Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Bootstrapping a startup with Craigslist arbitrage (cam.ly)
31 points by danecjensen on Jan 6, 2011 | hide | past | favorite | 26 comments


I appreciate the post, but in the end it just isn't that useful. It's like "I did all this stuff to sell MacBooks and didn't really make any money, so I started this startup."

Obviously you can't do it for long if you're not making coin, but either a more in-depth analysis of how many you sold for what prices, or any success / fail stories, would make this a more useful post for readers. Instead it's all the build-up of an objective study with no real payoff.

Hopefully this comes across as a useful critique and not something that's personally attacking. I like the idea, I just want something more out of a post like this.


Find a deal with the data and intuition on Craigslist. Then, meet in person and barter at least $50 off the asking price.

You mean, agree to $X over email and then demand it for $X-50 when you meet in person? That's really, really tacky.


The irony is that he says that immediately after "Selling on Craigslist sucks, it’s time consuming and you have to deal with flaky people that just want a deal." He's recommending exactly the behavior that makes Craigslist suck.


I came here just to post that exact thought!


craigslist isn't a fixed-price economy; approaching it as such is silly. Most informal market places in the world are not fixed price economies.


He said 'asking' price.

What he is saying is that the person is offering $X and when he arrives in person he counters with $X-50. (Probably more like offering $X-100 so you get a counter at $X-50).


Right... but if you agree to the price over the phone, don't negotiate in person. It's a dick move. If I'm selling my TV for $250, you agree to that, I drive to meet you and you low-ball me for $200, I hate you. It's that simple.


Rule #1: Never drive to meet someone if you're the seller. Let them come to you. If they can't, someone else will.

Half of people on CL are flakes or cheapskates, you have to assume they will either never show up or lowball you


Hah! Like I'm going to give my address out to people.

I meet people in a public place: Starbucks, a grocery store, McDonalds, etc.


No you say that you would like to look at the piece before you can decide on a price. And then you meet and then you price it.

If you take a look at any rural population, you'll see that every price is negotiable.


That's fine, because you've made your intentions clear. In that case, if I'm not willing to budge, I'll let you know that, and we can avoid a meeting. I've had people do it exactly as I described, and it's an infuriating waste of time for me.


Nobody here proposed using that as a tactic.


Perhaps not, but it's not unheard of. "Smart" bargainers sometimes resort to tactics like that, because they figure since you've already come this far to meet them, you'll lower your price just to get the deal over with. Which I generally won't do on principle, because that pisses me off.


If I'm meeting with you to sell you something, it's because we agreed on a price already. If you try to bargain with me after the fact, I'll just say "No," and leave. You've violated my trust and broken the deal.


Properly speaking, this is not arbitrage. There are two criteria for arbitrage, it has to be riskless and self-financing. Looking at an item and saying, I should be able to sell that for more is not an arbitrage. If you were able to locate a buyer willing to pay a certain price prior to purchasing the item, and then buy an under priced item and sell it to your buyer for more money, that would be arbitrage. Arbitrage is a sexy three dollar word that people like to throw around, but if you are risking loss in order to hopefully make a profit, then plain and simple it is not arbitrage.


He's doing statistical arbitrage.


What basis are you using for saying this is StatArb? Typically that involve hundreds or thousands of equities where you can use the laws of large numbers to your advantage. Buying one item that is below the average price of something and "hoping" you can sell it for more just doesn't strike me as any type of real arbitrage.


StatArb has no dependency on equities. This is arbitration, Arbitration between two markets, his local craigslist market and the rest of the online market. There is no hoping, hes priced the item in one market and it is above the price of another, that is pure arbitrage.


Arbitration does not mean what you think it does.


s/arbitration/arbitrage/g my bad.


> Here’s an interesting graph from the data that proves the price difference increases with price: [followed by a graph of nearly random data]

Drawing a straight line through loosely correlated data doesn't "prove" anything.

> I didn’t do Craigslist arbitrage for long because it was not worth my time and effort.

So a more appropriate title for this post would have been "Failing to bootstrap a startup with Craigslist arbitrage"?


> For example it quickly became apparent that a unibody Macbook was worth more than one that wasn’t a unibody.

Then that's just another variable. Not intuition like the author claims. Intuition would be in recognizing the important variables.


He posts a math formula, a fancy graph, and even the Python script but barely mentions the end result.

I would have loved to know how many he actually bought, sold, and for what profit.


Slightly unrelated to the blogpost, but how do your cameras work in low light settings? I'm looking into a similar system for a business my wife and I run.


I wrote a (fairly) successful e-book on this topic, freely available over at my site.


the word is "bargain" not "barter".




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

Search: