Craigslist!!
The UI is part of a truly unique brand identity – users’ trust in Craig. Users connect to the site because Craig still maintains the list, Craig isn’t marketing to us, Craig is just there if we want to use him and his list service. A snazzier UI or new product brand names would belie the very premise that makes us use Craigslist. I can’t think of a better example of monetizing trust. I bet you would’ve invested because you trusted Craig.
To be honest, that is my one nit with Craig, Craigslist and company; an almost Luddite-like sense of not wanting the outside world to extend their offering (via APIs, add ons, etc.). The “users” aren’t asking for that is a pretty common “tell” about innovation philosophy, inasmuch as via APIs third parties could build mobile and desktop apps, composite services and the like without mucking the core of Craigslist.com one iota.That said, given THEIR ambitions for the service, I think that they have done a remarkable job ignoring the noise of media cos, VCs and would-be partners in pushing product evolution for the sake of product evolution. The fact that my mom (total non techie), my wife (lite tech), my nephew (geekoid) and myself (no comment) all can get past the hurdles when utility calls, speaks volumes.I would underline your iconic reference to Craigslist as a model/iconic approach to emulate. It’s the ultimate return on equity.
Fred thanks for the post as somebody who spends all day running around 65 classified sites, 19 of which we’ve acquired or hold and interest in I hope I can add something to the conversation.As to Craiglist’s success I think your point concerning the most efficient and cost effective way to make a market are spot on but might be peculiar to the classified market.Prior to the web and in particular Craiglist the classified market was poorly served and certainly in the US exploited by the newspaper industry.Craig or perhaps more accurately the web simply made the marketplace more efficient – with no barrier to entry for sellers and no barrier (cost) to prospective buyers the marketplace was massively expanded. At some point a virtuous circle of more sellers generating ever more buyers generating ever more sellers was achieved. Craigslist have refrained from exploiting this which to some is admirable but to me with a mortgage to pay is odd.Craiglist is certainly not unique similar sites in Marktplaats in the Netherlands (mentioned by another commenter), Loquo in Spain and Gumtree in the UK have all achieved similar popularity – modified by the size of their market. All of course now owned by Ebay.I’ve seen and the business I’m now involved in own a number of classified businesses serving niches – boats, horses and birds to name a few. All started out in the same way to Craigslist – free to list, free to view, basic ad placement and basic navigation.The combination of free and a really simple user interface plus first mover advantage makes them work. At some point either the commercial / business part of their market hears about or is told of the audience and they start to make significant monies with a very low cost base.Are there lessons to be learned – making it free helps create an audience / users, making it simple (just good enough) also helps enable an audience – whether conciously or not both these solve an advertisers problems eg I want to place an ad but its difficult to do, or I want to place an ad but I don’t want to risk the expense if I don’t sell.Have to say I also share Craig’s general believe that people are generally good in any society of any size there are bound to be people who are fraudulent etc but don’t build the system / service for the exception.Apologies for the length of the post but just one further point – we do have several businesses that have now gone on beyond the basic CL interface and they have substantial social elements – are they necessary, possibly not now, but they certainly do differentiate our offerings and ultimately the users have decided they like them.
having lived in san francisco where Craigslist was ‘born’, I came to quickly realize that every area of potential commerce (rentals, dating, events, sales) were all done on the website. this site replaced most community boards and chat rooms. though it went to the extreme at times (with public heated arguements being broadcasted on craigslist to my own personal horror story) i was often grateful that the site existed. it was a common meeting place that rarely was discriminatory.
If Craigslist added all the fancy web 2.0 features, they would destroy their cost structure. At that point, they’d probably be forced to charge listing fees, etc.Consumers have clearly spoken – free classified ads with minimal functionality is preferable to ones requiring even small payment but better functionality
This system is in place for buyers and sellers to follow their best interests, and more times than not, it works out and the market wins. There are other, less efficient markets for buyers and sellers to use if they want to pay for protection. I think CL is fantastic.
Keeping this anonymous as NDAs require me to do so, I know of at least 2 other companies that have this type of revenue per employee number.One is an online division of a major retailer in the US. Employees number less than 100 total including a slight amount of “overhead” for management. The other is a wholesale, members only auction site where the number of employees is more like 50.The retailer’s estimates for their online division are close to a billion annually. The auction site gets close to that number as well.The interesting thing with both of my examples (again, sorry to be vague) is that they are both smaller parts of a larger corporate parents that have been around for a long time.Maybe this is more about Seth Godin’s view of companies building a tribe or following and then charging people for the privilege to interact with them.
Appreciating this discussion and learning from it. A few things I found worth thinking about as I read the comments here: if it is not pressure from investors that leads to counter-productive revenue maximizing strategies as businesses grow, what is it? Fear of competition? Status-seeking? Curiosity about trying new things? Some combination? I think it is inarguable that a resistance to running banner ads or doing other user-unfriendly things to maximize revenue (upselling, “premium” or “sponsored” listings, etc.) has contributed to craigslist success. What are the specific pressures they are resisting, and is this a general lesson that has meaning for others?I appreciate the case “for” and “against” craigslist, but I hope that the irony some noticed in the case against is understood as intentional. It is impossible to study the company for any length of time and not ask the obvious question: if it is so bad why is it so good? Or as the story says: “Every day the choristers of the social web chirp their advice about openness and trust; craigslist follows none of it, and every day it grows.”
Gary, I thought your article was brilliant. Craigslist is both frustrating and awe inspiring. You capture this confusing phenomenon perfectly. If and when they are beaten, it won’t be because of feature creep or overreach.
thanks!!—Gary WolfContributing editor, WiredMagazine home: http://www.wired.comCurrent project: http://www.quantifiedself.org
It is rare to get a comment from the author of a piece referenced on this blog. So I am thrilled and flattered by your comment garyYour last point about being closed does bother me. Our portfolio company Indeed has the second most popular jobs site on the web, second only to craigslist. Their model is different. They don’t take listings, the aggregate and provide job discovery for their users (a google model)The only jobs site on the web that they are not allowed to crawl is craigslist
This is perhaps the most interesting thing… the relationship between a sincere devotion to users and total deprecation of the the “openness” value – or, as they might think of it, despite Craig Newmark’s application of the idea in his NGO/government discussions – openness dogma. One hint there for all the rest of us might have to do with the “tax” aggregation creates on users’ attention through replication, failure to cancel old postings, general “noise.” Noise is generally considered just another opportunity for good filtering. At CL they cut out the intermediaries. An interesting old/new model!
Gary, I love the disintermediation model. Long live disintermediation!!
It grows because their aim isn’t to make money, it’s to serve people without getting in their way, and that goal is good in of itself.
Can’t agree more. CL is a model that most of us, if not all, crave for. While it can tough to re-create a CL like company, I think its important to make sure that a startup is built, from the ground up, to try to be like CL. Money has a lot to do with it as well. When you raise a few million dollars, it tough to keep the discipline to remain small. Having said that, I agree that most USV companies are diff than many other VC funded startups in terms of being small and trying to create a CL like model.I also think that every startup can be CL like and it doesn’t have to be a $100m business. If you make $2m and your costs are <$200K, I think that is good as well as every segment may not be as big as CL’s.
“The obvious is that which is never seen until someone expresses it simply.” (quote by Kahlil Gibran)Going from zero to $100 million in 14 years is not a trajectory most VC’s would have tolerated, especially passing through the Internet hey days where Craigslist looked perhaps more like a lifestyle business rather than a fundable business. It’s ironic they are being envied now.So, the lesson for VC’s if they want to emulate this- is to change their paradigm as well. Perhaps there are cases where business viability and repeatable dividends might trump a quick desire for X10-20 over 3-5 years.btw- the biggest Marketing department they have is Craig himself. He has become quite visible & quotable everywhere, so his exposure promotes Craigslist very efficiently.
Long term business strategies is something our economy needs a lot more of.
I would love to see Craigslist create an API that would allow others to experiment and create with the incredible following they’ve amassed. As a platform versus a website, developers would be free to deliver on the functionality they feel is not being delivered, and Craigslist could remain true to it’s personal vision of offering only what the community demands or requires. It’s the best of both worlds…but as it stands now, they offer only RSS feeds and block many of the attempts to use these to create an additional layer of functionality on top of their data. I also find it interesting that they’re not often criticized for this (at least not that I’ve seen) compared to other high-profile sites.
Checkmate. That is the only thing that pisses me off about them. I believe it will ultimately do them in
In today’s web, great design and iterative, agile development are touted as the best ways to build a great product. With everyone focused on rounding their corners to separate themselves from the pack, it is annoying to them that Craigslist separates itself by way of their product to such an extent that they can openly, proudly ignore everything else.I think this highlights a couple points.1. Great visual design has become a proxy for differentiation and a potential “competitive advantage”. When your design is a differentiator it means your functionality is a commodity. That should be a concern.2. Nothing so protects a company like network effects. In the CL case, it is the people, not the code that make the company unstoppable. They can screw up everything else, but because they have the people there is no stopping them.
Your second point is something we say to ourselves every day. It is the first law of web economics
That’s interesting. I agree that it is the people that make CL work. But they don’t know each other. And I know many believe that a social network wrapped around Craigslist-like listings would be a great idea (including myself… some time ago), yet anonmyity seems to be more important.The audience has no other connection to CL besides their own instinct to go there. Love that. No login. No pressure. However… Tough sell for a competitor startup looking for money.
edit: boring
A relevant piece to this discussion from today’s WSJ: How Long Does It Take To Build A Technology Empire?: Source: WSJ.com: Venture Capital Dispatch http://bit.ly/CSNAt (via @techvcnews) “Oracle Corp., one of the world’s largest software makers, reached $50 million in revenue in its 10th year. It took software king Microsoft Corp. eight years to hit that milestone.”And a thought on the the “Mess” thing: that mess is tolerable when compared to the value provided- so users will tolerate not-so-perfect solutions, as long there is something of great value in the middle of that mess. It’s almost like an “invisible mess” to most users.
As people in the tech industry, we tend to focus on the craft of what we do rather than what value we’re providing to our users. It’s like the Hollywood films that focus on special effects, movie stars, and gigantic marketing budgets. At the end of the day, people just want to be entertained by good story, which is why films like Blair Witch Project, My Big Fat Greek Wedding and Napoleon Dynamite explode onto the scene every few years, and every few years Hollywood folks are absolutely shocked, yet again. Sometimes it does take massive budgets, larger than life movie stars and out of this world CGI to make a blockbuster, but the mistake is to think that those things alone make it so. Just as it’s harder to write a short letter than a long one, it’s much harder to create a simple and extremely useful service than a really good looking and complicated/technologically advanced one. You don’t get high fives from peers for creating simple services, you get “oh, I can’t believe THAT has a million users, I could write that in a day”. But when we’re surrounded by industry folks, it’s easy to forget the high fives that really matter are the ones from users.
Great comment and so true. Whenever I think of simple and ugly I think of joshua schachter and delicious. That service was butt ugly but it was incredibly useful. Still is
i find it interesting that so few attribute much CLs success to when the business originated and being the the first to offer such a service. i believe the same of eBay, yet i believe CL will continue to thrive for the sake of it’s simplicity. it was Buckmaster “When he realized that every post had to be reviewed and published by hand, he created the automated process that allowed craigslist to grow.” – which instills my belief that the leader of a web service company should have at least some technical background. keep it simple, automate everywhere possible, innovate by making peoples lives easier, ill run with that and take it to the bank later.
I agree jason. When the marketing types take over its an issue
I love Craigslist’s approach and I’m glad you are sticking up for them.Especially it’s lo-tech look-and-feel. As an architect type, I really appreciate how they must be squeezing every drop of capabilities from their server ops. No need to bog down the pipes with fancy graphic designs – raw HTML serves the application well – although I can understand that such an approach may not fly if it had started in the past few years.Also, it’s still the purest example out there of Kopelman’s “shrinking-a-market”. But, I did comment on Brad’s earlier post about how shrinking markets and unlocking efficiencies through disintermediation may be at fundamental odds with a broader economic initiatives for “full employment”.
The portrayal (in the article) of CL’s leadership as Internet Neanderthals is well done but seems to undersell the genius behind the craigslist “anti-brand.” The site was already comparatively underdesigned when it was launched in its modern incarnation in 1998. The urge to redesign misses the point. http://bit.ly/DFWhD
1. the game is niche. when i go to craigslist, i go to miami.craigslist.org, or newyork.craigslist.org. failure to appreciate niche is i think one of the biggest plagues in the internet business community, everybody is too money hungry and focusing on the big thing. i’m reminded of seth godin’s money paradox: “the sooner you ask for money, the less you get.” CL is immensely disruptive, disruptions almost always take an unreasonable amount of time.2. one concept john hagel has talked about repeatedly is that of unbundling and understanding what type of company you are. most intenret companies are still raising a lot of money and doing too much shopping, building their own CMS, hiring their own designers, keeping their own server farm….IMHO the first question of every internet entrepreneur should be “how can i get someone else to do this for me?” the more you answer that question, the fewer employees you’ll need.3. CL is leaving money on the table, but i do not think this is virtuous. in fact i think this is terrible and we should all be dissing CL because of it. “CL 2.0” is not going to leave money on the table, is going to make transactions safer (i’ve gotten cheated on CL twice already)4. i’m overly biased given the nature of my business, but i think blogging is where it’s at. it can produce the passionate community needed to generate outrageously high profit margins. it can be built on top of commoditized technology (blog CMSes). and perhaps most importantly it can get the audience without spending the big marketing bucks. it’s tough, and only blog stars like fred will be able to pull it off, but IMHO that’s where the oppty is.
Hey Fred, “mess” brought back the days of CB Radios. It occurs to me that the CB Radio market can teach us what to anticipate for Web3.0.http://bit.ly/18PteyKatherine Warman Kern@comradity
It most certainly is not the most “efficient and cost effective way to make a market”. There “community moderation” has resulted in a dictatorship of children. Here in San Diego, there is a section called “gigs” where people are supposed to be able to post business and job opportunities. It used to be somewhat useful, but now self-appointed censors will “flag” anything that isn’t a traditional salaried job, making it worthless as a business or community site. Craigslist was a nice concept when it started, but now it is a dysfunctional monopoly that has outlived its usefulness.
Fred, I admire Craigslist, and it’s definitely worth studying. They have created a great system with minimal resources (28 employees). Their strategy of not developing features works for them now, but in the long term it does seem to prevent innovation. Craigslist has acquired critical mass high enough that others can’t compete just because of their huge user base. Thus, Craigslist is blocking startups who are willing to experiment. This can’t be good for the users, and this is why I too think Craigslist is a mess.PS btw, I love this quote: “Only programmers, customer service reps, and accounting staff work at craigslist. There is no business development, no human resources, no sales. As a result, there are no meetings.”–mihai
Craigslist is a success because Craig Newmark kept the purpose of his company from the beginning. That’s what a lot of Internet entrepreneurs forget when they start seeing jumps in their traffic. Every site you go on to these days, you see it crammed up with Google ads. They all seem to forget why they started the site in the first place. By having a simple design and by having a steady stream of revenue from job advertisements, he is doing good for society and he knows it. If he was out just for the money, he would have advertisements all over the place.Consistency is important when designing a website, and I make sure I follow that every time I pursue one of my ideas.
Drudgereport is a mess too! Man those 1 million visitors PER HOUR are pathetic.Good post Fred. Design that does not ADD to usability/utility is at best a waste.
I read the Wired article earlier this week, and the thing that struck me was the issue of antitrust. I see a lot of parallels between Craigslist and Microsoft’s IE antitrust battle. In both cases you have a dominant, free product, it generally serves people well but is criticized for stifling innovation, and it steamrolls competition. The fundamental difference is that IE is (was) controlled by Bill Gates, the richest man in the world. Craig is an odd little guy who isn’t interested in money.Would Craigslist get the pass it does if Craig essentially ran the business the same way, but was openly seeking fortune and market share? This guy is really disrupting a lot of businesses and industries with what could and has been called predatory pricing.I’ve always been against antitrust prosecutions — they almost always are brought on by competitors, not customers. There’s also a strong populist streak that runs through antitrust campaigns. People love(d) hating Microsoft and Bill. Has Craig has sidestepped the blowback with his lumpen approach? If so, what does that say about the motives of antitrust campaigns?
People’s aversion to CL’s design at this point are totally missing the mark. Sparse design is their branding, and it is a branding decision that only increases in value as other sites adopt more widgets, AJAX, layers, panels, and where audiences are sold “function” but only receive “form”. In a sea of people swimming toward the shiniest lure, Craig landed value in the opposite direction. It is a differentiating maneuver with teeth. After a century of classified ads in newspapers, the metaphor is pretty established people. Everyone who has tried to make a classified ad into something that it is not has failed miserably.
Here is a great chance to drive a large number of targeted visitors to your blogs and websites for free.Submit your websites, blogs, videos to http://www.zillionsb.com and get 1000s of visitors everyday for free. It also helps your websites/blogs gain valuable backlinks.Let the other bloggers cast their votes to push your posts up for a greater visibility. Enjoy free huge traffic to your sites.ThanksSarahttp://www.zillionsb.com
I think Craigslist is an absolutely fascinating business. As someone who has studied the classified market quite closely (as a fmr VP Biz Dev for Monster), I think that Craigslist is totally unique in the classified or ecommerce markets. I won’t go into all of my opinions on the CL functionality, many of which are in these comments. Suffice it to say it’s good enough for many things, including entry level jobs or infrequently posted jobs. I may blog about it soon.But regarding the CL business model, I think it is only fair to call it “bait and switch.” CL attracted all of its 10s of millions of users with a very simple interface and a simple value proposition- free to post. AFTER it had all of those users, including in the jobs vertical, they launched the only real money making aspect of the site, pay to post jobs in major markets. Yes, the price is WAY lower than a newspaper ad or Monster posting, but it’s still pay to post. I highly doubt if Craig had launched the business as ‘almost free’ he would have achieved the same traffic levels.
Craigslist appeals to that natural instinct to find the classified ads. More accurately, to find the main marketplace where you can post a listing or view other listings (for goods and services), with the confidence that you are going to the right place.If you focus only on the market sectors where Craigslist and its “competitors” make money, then you are missing the point. You have to think about the other categories that you and others visit much less frequently. On an individual level, it might seem inconsequential — like that very occasional need to browse/search the listings for kids bicycles, either for buying or selling. (obviously not a new concept: aka, long tail)And Craigslist treats these categories as first-class citizens. When you are the beneficiary of a Craigslist transaction, that reinforces your desire to come back to Craigslist when you need to buy or sell something.Are there are other companies/startups that recognize this and want to serve these market sectors without any profit motive? Indefinitely? It’s a tough arena for any company to tackle, especially if zero dollars and critical mass are the numbers you are trying to beat.So. Personal example. I tried to arrange a vacation apartment rental in Denmark through Airbnb.com and my transaction was cancelled by airbnb to protect me from a suspected fraudulent listing. I didn’t have time to find another rental and so I turned to Craigslist. In Copenhagen. The safety and features of Airbnb were comforting, yet the deal did not close.Through some vague feeling of confidence and from my conversation with the person renting the apartment, it worked out just fine. Just showed up with some cash and received a key. Both parties had to put some trust into the deal. That might sound dumb and risky, but that’s what classified listings are like sometimes. And when it works out once, it makes you want to do it again. (Apologies to those who have had the opposite experience. Follow the warnings. )peace(sorry for the repost. this page seems to have two comment entry forms. one for disqus and the other… for the clueless)
They have passionate users. Have reinvented the classifieds market place. Are kicking their competitions asses. Have little overhead. AND apparently love going to their jobs every day.Um yeah. If that’s a mess? Bring it on!
Craigslist succeeded because of luck and timing, not because of a genius model, a masterful business plan, incredible insight, or brilliant management. This post should really be titled “What We Can Learn From Winning the Lottery” and since the lotto is currently @ $278 M, I would suggest that you invest $4 M in lottery tickets, which will give you better odds than investing in a company that presents you with no business model, no tech, no management, etc.Something most investors don’t understand is what passion, art, and expression mean to a product. Craigslist was a labor of love, not a creation for a spectacular exit and Craig’s altruistic intentions shine bright in the end user experience. Again, timing… While other companies were scheming to get rich, Craig whipped up a site that was uncomplicated and provided free access to the same information others were trying to broker. New visitors were spared the blatantly capitalistic overtones that were so common with sites at the time. It’s pretty hard to compete with free. The .org extension was also very disarming for new visitors. At a time when get-rich-quick fever was rampant and the .com gold rush was in full swing, it was nice to visit a friendly .org that wasn’t caught up in all the madness. Modern sociology is another topic but having user interest driven by user entries, once the user base reached its critical mass, the site took on its own life (what we now call “viral”).Companies with the burden of investor expectations, exit strategies, development costs, staff salaries, etc. are simply not allowed to build something like Craigslist.org. If I were to present you with a business that had no user acquisition strategy or business model, I think it’s safe to say your face would produce a pretty confused look. Obviously, after we have full hindsight with Craigslist, anyone with a pulse would like to “invest in a company like this” but winning the lottery isn’t something you get to control.
Its not all luck
Thanks for this post, Fred. People underestimate the need to learn from the success of companies like craigslist, who in many ways fly against conventional wisdom; precisely because by definition ‘conventional wisdom’ alone does not create the top 0.1% (or even 1%) of companies.Perhaps one day you should write a post comparing and contrasting the strategies of Craigslist and Twitter, who likewise are one of the very few companies ever who have created $1B of value with less than 50 employees.
The first time I used Craigslist will also remain the last time.
The first time I used Craigslist will also remain the last time. Craigslist has done that by doing very little indeed.
Sorry for the double comment.
Huh? Craiglist isn’t a mess, it’s dirt simple. The interface looks like a 90s interface so that means it’s user-friend and not geekified with slow-loading junk and loads of white space and tiny olive-green ink like a contest for who can design the best Islamic country’s flag — which is what every other fashionable website looks like these days I don’t care for Craigslist lately because @craignewmark is so politicized, adopting all the hard left geeky freaky causes and applying the escort ad dough to those causes, which is something I don’t wish to assist him in doing, coming or going. Still, I recognize that for ride-sharers and job-seekers, it’s great.
“Craigslist gets more traffic than either eBay or Amazon.com. eBay has more than 16,000 employees. Amazon has more than 20,000. Craigslist has 30.”Mind boggling. And Craig manages to reply to all tweets I send his way.”Only programmers, customer service reps, and accounting staff work at craigslist. There is no business development, no human resources, no sales. As a result, there are no meetings.”Wow.
Fred, the Craiglist model requires something that’s very hard for many businesses, especially businesses backed by investors such as VCs, to emulate, and it’s not financial or a business model. It’s a mindset in which the founder and chief executive are willing to forgo financial gain and maximizing shareholder (or stakeholder) value in pursuit of “social capital.” I’ll be talking to Craig and other panelists at the Economist media convergence conf about this mindset issue, but I’d be curious your thoughts if any.
You have to be patient. Google, Facebook, Twitter, Tumblr, Disqus, Etsy, etc, etcVenture capital is supposed to be patient capital. When it is patient, great things can happen
I hear you. But doesn’t it seem Craig and Jim are not only being patient, but also being willing to forego some gains, long- or short-term? And isn’t that anathema to financial investing?
Nope. I’ve blogged about this before. I think one way to create sustainability over the long run is to avoid over optimizing monetization in the short run and the earnings multiples go up if investors think you’ve got long term sustainability
Thanks, much. I have ready you write this, and appreciate you pointing outhow it applies here.Also, fyi, loving Disqus. How I can log onto your blog, or keep theconversation going here through email — and also aggregate my commentselsewhere.db